2009 IEEE International Conference on Automation and Logistics

August 5-7, 2009 Shenyang, China

Cosponsored by Institute of Electrical and Electronics Engineers, Inc. Shandong University, China IEEE Robotics and Automation Society

Technically cosponsored by: Northeastern University The ASME Dynamics Systems & Control division Chinese Mechanical Engineering Society Federation of Logistics and Purchasing of China Chinese University of Hong Kong Logistics Engineering Institute of China

2009 IEEE International Conference on Automation and Logistics

August 5 – 7, 2009 Northeastern University, Shenyang, China

Conference Program Digest

Sponsored by

Institute of Electrical and Electronics Engineers, Inc. IEEE Robotics and Automation Society Shandong University

Technically Sponsored by

Northeastern University The ASME Dynamics Systems & Control division Chinese Mechanical Engineering Society Federation of Logistics and Purchasing of China Chinese University of Hong Kong

IEEE ICAL 2009 CD-ROM Proceedings

Additional Copies may be ordered from:

IEEE Service Center 445 Hoes Lane P.O. Box 1331 Piscataway, NJ 08855-1331 U.S.A.

IEEE Catalog Number: CFP09CAL ISBN: 978-1-4244-4795-4 Library of Congress: 2009905379

Copyright and Reprint Permission:

Copyright and Reprint Permission: Abstracting is permitted with credit to the source. Libraries are permitted to photocopy beyond the limit of U.S. copyright law for private use of patrons those articles in this volume tha carry a code at the bottom of the first page, provided the per-copy fee indicated in the code is paid through Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923. For other copying, reprint or republication permission, write to IEEE Copyrights Manager, IEEE Operations Center, 445 Hoes Lane, P.O.Box 1331, Piscataway, NJ 08855-1331. All rights reserved. Copyright © 2009 by the Institute of Electrical and Electronics Engineers, Inc.

2 Welcome

It is our great pleasure to welcome you to the 2009 IEEE International Conference on Automation and Logistics (IEEE ICAL 2009), which takes place from August 5 to 7, 2009 at the Hanqing Conference Hall of Northeastern University in Shenyang, China.

IEEE ICAL 2009 marks the third event of the IEEE ICAL annual conference series. We have received 903 paper submissions from 20 countries and regions. After a rigorous full-paper peer-review process, 409 papers were accepted for oral or poster presentation at the conference in 28 technical sessions, resulting in an acceptance rate of 45%. These papers reflect the dynamism of the research and development activities in automation and logistics, as well as the emergence of new research and development topics in addressing the ever increasing challenges from the related industrial and societal needs.

The conference program is highlighted by six invited plenary speeches to be delivered by Prof. David D. Yao of Columbia University, USA, Prof. Kazuhiro Kosuge of Tohoku University, Japan, Prof. Zhongtuo Wang of Dalian University of Technology, China, Prof. Warren B. Powell of Princeton University, USA, Prof. Yoshimi Takeuchi of Osaka University, Japan, and Prof. Mengchu Zhou of New Jersey Institute of Technology, USA. The main objective of the IEEE ICAL conference series is to provide a forum for researchers, educators, engineers, and government officials involved in Automation and Logistics to disseminate their latest research results and exchange views on the future research directions of the related fields. IEEE ICAL 2009, hosted in the famous historical city of Shenyang, China, promises to be a great event for participants from all over the world, with an excellent technical program as well as social activities.

We wish to express our gratitude to all of our sponsoring societies and organizations and to all the individuals who have contributed to the organization of this conference. Special thanks are extended to our colleagues in the Program Committee for their thorough review of all the submitted papers, which is vital to the success of this conference. We must also extend our thanks to our Organizing Committee and our volunteers who have dedicated their time toward ensuring the success of this conference. Last but not least, we thank all the contributors from 20 countries and regions for their support and participation in making this conference a great success. Finally, if your travel plans permit, we encourage you stay beyond your meeting to enjoy visiting Shenyang and China. We wish you a great conference experience and enjoyable stay in Shenyang, China.

Tianyou Chai Tzyh-Jong Tarn Toshio Fukuda Huaguang Zhang General Co-Chair General Co-Chair Program Co-Chair Program Co-Chair

3 Table of Contents

Welcome ······························································································· 3 Organization Committees ············································································ 5 Program Committee ·················································································· 8 General Information ················································································ 10 Transportation Information ········································································ 12 Program Schedule ··················································································· 15 Floor Plan of Conference Venue ·································································· 16 Plenary Talks ························································································· 18 NEU 111 Project Panel Automation and Logistics Forum ····································· 30 Symbols for Sessions and List for Session Titles and Session Chairs ······················· 31 Session IDs, Titles & Chairs ······································································· 32 Oral Sessions Wednesday Sessions ······································································ O-1 Thursday Sessions ······································································· O-11 Friday Sessions ·········································································· O-16 Poster Sessions Wednesday Sessions ······································································ P-1 Thursday Sessions ······································································· P-13 Friday Sessions ·········································································· P-25

Session Chair Index ················································································ S-1 Author Index ······················································································· A-1

4 Organization Committees

Advisory Committee CHAI Tianyou, Northeastern University, China CHEN Hanfu, Chinese Academy of Sciences, China DAI Ruwei, Chinese Academy of Sciences, China DE SILVA Clarence, U. of British Columbia, Canada FUKUDA Toshio, Nagoya University, Japan GUO Lei, Chinese Academy of Sciences, China LUH Peter B., University of Connecticut, USA SICILIANO Bruno, University of Naples, Italy TARN Tzyh Jong, Washington University, USA XU Yangsheng, Chinese University of HK, China

General Chairs CHAI Tianyou, Northeastern University, China TARN Tzyh Jong, Washington University, USA

General Co-Chairs JIA Lei, Shandong University, China KOSUGE Kazuhiro, Tohoku University, Japan LIU Jianchang, Northeastern University, China VOSS Stefan, University of Hamburg, Germany WANG Yuechao, SIA, CAS, China ZHOU Xunyu, University of Oxford, UK

Program Chairs FUKUDA Toshio, Nagoya University, Japan ZHANG Huaguang, Northeastern University, China

Regional Program Co-Chairs CHEN Ben M., National U. of Singapore, Singapore KERSTEN W., Technical U. of Hamburg, Germany LIU Derong, University of Illinois at Chicago, USA TANG Lixin, Northeastern University, China WANG Zhidong, Chiba Institute of Tech., Japan YAN Houmin, Chinese University of HK, China

5 Organizing Chair MENG Q.-H. Max, The Chinese University of HK, China

Organizing Co-Chairs CHEN Tongwen, University of Alberta, Canada DUAN Guangren, Harbin Institute of Tech., China FU Minyue, University of Newcastle, Australia MA Shugen, Ritsumeikan University, Japan TANG Jiafu, Northeastern University, China WANG Hong, University of Manchester, UK WU Yaohua, Shandong University, China XIE Lihua, Nanyang Tech. University, Singapore

Organized & Focused Session Chairs ASAMA Hajime, University of Tokyo, Japan CHIAVERINI Stefano, University di Cassino, Italy FENG Gang Gary, City Univ. of Hong Kong, China GE Shuzhi Sam, Nat’l U. of Singapore, Singapore HU Huosheng, University of Essex, UK KWON Dong Soo, KAIST, Korea LIN Wei, Case Western Reserve University, USA SU Chunyi, Concordia University, Canada ZHANG Jianwei, University of Hamburg, Germany ZHOU Mengchu, New Jersey Inst. of Tech., USA

Tutorial & Workshop Chairs HUANG Jie, Chinese University of HK, China QIN Sizhao Joe, U. of Southern California, USA YANG Guanghong, Northeastern Univ., China

Panel Organizing Chairs CHEN Guanrong Ron, City University of HK, China GU Jianjun Jason, Dalhousie University, Canada JIANG Zhongping, Polytechnic University, USA

Poster Session Chairs TIAN Guohui, Shandong University, China WANG Dingwei, Northeastern University, China YANG Xianyi Simon, University of Guelph, Canada

6 Publicity Chairs CHEN Jie, University of California, Riverside, USA GUO Shuxiang, Kagawa University, Japan GUAN Xinping, Yanshan University & SJTU, China LIN Zongli, University of Virginia, USA LIU Xiaoping Peter, Carleton University, Canada PARK Jong Hyeon, Hanyang University, Korea SUN Dong, City University of Hong Kong, China WEN Changyun, Nanyang Tech. Univ., Singapore XI Ning, Michigan State University, USA ZHENG Weixing, U. of Western Sydney, Australia

Awards Committee Chairs LI Yibin, Shandong University, China WANG Chengen, Northeastern University, China ZHANG Hong, University of Alberta, Canada

Publications Chairs HU Chao, CAS/CUHK SIAIT, China MING Aiguo, Electro-Communications U., Japan YUE Heng, Northeastern University, China

Finance Chair

ZHANG Chengjin, Shandong University, China

Local Arrangement Chairs FAN Songtao, Northeastern University, China LU Shaowen, Northeastern University, China ZHENG Xiuping, Northeastern University, China

Secretariat LIU Zhuo, Northeastern University, China

7 Program Committee

Program Chairs FUKUDA Toshio, Nagoya University, Japan ZHANG Huaguang, Northeastern University, China

Regional Program Co-Chairs CHEN Ben M., National U. of Singapore, Singapore KERSTEN W., Technical U. of Hamburg, Germany LIU Derong, University of Illinois at Chicago, USA TANG Lixin, Northeastern University, China WANG Zhidong, Chiba Institute of Tech., Japan YAN Houmin, Chinese University of HK, China

Committee Members

Gholam Reza Amin William Gruver Yibin Li Hisayuki Aoyama Jason Gu Jingsheng Liao Jacky Baltes Dongbing Gu Li Liao Evgueni Bordatchev Qi Hao Xiaoping Liu Robert Brennan Minai He Ming Liu Zhili Cai Yasuhisa Hirata Kefu Liu Chunqi Chang Xuezhang Hou Chao Liu Chien-Chern Cheah Zeng-Guang Hou Peter Liu Zhenxue Chen Huosheng Hu Rong Liu YangQuan Chen Chao Hu Shanzuo Lou Chin-Yin Chen Behrad Khamesee Yide Ma Hao Chen Shunichi Kobayashi Xiangru Meng Xinkai Chen Naoyuki Kubota Hao Meng Bing Chen Ti-Chung Lee Max Q.-H. Meng Mantis Cheng Zhenhua Li Mehrdad Moallem Ozren Despic Youfu Li Petr Musilek Zuren Feng Zhijun Li Shigeaki Ogese Mark Fiala Baopu Li Gyei Kark Park Reza Fotouhi Yangmin Li Gary Parker Yunjian Ge Xiaoli li Geof Pond Sam Ge Howard Li Erwin Prassler

8 Hairong Qi Zhizhong Tarn Guilin Yang Ronghu Qi Abdelhamid Tayebi Lucy Yang Zheng Qin Guohui Tian Simon Yang Hongliang Ren Gancho Vachkov Xiufen Ye Hideyuki Sawada Changjun Wang Jingang Yi Kiam Tian Seow Bing Wang Hyun Joong Yoon Nariman Sepehri Ying Wang Yong Yu Weiming Shen Xiaona Wang Chengxin Yu Yanjun Shen Shuyun Wang Cao Yun Jie Sheng Xin Wang Lilin Zang Weihua Sheng Xingmei Wang Chengjin Zhang Xianliang Shi Kerwin Wang Yunfeng Zhang James Smith Zhidong Wang Jianwei Zhang Rui Song Changhua Wu QiJun Zhang Kai-Tai Song Jonathan Wu Yuan Zhang Quanjun Song Mianhong Wu Yunong Zhang Koichi Sugimoto Jiwei Xiao Xinzhi Zheng Guohua Sun Lisheng Xu Huiyu Zhou Fengchi Sun Jinxue Xu Fangwen Zhu Dong Sun Na Xu Qun Zong Jin Tang

9 General Information

Conference Venue

Han-Qing Conference Hall Northeastern University Shenyang China

Language

The conference and all its activities will be conducted in English.

Conference Secretariat

IEEE ICAL 2009, c/o Ms. Zhuo Liu Northeastern University, China E-mail: [email protected]; telephone: +86 (24) 2390-9157

Dietary Needs

Conference delegate or partner with special dietary needs etc. is invited to indicate his/her special dietary needs to the organizing committee at the Registration Desk.

Welcome Reception

This welcome dinner on Wednesday night features famous Liao Cusine of Shenyang region at the favoured New Hong’s Restaurant for all registered ICAL 2009 participants. Don’t miss this opportunity to enjoy local cusine and to meet friends and colleagues.

Date: August 5th, 2009 Time: 18:00 - 20:30 Venue: New Hong’s (新洪记) Restaurant

Conference Registration

Tuesday, August 4th, 2009: 14:00 - 20:00 Venue: Registration desk on the second floor of Sheraton Hotel Shenyang

Wednesday August 5th, 2009 08:00-17:00 Thursday August 6th, 2009 08:00-17:00 Friday Augustr 7th, 2009 08:00-12:00 Venue: Han-Qing Conference Hall Lobby, Level 1, Northeastern University

10 Conference Banquet

Date: Thursday, August 6th, 2009 Time: 18:00 - 21:00 (Please gather at the lobby of Han-Qing Hall to take bus at 18:00) Venue: Banquet Hall of Liaoning Tower

Additional Equipment

Please note that each session room will be equipped with a LCD projector, screen, laser pointer, and microphone. Laptop and/or desktop computers will be provided. General software presentation package are preinstalled, such as Microsoft PowerPoint, Adobe Acrobat reader. If you plan to use specific software presentation packages other than stated above, you are required to bring your own computer and install them before your presentation. Please include your paper number in all correspondence. Each presentation is allocated 20 minutes inclusing question and answer period.

Conference Proceedings

Conference registration fee includes a copy of the Conference Proceedings on USB Key. Additional copies are available at a cost of US$50.00 (RMB350) each. Please enquire at the Registration Desk.

Internet Access

Free wireless access will be provided for all delegates at the conference venue: Han-Qing Conference Hall and in all presentation rooms on Wednesday, Thursday and Friday.

11 Transportation Information

The main airport in Shenyang is the Shenyang Airport (SHE) is the largest airport of northeast China offering direct flights to cities around China including countries such as

Japan, Russia, and Korea Here are some of airlines out of Shenyang:

ANA: +1 800 235 9262; http://www.fly-ana.com

Asiana Airlines: +1 888 437 7718; http://www.flyasiana.com

China Northern Airlines: +86 23 197 188; http://www.cna.com.cn/Eng/index-en.html

China Southern Airlines: +86 20 950 333; http://www.cs-air.com/en/index.asp

By plane The city is served by the Shenyang Taoxian International Airport (沈阳桃仙国际机场, airport code SHE), as well as by several smaller, regional airports. Direct flights from Shenyang go to Beijing, Changsha, Chaoyang, Chengdu, Chongqing, Dalian, Fuzhou, Guangzhou, Haikou, Hangzhou, Harbin, Hefei, Hong Kong, Jinan, Kunming, Lanzhou, Nanjing, Ningbo, Qingdao, Qiqihar, Sanya, Shanghai, Shantou, Shenzhen, Shijiazhuang, Taiyuan, Tianjin, wuromqi, Wenzhou, Wuhan, Xiamen, Xian, Xuzhou, Yanji, Yantai, Zahuang, Zhengzhou, and Zhuhai.

Direct international flights go to Seoul and Cheongju, South Korea; Pyongyang, North Korea; Irkutsk and Khabarovsk, Russia; Osaka and Tokyo, Japan with connections to Frankfurt, Germany, Sydney, Australia, Los Angeles, USA and other cities. The airport is located 30 km south of the city. An airport shuttle runs from the airport to the main China Northern Airlines ticket office (Zhonghua Lu 117) and back frequently.

By car Shenyang is connected by a major expressway, the Jing-shen 6-lane Expressway, to the city of Beijing, some 658 kilometers away. Expressways also link Shenyang with Jilin and Heilongjiang provinces. The Shen-Da Expressway connecting Shenyang and Dalian was the first expressway ever built in China. It is the fastest highway (8-lane) linking one of the largest port cities to Shenyang.

12 By train Train service is provided through two train stations, the north train station and the south train station. The north train station handles the majority of the long-distance trains while the south train station handles most of the Dongbei (northeast) trains and does local service around the province. Some trains travel to both stations so be aware of which station you need to board your train.

North Train Station (Shenyang Bei Zhan) at Beizhan Lu

• Beijing - takes about 8 hours (or if you are lucky to get a ticket for the new fast train 'D Train', about 4 hours) • Dalian - takes about 6 hours (express trains take 3 - 4 hours) • Dandong - takes about 5 hours • Guangzhou - takes about 30-48 hours • Harbin - takes about 7 hours • Jilin - takes about 6-9 hours • Shanghai - takes about 27-34 hours • Tianjin - takes about 10-14 hours • Qinhuangdao - takes about 5-6 hours

South Train Station (Shenyang Huoche Zhan) at Shengli Beijie

• Baihe via Songjianghe - takes about 9 hours • Beijing - takes about 8-10 hours • Changchun - takes about 4 hours - 2 hours for express trains • Dalian - takes about 5 hours • Dandong - takes about 4 hours • Harbin - takes about 7 hours • Jilin - takes about 7 hours • Tianjin - takes about 9 hours • Tonghua - takes about 7 hours • Tumen - takes about 16 hours

13 By bus Long distance bus service is handled by the following two bus stations:

Main Long-Distance Bus Station (Shenyang Qiche Kuaisu Keyun Zhan): Located on Huigong Jie, this station serves routes to and from major cities. You can also find night buses to Beijing being hawked outside the north train station.

• Beijing - takes about 8 hours • Changchun - takes about 4 hours • Dandong - takes about 3 hours • Dalian - takes about 6 hours • Harbin - takes about 7 hours • Jilin - takes about 5 hours • Jinzhou - takes about 3 hours

South Bus Station: Buses queue along Minzhu Lu (opposite the south train station), around Minzhu Square and some nearby streets. On most buses you pay the conductor on the bus but for Anshan and Liaoyang you must purchase tickets at the office on the east side of Minzhu Square. These buses run medium distance routes to neighbouring cities within Liaoning. Buses stop departing this bus stations around 6pm. Passengers arriving to this station may be offloaded from the bus on a nearby street or some distance away at a yard used for storing coaches, sited behind the post office on Zhongshan Jie.

• Anshan - takes about 1 hour 40 minutes • Benxi - takes about 1 hour • Liaoyang - takes 1 hour • Fushun - takes 1 hour

14 Program Schedule

Tuesday, August 4th 2009

14:00‐20:00 Registration on second floor of Sheraton Hotel Shenyang

Wednesday, August 5th 2009

08:30‐18:00 Registration at Zhang Xue‐Liang Conference Hall Lobby 08:30‐09:00 Opening Ceremony at HQHT Auditorium 09:00‐10:00 Plenary Talk #1 by Professor David D. Yao 10:00‐10:20 Morning Break with tea and coffee 10:20‐12:00 Technical Sessions WM 12:00‐14:00 Lunch Break (University Town Restaurant buffet lunch) 14:00‐15:40 Technical Sessions WP 15:40‐16:00 Afternoon Break with tea and coffee 15:40‐16:40 Poster Sessions WPP 16:40‐17:40 Plenary Talk #2 by Professor Kazuhiro Kosuge 18:00‐20:30 Welcome Dinner at New Hong’s (Xin Hong Ji) Restaurant

Thursday, August 6th 2009

08:30‐18:00 Registration at Zhang Xue‐Liang Conference Hall Lobby 09:00‐10:00 Plenary Talk #3 by Professor Zhongtuo Wang 10:00‐10:20 Morning Break with tea and coffee 10:20‐12:00 Technical Sessions TM 12:00‐14:00 Lunch Break (University Town Restaurant buffet lunch) 14:00‐15:00 Plenary Talk #4 by Professor Warren B. Powell 15:00‐16:00 Poster Sessions TPP 15:40‐16:00 Afternoon Break with tea and coffee 16:00‐17:00 NEU 111 Project Panel Automation & Logistics Forum 17:00‐18:00 Plenary Talk #5 by Professor Yoshimi Takeuchi 18:00‐21:00 Award Banquet with performance (Banquet Hall of Liaoning Tower)

Friday, August 7th 2009

08:30‐12:00 Registration at Zhang Xue‐Liang Conference Hall Lobby 09:00‐10:00 Plenary Talk #6 by Professor Mengchu Zhou 10:00‐10:20 Morning Break with tea and coffee 10:20‐12:00 Technical Sessions FM 12:00‐14:00 Lunch Break (University Town Restaurant buffet lunch) 14:00‐15:20 Technical Sessions FP 15:20‐15:40 Afternoon Break with tea and coffee 15:40‐16:40 Poster Sessions FPP 16:40‐17:40 Poster Sessions FEP 18:00‐20:30 Farewell Party at Sheraton Hotel Shenyang

15 Floor Plan of Zhang Xue-Liang Conference Hall

Plenary talks and parallel technical sessions will be held at Zhang Xue-Liang Conference Hall on two different floors: ground floor (Level1) and third floor (Level 3).

For ground floor (Level1), will be used for Plenary talks, poster sessions on Wednesday Thursday and Friday.

Ground Floor (Level 1)

16 For third floor, function room HQHT 304, 308, 312, 313 and 314 will be used for all oral sessions on Wednesday, Thursday and Friday.

Third Floor (Level 3)

17 IEEE ICAL 2009 Plenary Talk I

09:00-10:00, August 5th (Wednesday) Han-Qing Conference Hall Auditorium

Resource Control in Stochastic Processing Networks

Professor David D. Yao Department of Industrial Engineering & Operations Research Columbia University, USA

Abstract

New models of stochastic networks in supply chain and logistics management often involve features that are beyond the capabilities of classical tools in control theory and operations research. These include, for example, simultaneous resource occupancy, sharing of service capacities, and the real-time resource allocation among different job classes. These features are widely present in applications ranging from internet and web servers to assemble-to-order and revenue management systems. We overview some of the recently developed methodologies and applications in this area, focusing on performance analysis, limiting regimes and asymptotic optimality.

18 Biography

David Yao has been on the faculty of Columbia University since his PhD in 1983, where he rose to full professorship in 1988 and was appointed to the Edison Chair in 1992. He has also held academic and professional appointments at Harvard, Yale, Tsinghua, Chinese University of Hong Kong, and IBM T.J. Watson Research Center.

He is an IEEE Fellow, an INFORMS Fellow, and a recipient of honors and awards including the Guggenheim Fellowship; the Presidential Young Investigator Award from the U.S. National Science Foundation; the Franz Edelman Award from the Institute for Operations Research and Management Sciences; the SIAM Outstanding Paper Prize from the Society for Industrial and Applied Mathematics; and from IBM, the Outstanding Technical Achievement Award, Research Division Award, Invention Achievement Award, and the IBM Faculty Award, Author/co-author of over 180 refereed publications, three books and five edited volumes, he has been a principal investigator of over thirty research grants and contracts from government and industrial sources. He is also a holder of six U.S. patents in manufacturing operations and supply-chain logistics.

19 IEEE ICAL 2009 Plenary Talk II

16:40-17:40, August 5th (Wednesday) Han-Qing Conference Hall Auditorium

iCART: Intelligent Cooperative Autonomous Robot Transporters

Professor Kazuhiro Kosuge Department of Bioengineering and Robotics Tohoku University, Japan

Abstract

Mechanical parking systems, such as elevator/tower parking systems, convey parking systems, shuttle parking systems, etc., are very popular in Japan especially inside of a crowded city. These systems are used to park automobiles efficiently in narrow space available for parking in department stores, hotels, etc., although users are required to maneuver their cars skillfully and guide their cars into a specified narrow place. iCART (Intelligent Cooperative Autonomous Robot Transporters) is a robot system which assists the users to maneuver their cars in a narrow space.

The current robot system consists of two mobile robots and each robot consists of three parts; a mobile base module, which is nonholonomic but having omni-directional mobility, a lifter module, which lifts up automobile without touching its body, and a

20 connecting module, which connects the mobile base module and the lifter module and plays both roles of a suspension for the system and a three degrees of freedom force/torque sensor. The use of two mobile robots enables the robot system to handle a variety of cars having different dimensions.

Coordination of multiple mobile robots handling an automobile together is challenging. The slippage between the mobile base and the ground is unavoidable and odometry could not be used for the coordination of multiple robots having physical interaction. The estimation error of robot pose caused by the slippage could damage the automobile. A new coordinated motion control scheme has been designed inspired by a caster mechanism and implemented in the robots to overcome the problem. Several laser range finder systems have been also installed for localization of robots and the automobile and for robot motion planning. Through the development of iCART, we will discuss issues relating to robotics as systems integration.

Biography

Kazuhiro Kosuge is a Professor in the Department of Bioengineering and Robotics at Tohoku University, Japan. He received the B.S., M.S., and Ph.D. in control engineering from the Tokyo Institute of Technology, in 1978, 1980, and 1988 respectively. From 1980 through 1982, he was a Research Staff in the Production Engineering Department, Nippon Denso Co., Ltd. (DENSO Co., Ltd. at present). From 1982 through 1990, he was a Research Associate in the Department of Control Engineering at Tokyo Institute of Technology. From 1990 to 1995, he was an Associate Professor at Nagoya University. From 1995, he has been at Tohoku University. He received the JSME Awards for the best papers from the Japan Society of Mechanical Engineers in 2002 and 2005, the RSJ Award for the best papers from the Robotics Society of Japan in 2005. He is an IEEE Fellow, a JSME Fellow, a SICE Fellow and RSJ Fellow. He is IEEE RAS President Elect for 2008-2009.

21 IEEE ICAL 2009 Plenary Talk III

09:00-10:00, August 6th (Thursday) Han-Qing Conference Hall Auditorium

Supernetwork and its Applications

Professor Zhongtuo Wang School of Management Dalian University of Technology, China

Abstract

In this presentation a new concept Supernetwork and its applications are introduced. Supernetwork can be referred as Network of networks and has the following features: - Multi-tiered; - Multi-level; - Multi-mode network flows; - Congestion; - Alternative behavior of users of the network; - Multi-criteria; and - Conflict between user’s optimization and system optimization. Concept of supernetwork has a wide range of applications and only a small part of those applications has been explored thus far. Some specific applications of supernetworks are: supernetworks consisting of social networks interacting with supply chain networks, supernetworks consisting of social networks interacting with financial networks, and knowledge supernetworks. This framework captures the different interacting networks in one model. It allows one to compute optimal solutions under different scenarios and to test how the equilibrium will change

22 when certain cost and benefit functions are changed. Application areas of supernetwork models: • Telecommuting/Commuting Decision-Making • Teleshopping/Shopping Decision-Making • Supply Chain Networks with Electronic Commerce • Financial Networks with Electronic Transactions • Reverse Supply Chains with E-Cycling • Energy Networks/Power Grids • Knowledge Networks

Biography

Zhongtuo Wang is the professor of School of Management, Dalian University of Technology (DUT), head of the PhD Program of Systems Engineering, director of Research Center of Knowledge Science and Technology, DUT. He is the member of Chinese Academy of Engineering. He was the vice-president of Systems Engineering Society of China. In fifties of last century, he joined the department of Electrical Engineering, Dalian University of Technology. As the founder of Department of Control Engineering of DUT, he made a lot of contributions to the teaching and research works in the field of Optimizing control and Computer applications. In the year of 1977, he moved to the area of Systems Engineering. As one of the originators of PhD Program and research works of Systems Engineering in China and founder of Institute of Systems Engineering DUT, he devoted himself to the task of theoretical research in decision analysis, complex adaptive system and network optimization. As a leader he has organized a lot of projects of applying methodology and techniques of Systems Engineering to the Chinese economic and engineering endeavors, including the strategic analysis of regional economic development, production planning of the petroleum refinery, planning and scheduling of the construction projects, and impact of information technology to the management transformation. At the same time, he organized the education task of MS and PhD program of Systems Engineering and Management Science in DUT. In the years 1986-1988, He worked in International lnstitute for Applied Systems Analysis (lIASA) in Vienna, Anstria as research scholar and head of international collaborative project and known internationally for his contribution. He has published 10 books and 9 translations and more than 170 papers and reports. He received 2 national awards, 9 awards from ministries of Chinese government. He is now working in Knowledge Management and Technological Innovation.

23 IEEE ICAL 2009 Plenary Talk IV

14:00-15:00, August 6th (Thursday) Han-Qing Conference Hall Auditorium

Approximate Dynamic Programming for High-Dimensional Resource Allocation Problems

Professor Warren B. Powell Department of Operations Research and Financial Engineering Princeton University, USA

Abstract

Managing fleets of trucks, locomotives, business jets, vaccines, medical resources and energy all share something in common: they are all high-dimensional stochastic resource allocation problems. They can be formulated as dynamic programs with state, information and action variables with thousands or even millions of dimensions, a characteristic we refer to as the "three curses of dimensionality." Classical techniques in approximate dynamic programming solve only part of the problem. We show how the use of the post-decision state variable allows us to break a problem into three distinct components: simulation, deterministic optimization and statistical learning. This strategy decomposes problems over time, allowing us to use commercial optimization packages to

24 solve the problem at each point in time using value function approximations to produce good solutions over time. The challenge is designing and estimating value function approximations. I will describe a simple method that works well for a wide range of problems, and discuss some of the algorithmic challenges that we are still facing. I will describe several applications of ADP in transportation and energy policy modeling.

Biography

Warren B. Powell has been a faculty member at Princeton University since 1981, with specialization in stochastic optimization problems arising in a variety of resource allocation problems, with applications encompassing freight transportation, military operations, energy resource management, health and finance. He founded and currently heads CASTLE Laboratory within the Department of Operations Research and Financial Engineering at Princeton where he has developed advanced optimization models and algorithms for some of the largest freight transportation companies in the country.

His research achievement includes fundamental contributions to stochastic optimization, computational advances in the solution of large-scale stochastic optimization problems, and the formulation and solution of complex resource allocation problems arising in transportation. He has also recently begun a new line of research in optimal learning that addresses the challenges of collecting information in an efficient way. These problems arise in drug discovery and medical testing, business, R&D portfolio optimization, engineering design, optimization of expensive simulations, and biosurveillance, to name a few.

He has authored or coauthored over 140 refereed publications, and he is the author of Approximate Dynamic Programming: Solving the curses of dimensionality, published by John Wiley and Sons. This research has led to the first stochastic, multiscale model for energy policy analysis, and his techniques have been used in applications in finance and health. His work in freight transportation has spawned two consulting firms, and he was twice a finalist in the Edelman competition. A recipient of the Informs Fellows Award, Professor Powell has served in a variety of editorial and administrative positions for Informs, including Informs Board of Directors, Area Editor for Operations Research, President of the Transportation Science Section, and numerous prize and administrative committees.

25 IEEE ICAL 2009 Plenary Talk V

17:00-18:00, August 6th (Thursday) Han-Qing Conference Hall Auditorium

Automatic Fabrication of Ultraprecision Microparts with Complicated Shape

Professor Yoshimi Takeuchi Department of Mechanical Engineering Osaka University, Japan

Abstract

Ultraprecision micromachining is a key technology to realize the fabrication of microparts, microrobots, etc. Until now, etching and lithography based technologies are well known in the microfabrication fields, where such a material as silicon oxide is generally utilized. A large problem of these technologies is the difficulty of producing 3-dimensional shapes with inclined and sculptured surfaces. Utraprecision mechanical micromachining technology is utilized to meet the requirement, using a diamond cutting tool and an ultraprecision machine tool such as a turning lathe or a milling machine, especially with the multi-axis control machining function. Multi-axis control machine tools are one solution for improving productivity and accuracy, however, it is difficult to actually implement due to the interference between a cutting tool and a workpiece. Thus, the

26 speech presents the current state of multi-axis control ultraprecision machine tools in Japan and a machined example of small workpieces with a complicated shape by making use of multi-axis control machining function. In order to set a cutting tool at an arbitrary position with an arbitrary attitude, 6 degrees of freedom are required, that is, 3 translational positioning mechanisms, X, Y, and Z, and 3 rotational positioning mechanisms, A, B, and C around each corresponding translational axis. Usually, conventional machine tools consist of 3 translational axes. On the contrary, multi-axis control machine tools stand for machine tools having more than 3 control axes. 5-axis control machine tools, the representative of multi-axis control ones, are usually equipped with 3 translational axes and 2 rotational axes, for example, X, Y, Z, A, and B. The axis C is used as the main spindle to rotate a diamond cutting tool or a workpiece so that high cutting speed can be realized. In case of 6-axis control machining, non-rotational diamond cutting tools are used, which enable the characteristic machining in spite of extremely low cutting speed. In general, ultraprecision machine tools have the translational positioning accuracy of 1 nm and the rotational one of 0.00001 degree. In Japan, there are currently more than 5 ultraprecision machine tool builders. Ultraprecision machine tools can be classified into several types in terms of driving mechanism of movement axes such as ball- screw, pneumatic/hydrostatic screw or linear drive, and supporting mechanism/bearing such as ball/roller bearing or pneumatic/hydrostatic bearing. In order to fabricate microparts with complicated shape, advanced CAM systems are essential. Such a CAM system can generate NC data for ultraprecision machine tools, based on 3-D CAD systems. In other words, it is impossible to create microparts with complicated shape without advanced CAM systems. As a representative of complicated micromachines, there is a miropump useful for water-cooling of heat generation from CPU, fuel-supplying to fuel cells, blood circulation and so on. Let us introduce the machining of an inducer of axial flow type micropump for blood circulation. Then, several machined results by means of our own multi-axis control CAM system for ultraprecision milling are presented.

Biography

Yoshimi Takeuchi is a professor of the Department of Mechanical Engineering, Graduate School of Engineering at Osaka University. He graduated from the University of Tokyo in 1971, and received Master Degree in 1973 and Doctor Degree in 1976 respectively. He was a visiting researcher of Machine Tool Laboratory at Technical University of Aachen, Germany during 1978.4 to 1979.3 under the support of Alexander von Humboldt foundation. He is a fellow of CIRP, editor-in-chief of International Journal of Automation Technology and International Journal of Precision Engineering. He received Paper Awards from JSPE at 1997, 2000 and 2008, and from JSME at 1998 and 2004, respectively. He is now a vice-president of JSPE. His research interests are the development of ultraprecision micromilling technology to make small size and complicated shape parts, and the application of 3D-CAD/CAM to multi-axis control machining.

27 IEEE ICAL 2009 Plenary Talk VI

9:00-10:00, August 7th (Friday) Han-Qing Conference Hall Auditorium

Modeling, Scheduling and Control of Cluster Tools in Semiconductor Manufacturing Automation

Professor Mengchu Zhou Department of Electrical and Computer Engineering New Jersey Institute of Technology, USA

Abstract

Cluster tools provide a flexible, reconfigurable, and efficient environment for semiconductor manufacturing. They become difficult to operate because of residency time constraints and process time variation. This talk addresses their modeling, scheduling and control issues. A generic Petri net model is developed to model them. It describes the robot activity sequence with robot waits included. Hence, to operate a cluster tool is to determine robot wait times. A two-level operational architecture is proposed and discussed. It includes an off-line optimal periodic scheduler and real-time controller. This proposed approach allows a cluster tool to adapt to activity time variation while operate at its highest throughput at the steady-state. The presented Petri net

28 models play a critical role in ensuring proper transient processes and optimal steady-state operations of cluster tools. The advantages of the propose methodology over the existing Petri net methods and mathematical programming approaches are discussed.

Biography

MengChu Zhou is a Professor of Electrical and Computer Engineering and the Director of Discrete-Event Systems Laboratory at New Jersey Institute of Technology. His research interests are in Petri nets, semiconductor manufacturing, sensor networks, system security, and life-cycle engineering product design. He has over 300 publications including 9 books, 130+ journal papers, and 17 book-chapters. Based on the data analysis from Scopus database (the world’s most extensive database covering science and engineering papers), in the area of Petri nets, Dr. Zhou ranks the top one in terms of the number of publications; and top two in terms of the number of citations. He ranks the top one in the area of automated manufacturing systems in both indices. He is Editor of IEEE Transactions on Automation Science and Engineering, Associate Editor of IEEE Transactions on Systems, Man and Cybernetics: Part A and IEEE Transactions on Industrial Informatics. He was General Co-Chair of 2003 IEEE International Conference on System, Man and Cybernetics, Washington DC, October 5-8, 2003, Founding General Co-Chair of 2004 IEEE Int. Conf. on Networking, Sensing and Control, Taipei, March 21-23, 2004, and General Chair of 2006 IEEE Int. Conf. on Networking, Sensing and Control, Ft. Lauderdale, Florida, U.S.A. April 23-25, 2006 and IEEE Conf. on Automation Science and Engineering, Washington D.C., August 23-26, 2008. He was Program Chair of 1998 and 2001 IEEE International Conference on System, Man and Cybernetics (SMC) and 1997 IEEE International Conference on Emerging Technologies and Factory Automation. Dr. Zhou was the recipient of NSF’s Research Initiation Award, CIM University-LEAD Award by Society of Manufacturing Engineers, Perlis Research Award by NJIT, Humboldt Research Award for US Senior Scientists, Leadership Award and Academic Achievement Award by Chinese Association for Science and Technology-USA, Distinguished Lecturer of IEEE SMC Society, and Chang Jiang Scholars Program Award from PRC Ministry of Education. He was the founding chair of Discrete Event Systems Technical Committee of IEEE SMC Society, and founding chair of Semiconductor Manufacturing Automation Technical Committee of IEEE Robotics and Automation Society. He is a Fellow of IEEE.

29 NEU 111 Project Panel Automation & Logistics Forum

16:00 - 17:00, August 6th, 2009 Han-Qing Conference Hall Auditorium

Chairs: Tianyou Chai and Tzyh-Jong Tarn

Panelists

Tianyou Chai, Northeastern University, China Tongwen Chen, University of Alberta, Canada Toshio Fukuda, Nagoya University, Japan Sam Shuzhi Ge, National University of Singapore, Singapore Kazuhiro Kosuge, Tohoku University, Japan Wei Lin, Case Western Reserve University, USA Derong Liu, Universioty of Illinois at Chicago, USA Max Q.-H. Meng, The Chinese University of HongKong, China Warren B. Powell, Princeton University, USA Sizhao Qin, University of Southern California, USA Chunyi Su, Concordia University, Canada Tzyh-Jong Tarn, Washongton Univerrsity, USA Yoshimi Takeuchi, Osaka University, Japan Hong Wang, University of Manchester, UK Zhongtuo Wang, Dalian University of Technology, China David D. Yao, Columbia University, USA Hong Zhang, University of Alberta, Canada Mengchu Zhou, New Jersey Institute of Technology, USA

The Northeastern University 111 project participants and the plenary speakers of IEEE ICAL 2009 conference, with research interests in the areas of automation and logistics, will form the panel of this forum to share and discuss their perspectives, experience, and advices on conducting high- impact engineering research in the general research areas of automation and logistics. Important issues common to various engineering disciplines and specific issues in automation and logistics will be discussed. This forum is open to all participants of the conference. It will be particularly useful for graduate students as well as and researchers and professors in the early stages of their career development. The panel will entertain questions from the audience during the forum.

30 Symbols for Technical Sessions

Symbols for Days of a Week W = Wednesday T = Thursday F = Friday

Symbols for Session Time Slots Session XM = 10:20 - 12:00 Session XP = 14:00 - 15:40 Session XPP = 15:40 - 16:40 Session XEP = 16:40 - 17:40

Symbols for Room Assignments HQHT = Han-Qing Conference Hall

Room ID 1 2 3 4 5 P

HQHT HQHT HQHT HQHT HQHT HQHT Room 304 308 312 313 314 Hallway

31 Session IDs, Session Titles & Session Chairs

Wednesday August 5th, 2009

Morning Session Title Session Chairs WM-1 Robotics I Aziza Zaki Jong Hyeon Park WM-2 Modeling I Jason Gu Na Xu WM-3 Automatic control I Qiang Wei Hong Wang WM-4 Intelligent control and automation I Sam Ge Simon Yang WM-5 Supply chain management Qiqi Sun Chao Hu Afternoon WP-1 Robotics II Lingfei Wu Yangmin Li WP-2 Modeling II Mingzhu Zhang Dayou Li WP-3 Automatic control II Donglian Qi Shi Jin WP-4 Intelligent control and automation II Yuqing Chen Fujun He WP-5 Logistics planning Na Xu Chao Hu WPP Automation Chao Hu Yanjun Shen

Thursday August 6th, 2009 Morning Session Title Session Chairs TM-1 Robotics III Xuezhang Hou Aiguo Ming TM-2 Modeling III Sam Ge Simon Yang TM-3 Distribution systems Dengpan Zhang Xiaotong Du TM-4 Process automation Chao Hu Hong Wang TM-5 Modeling and applications Ming Zhong Xiaohua Yu Afternoon TPP System modeling Guohui Tian Xianming Li

Friday August 7th, 2009 Morning Session Title Session Chairs FM-1 Optimization I Xiaohua Yu Dayou Li FM-2 Simulation I Ming Zhong Zaiwen Liu FM-3 Image processing Yangmin Li Fei Lu FM-4 Network-based systems I Yingwei Zhang Chunyi Su FM-5 Planning and scheduling Yingzi Wei Ze Tao Afternoon FP-1 Optimization II Jing Bai Shuli Sun FP-2 Simulation II Yunong Zhang Yanjun Shen FP-3 Intelligent transportation system Chenjin Zhang Hong Zhang FP-4 Network-based systems II Aiguo Ming Yingwei Zhang FPP Transportation systems Jason Gu Simon Yang Evening FEP Modeling and simulation Yibin Li Chengjin Zhang

32 Oral Session

Wednesday August 5, 2009

WM-1 Robotics I

WM-2 Modeling I

WM-3 Automatic control I

WM-4 Intelligent control and automation I

WM-5 Supply chain management

WP-1 Robotics II

WP-2 Modeling II

WP-3 Automatic control II

WP-4 Intelligent control and automation II

WP-5 Logistics planning

WM-1:Robotics I

Session Chair: Aziza Zaki and Jong Hyeon Park

HQHT-304 10:20-12:00 Wednesday, August 5th

HQHT-304(1) 10:20-10:40 HQHT-304(2) 10:40-11:00 Robotic Gripper Based on Advanced System Continuous Steps toward Humanoid Push Set-up and Fuzzy Control Algorithm Recovery *A. M. Soliman, *A. M. Zaki, **A.M. El-Shafei, and **O. A. Mahgoub *Electronics Research Institute, Egypt **Faculty of Engineering, Cairo University, Egypt Wentao Mao, Jeong-Jung Kim and Ju-Jang Lee KAIST , Korea • Gripper importance appears in hazardous • Continuous Steps Method proposes and routine work. a method which employs continuous • Applying the minimum force to grasp and steps reaction to recover humanoid handle unknown object without dropping it from a big push. or smashing it is the main purpose of this research. • Continuous Steps Method has more • New grasping scenario using fuzzy logic stable and soft responses compared to the previous methods. controllers is developed and presented. 1- Drive card for the actuator 2- BLDC motor as an actuator • Economical system configuration is presented 3- Movable finger • Mathematical Dynamic modeling of 4- Fixed finger and good system performance was obtained 5- Object which will be handled steps proposes a reasonable, strong 6- Force sensor 7- Force Sensor and reliable reference to humanoid through experimental results. 8- Slip Sensor Robotic Gripper push recovery.

HQHT-304(3) 11:00-11:20 HQHT-304(4) 11:20-11:40 Predictive Function Cascade Control Scheme Object-Centric Programming and its Application to Hydraulic Robot Systems Andreas Angerer, Alwin Hoffmann, Frank Ortmeier, Michael Vistein and Wolfgang Reif Zhihuan Zhang Institute for Software and Systems Engineering, University of Augsburg Ningbo Institute of Technology, Zhejiang University Augsburg, Germany Ningbo, China • Predictive • Manipulator-level programming is often Disturbance Function Cascade Base function inflexible and non-intuitive Control Scheme . )k(r )k(y Ref. Opt. PFC • Goal: Focus on the task that shall be Model . Plant • PFC-P controller accomplished and objects in this task for hydraulic robot . Model α )k(y • Goal can be accomplished by a proper object-oriented robotic architecture Æ “Object-Centric Programming”

Predictor & Error emendation • Example: Palletizing Box1.Bottom.Move(Box2.Top) Block scheme of predictive A central step in palletizing tasks function control

HQHT-304(5) 11:40-12:00 The Model-Based Iterative Learning Control for Industrial Robot Manipulators HQHT-312 HQHT-313 HQHT-314 Je Sung Yeon, Jong Hyeon Park, Seung-Woo Son, and Sang-Hun Lee School of Mechanical Engineering, Hanyang University, Korea

• This paper proposes a model- based iterative learning control

(MB-ILC) for industrial robot. HQHT-308 • The proposed MB-ILC modifies the desired trajectory but not the control input signals. • To achieve stability of the MB-ILC algorithm in the HQHT-304 HQHT-305 HQHT-306 iteration domain, two filters 6DOF Industrial Robot Han-Qing Conference Hall - Level 3 are used. (东北大学汉卿会堂)

O-1 WM-2:Modeling I

Session Chair: Jason Gu and Na Xu

HQHT-308 10:20-12:00 Wednesday, August 5th

HQHT-308(1) 10:20-10:40 HQHT-308(2) 10:40-11:00 Modeling and Simulation of Yard Trailer Kinematics Modeling and Simulation for Dispatching at Container Terminals an Amphibious Robot: Bin Li, Wen-feng Li, Yu Zhang, Yan-hong Ge, Huan Chen and Xiao-lei Liang School of Logistics Engineering, Wuhan University of Technology Design and Implementation Wuhan, China Rui Ding, Junzhi Yu, Qinghai Yang, and Min Tan • Whether yard trailer dispatching at Key Laboratory of Complex Systems and Intelligence Science container terminals is sound or not Institute of Automation, Chinese Academy of Sciences, Beijing, China possesses of important significance to • Wheel-driving mechanism improve on the operational efficiency. • Kinematics analysis on • This paper utilizes hybrid flow shop with blocking based on attributes to land build the mathematical model of yard • Kinematics model trailer dynamic dispatching and presents the corresponding established by Solidworks formalization model based on Harvard and ADAMS architecture and multi-agent. • Simulation of straight- • The following simulations validate the going, turning, and slope feasibility and creditability of the above an amphibious robot modeling methodology. Modeling Framework of YTDD climbing capable of multi-mode motion

HQHT-308(3) 11:00-11:20 HQHT-308(4) 11:20-11:40

The Finite Element Model of Transient PSO Algorithm for a Single Machine Scheduling Temperature Field of Airplane Brake Disks with Problem with Batching in Chemical Industries Ping Yan, and Lixin Tang Rough Surface Profile Liaoning Key Laboratory of Manufacturing System and Logistics, The Logistics Institute, Xue Jing, Li Yuren, Liu Weiguo, Tian Guanglai and Wei Hanbing Northeastern University College of Automation, Northwestern Polytechnical University Shenyang, China Xi’an, China • A single machine batch process scheduling problem (SBPSP) integrating batching decision is • Contact model of surface roughness investigated. The integration problem of batching and scheduling is to • Finite element model of allocate the demands from different customer orders to sets of batches and temperature field of brake disks. schedule these batches such that the total weighted tardiness costs and the • Performance simulations in ANSYS total set-up costs are minimized. The Temperature Field • A PSO algorithm incorporated AIA mechanism is proposed to solve this The Particle Swarm problem effectively.

HQHT-308(5) 11:40-12:00 Assigning Skus to Multiple Automated-Picking

Areas over Multiple Periods HQHT-312 HQHT-313 HQHT-314 Peng Liu, Yaohua Wu, Na Xu School of Control Science and Engineering, Shandong University Jinan, China • Based on fluid model, the optimal allocation of storage resources to each sku for the automated picking system is presented to minimize the total restocking cost. HQHT-308 • For a given period, greedy heuristic algorithm is developed to assign skus to automated-picking areas. • Sequence viscosity is formulated to evaluate skus over multi-period by

minimizing the total cost of picking, Automated Picking System HQHT-304 HQHT-305 HQHT-306 restocking, and reslotting. • Efficient heuristic algorithms are Han-Qing Conference Hall - Level 3 adopted to assign skus to automated- (东北大学汉卿会堂) picking areas over multiple periods.

O-2 WM-3:Automatic control I

Session Chair: Qiang Wei and Hong Wang

HQHT-312 10:20-12:00 Wednesday, August 5th

HQHT-312(1) 10:20-10:40 HQHT-312(2) 10:40-11:00 Neural Network Control of Scanning Probe Wall Climbing Robot Enabled by a Novel and Microscope Precision Stage Robust Qiang Wei, Chengzhong Hu, Dong Zhang, and Xianming Li Vibration Suction Technology Deptartment of Physics and Electronic Science, Taishan University, Taian, China Qingfeng Hong , Rong Liu, Hao Yang and Xudong Zhai School of Control Science and Engineering, Shandong University, Jinan, China Robotics Institute, Beijing University of Aeronautics & Astronautics Beijing, China • Precision stage is actuated by the 16 piezoelectric ceramics, and a dynamics • A new robotic foot with higher model based on Neural Network is given. 14 suction performance was designed 12 m)

• The traditional PID controller is replaced μ 10 based on Vibration Suction Method. by a neural network PID controller. Weights in the network and parameters 8 • A mini multi-joint wall climbing in the PID controller are adjusted. 6 robot was built based on the new • Experimental verification and validation Displacement( 4 robotic foot and gait planning was of the proposed method using a nano- 2 0 made. positioning stage is designed. 1 1.1 1.2 1.3 1.4 1.5 1.6 • The response time is reduced from 0.38s Time(s) • The robot can walk from the ground to 0.18s, and the steady-state error is Output in different control schemes to the wall, crawl and turn on the Solid for neural network PID control, dashed for traditional shorten from 4.2% to 1.9% under a wall and then walk from the wall to The Mini Multi-joint Wall displacement of 10μm. PID control, and dotted for desired displacement the ceiling. Climbing Robot

HQHT-312(3) 11:00-11:20 HQHT-312(4) 11:20-11:40 Design of Plant-Friendly PID Controllers Based Concise Nonlinear Adaptive Robust Control of on Constrained Optimization ship Steering Systems Dawei Shi, Junzheng Wang, Liling Ma, and Jiangbo Zhao Guan Wei, Zhang Xian-ku and Wang Xin-ping School of Automation, Beijing Institute of Technology, Beijing, China Laboratory of Marine Simulation and Control, Dalian Maritime University, China

• Plant-friendliness is evaluated based • For ship steering systems, a

on the total variations of the control (deg) CNARC derived from Backstepping ψ variable. design method based on CGSA. • The controller is designed by • The CNARC adaptive nonlinear optimizing plant-friendliness while compensator overcomes the high angle Yaw Ship considering transient response and conservative of the pure nonlinear The navigation time t (sec) robustness. robust. (deg) (deg) • The trade-off between the transient • The CNARC design procedure is δ performance and plant-friendliness is simple and direct, easy to reach analyzed. information symmetric.

• Applications to typical process • Through the simulations, we can angle rudder Ship models illustrate the effectiveness of see that the CNARC has a superior The navigation time t (sec) the proposed method. performance.

HQHT-312(5) 11:40-12:00 Semi-parametric Adaptive Control of Discrete-time Nonlinear Systems HQHT-312 HQHT-313 HQHT-314 C. Yang1, T. Chai2, L. Zhai2, S. S. Ge1 and T. H. Lee1 1.ECE Dept. National University of Singapore, Singapore 2. Research Center of Automation, Northeastern University, Shenyang, China

Asymptotical tracking adaptive control for a class of nonlinear discrete-time systems with unknown vector parameter and perturbed by nonparametric nonlinear mode uncertainty is studied.

HQHT-308 • When the control gain is partially known, a novel deadzone with threshold converging to zero has been proposed to modify the traditional gradient update law for parameters estimation, while an auxiliary output including both uncertainties is introduced to facilitate the nonparametric uncertainties compensation. • When the control gain is totally unknown, discrete Nussbaum gain is incorporated into the design method to make the adaptive control insensitive to the control

direction. HQHT-304 HQHT-305 HQHT-306 • It is rigorously proved that the designed adaptive control guarantees the boundedness of all the closed-loop signals and achieves asymptotic tracking Han-Qing Conference Hall - Level 3 performance. (东北大学汉卿会堂)

O-3 WM-4:Intelligent control and automation I

Session Chair: Sam Ge and Simon Yang

HQHT-313 10:20-12:00 Wednesday, August 5th

HQHT-313(1) 10:20-10:40 HQHT-313(2) 10:40-11:00 Neural Network Tracking Control of Ocean Decentralized Cooperative Control for Swarm Surface Vessels with Input Saturation Agents with High-Order Dynamics Mou Chen, Shuzhi Sam Ge and Yoo Sang Choo B. Ren, S. S. Ge and T. H. Lee H. L. Pei and Z. D. Sun Department of Electrical and Computer Engineering NUS, Singapore SCUT, Guangzhou, China National Universityof Singapore, Singapore • Objective: To drive a swarm [m]

• Design of an approximation based xd

η 1 of multi-agent systems with - x η control in combination with back- 0 high-order nonlinear dynamics into a moving target stepping control. -1 tracking error 0 20 40 60 80 100 120 140 160 180 200 region while avoiding

[m] 2

• Account for control input saturation yd

η collisions among themselves.

- 1 y explicitly in the control design stage η 0 • Control design is based on a -1 for both full state feedback control -2 fusion of two kinds of new tracking error 0 20 40 60 80 100 120 140 160 180 200 potential functions,

and output feedback control. [rad] 1 d ψ η 0.5 backstepping technique and - ψ • Rigorous stability analysis via Lya- η 0 Lyapunov synthesis. -0.5 punov analysis which shows that the -1 0 20 40 60 80 100 120 140 160 180 200 • Flexibility and scalability are tracking error semiglobal uniform boundedness of Ti me [s] improved. the closed-loop signals are guaranteed. The Tracking error Simulation Results

HQHT-313(3) 11:00-11:20 HQHT-313(4) 11:20-11:40

Nonlinear Systems Identification and Control using Inverse Rate-Dependent Prandtl-Ishlinskii Model Dynamic Multi-Time Scales Neural Networks for Hysteresis Nonlinearities Compensation Xuan Han and Wen-Fang Xie, Member, IEEE M. Al Janaideh, S. Rakheja, C-Y. Su Department of Mechanical & Industrial Engineering, Concordia University, Montreal, Canada Department of Mechanical and Industrial engineering, Concordia University,

• On-line identification algorithm via • Inverse of the rate-dependent 6 dynamic neural networks with Prandtl-Ishlinskii model is proposed different time-scales to compensate rate-dependent 4 • Adaptive control method for hysteresis effects in smart actuators. 2

Dynamic systems with nonlinearity • The rate-dependent hysteresis 0 and uncertainty properties were characterized using -2 • Lyapunov function analysis, the rate-dependent play operators . -4 singularly perturbed technique and • The inverse model is subsequently -6 sliding mode methodology obtained analytically from the rate- -10 -8 -6 -4 -2 0 2 4 6 8 10 • Guarantee trajectory tracking and dependent play operators. stability Identification and Hysteresis Control Scheme

HQHT-313(5) 11:40-12:00 Fuzzy Based Control of Dissolved

Oxygen Concentration in a Bioreactor HQHT-312 HQHT-313 HQHT-314 for Wastewater Applications Jesse Rocca and Simon X. Yang Advanced Robotics and Intelligent Systems Laboratory, University of Guelph Guelph, Ontario, Canada • A new fuzzy controller for robust control of the dissolved oxygen concentration in a wastewater bioreactor is presented HQHT-308 • The fuzzy controller is developed and tested using first order plus dead time models obtained from a real bioreactor • The fuzzy controller performance is benchmarked against a similarly tuned PID controller and shows several enhancements HQHT-304 HQHT-305 HQHT-306 Han-Qing Conference Hall - Level 3 (东北大学汉卿会堂)

O-4 WM-5:Supply chain management

Session Chair: Qiqi Sun and Fei Lu

HQHT-314 10:20-12:00 Wednesday, August 5th

HQHT-314(1) 10:20-10:40 HQHT-314(2) 10:40-11:00

A Joint Decision Model of Inventory Control and Promotion Optimization based on Demand Forecasting Studies on Fashionable Products’ Demand Gao Jun-jun, Kang Ting Forecast and Stock Control Sydney Institute of Language and Commerce, Shanghai University Shanghai, China Demand • We study the problem of jointly determining the promotion 100 品种1拟合需求值 品种2拟合需求值 品种3拟合需求值 品种1实际需求值 品种2实际需求值 optimizing and inventory control of multiple beer product variants 品种3实际需求值 90 in a Chinese retail supermarket based on demand forecasting. 80 • We examine the joint effect of two modes of promotional policies: price discount and inventory stimulation. 70 • A numerical solving algorithm is proposed and an experimental 60

example is given. 50

40 • Our experimental analysis reveals that the two-stage joint 0 5 10 15 20 25 30 Time/day35 optimization model has good feasibility and robustness and Fig1.The Actual Demand Data and appropriate inventory control strategies and promotional methods Fitted Demand Data would achieve higher profit. Table 3.The Comparison between the Promotion Period and the Non-promotional Period in Pricing, Shelf-space and the Total Profit Non-promotion period Promotion period Profit Item Product Shelf- total Promotion Discount Profit of each growt Shelf-space h rate i price space profit strategy rate brand Promote 1th 1 3.0 130 (0.6,1,1) (180,140,30) 11896 59.2% brand Promote 2th 2 3.1 140 7472 (1,1,1) (130,190,30) 12423 66.3% brand Promote 3th 32.830 (1,0.7,1) (130,140,80) 9377 25.5% brand

HQHT-314(3) 11:00-11:20 HQHT-314(4) 11:20-11:40 The study of inter-organizational collaboration Research on Collaborative Quality Management by cluster supply chain System into Supply Chain Xiao Xue, Biqing Huang and Tianyuan Xiao National CIMS Engineering Research Center, Tsinghua University Xingyu Jiang Beijing, China School of Mechanical Engineering, Shenyang University of Technology Shenyang, China • Cluster supply chain is a set of networks comprised of horizontal ties between To cope with the challenges of traditional quality management system firms within a particular industrial applied in supply chain, which consists of internal practices that are cluster, which are sequentially arranged contained within a firm, external practices that cross organizational based on vertical ties between firms in boundaries integrating a firm with its customers and suppliers, a different layers. collaborative quality management system model into supply chain, and corresponding functional modules and framework were all put forward. • Cluster supply chain analysis integrates This mode dealt mainly with constructing and running collaborative supply chain and network perspectives quality management system into supply chain, integrated several enabling on inter-organizational collaboration technologies such as collaborative-supported-computer-work), Statistical with particular emphasis on the value Process Control and comprehensive evaluation. Furthermore, some of key creating and coordination mechanism. enabling technologies were studied in detail, including collaborative quality design based on CSCW, dynamic process quality control, quality evaluation oriented to product lifecycle. The prototype system of • We posit that sources of value and collaborative quality Management system into supply chain is developed, coordination mechanisms correspond to which was introduced to demonstrate the rationality and validity of the distinct types of collaborations: vertical, method as an example of some Heavy Machinery Ltd. It provided an pooled horizontal and reciprocal An example of cluster supply chain effective method with enterprise implementing supply chain management. horizontal.

HQHT-314(5) 11:40-12:00 Response to the Supply Chain Disruptions with Multiple Sourcing HQHT-312 HQHT-313 HQHT-314 ZHOU Xiaoqiang and FANG Huijiang Jiangxi University of Finance and Economics Nanchang, Jiangxi Province, China

Toyota • In a two-echelon supply chain brake Asia Long- suppli economic Network Philips Fisland crisis strikes and er fire economi fire’s o Hur consisted of one purchaser and two cfoam impact on rWest Coast rica Ericsson t Port I ne and Nokia ’closure Kat s r homogeneous suppliers, the a rina q and t Rita y purchaser could synchronously order r w e a r from two suppliers, but the suppliers r HQHT-308 e 1997 1998 1999 2000 2001 200 c 200 2004 200 2006 maybe don’t fulfill the orders 2 a 3 5 l UPS GM Taiwa 911 l Seabed because of emergencies. strike strike n incident SAR cable earth- S disruption quake by Taiwan • If choosing the multiple sourcing earthquake strategy, the purchaser orders equal- quantity materials from two suppliers. • The total order quantity of multiple HQHT-304 HQHT-305 HQHT-306 sourcing is greater than single sourcing. Multiple sourcing can Han-Qing Conference Hall - Level 3 provide backup inventory to the (东北大学汉卿会堂) purchaser.

O-5 WP-1:Robotics II

Session Chair: Lingfei Wu and Yangmin Li

HQHT-304 14:00-15:40 Wednesday, August 5th

HQHT-304(1) 14:00-14:20 HQHT-304(2) 14:20-14:40 Bio-inspired Design and Realization of a Novel Accurate Localization in Combination with Multimode Amphibious Robot Wireless Sensor Networks and Laser Localization Weibing Wang1, Junzhi Yu2, Rui Ding2, and Min Tan2 Lingfei Wu, Max Q.-H. Meng, Huawei Liang, Wen Gao 1Shihezi University, Xinjiang Uygur Autonomous Region 832003, China Center for Biomimetic Sensing and Control ResearchResearch, 2Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China IIM, Hefei, China • An Accurate Localization in Combination with Wireless Sensor • Related work of amphibious robot Networks and Laser localization scheme • Biological inspiration and updated for Lunar Rover is proposed. schemes • The lunar rover position system consisted of the Lander, the lunar rover and a • Improved buoyancy designand wireless sensor network is introduced. wheel-propeller-fin mechanism • In the proposed scheme, the wireless • Mechanical implementation and sensor networks firstly localize testing results Concept design of themselves, and then help lunar rover to get position in the special conditions. the AmphiRobot-II • In initial experiment, we validate the Experiment Field to test proposed method using the test platform the Proposed Algorithm designed by us.

HQHT-304(3) 14:40-15:00 HQHT-304(4) 15:00-15:20 Minimum Fuel Guidance Law for the Parameter Design and Optimization for Powered Descending Phase of a Lunar Mobile Mechanism of a Transformable Module Wheel-Track Robot Feng Zhang, Guangren Duan, and Yongzheng Shan Centre for Control Theory and Guidance Technology, Harbin Institute of Technoloty Zhiqing Li, Shugen Ma, Bin Li, Minghui Wang, Yuechao Wang Harbin, China State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of • Based on 3-dimentional kinetic model, Sciences, Shenyang, China. • Propose a TWTR robot to adapt autonomously the fuel optimal problem is considered. to the irregular environment. • With a time scaling transform and • Each wheel-track unit, equipped with one constraint transform, the problem is actuator, has the following features: translated into a standard constrained ¾ transforming the track configuration ¾ changing the locomotion mode between track optimal control problem. mode and wheel mode. • An open-loop parametric optimal • Design and optimize the parameters for the guidance law is derived. TWTR mechanism based on genetic algorithm. • The validity of the parameters is demonstrated • With the parametric eigenstructure by simulation on transforming the track assignment method, a state feedback configuration and changing the locomotion The TWTR Robot control law is obtained. Lunar Exploration mode.

HQHT-304(5) 15:20-15:40 Comparative Analysis for the Inverse Kinematics of Redundant Manipulators based on Repetitive Tracking Tasks Jingguo Wang and Yangmin Li HQHT-312 HQHT-313 HQHT-314 Department of Electromechanical Engineering, University of Macau Macao, China • The closed-loop inverse kinematics algorithm via pseudoinverse method is used to solve the joint trajectories.

• The repetitive trajectories are HQHT-308 tracked in the simulations made on the 3-DoF planar and 4-DoF spatial manipulators respectively. • Two optimization methods with joint limits avoidance are considered. The block scheme of improved CLIK algorithm with performance criterion • The simulation results provide HQHT-304 HQHT-305 HQHT-306 detailed comparative analysis among different methods and cases. Han-Qing Conference Hall - Level 3 (东北大学汉卿会堂)

O-6 WP-2:Modeling II

Session Chair: Mingzhu Zhang and Dayou Li

HQHT-308 14:00-15:40 Wednesday, August 5th

HQHT-308(1) 14:00-14:20 HQHT-308(2) 14:20-14:40 Efficiency Analysis of an Innovative The Analysis of the Energy Efficiency of Multi-range Hydro-mechanical Refrigeration Stations in HVAC Xiaotong Du Xiaomei Qi Maoliang Ai Continuously Variable Transmission School of Control Science and Engineering , Shandong University Jinan, China Zhang Mingzhu, Zhou Zhili, Xu Liyou Henan University of Science and Technology, Luoyang, China • In buildings refrigeration stations account for the most of • Study the calculating method of the total energy consumption. efficiency of the innovative multi-range • Bring forward the concept of hydro-mechanical continuously variable energy efficiency of transmission (HMCVT). refrigeration stations, define • Factors such as the power cycle and the the expression of energy variation of the efficiency of hydraulic efficiency and energy units with their input power are consumption of refrigeration considered. stations • Give an efficiency algorithm of the • Give out relation between The largest COP changing closed pump-motor system. Hydro-mechanical energy efficiency in different tendency of frozen station • Describe a reasonable efficiency Continuously Variable cooling water temperature and calculating procedure and calculate the Transmission variation of load efficiency of the multi-range HMCVT.

HQHT-308(3) 14:40-15:00 HQHT-308(4) 15:00-15:20 Design and Dynamic Simulation of Permanent Multi-dimensional Fuzzy Interpolation Neural Network Magnet Motor Actuator on High Voltage Circuit Dayou Li1, Hua Qiu2, Yong Yue1, Casten Maple1, Vitaly Schetinin1 Breaker 1Department of Computer Science and Technology, University of Bedfordshire, UK 2Department of Mechanical Engineering, Kyushu Sangyo University, Japan Lin Xin, Li Yong-xiang, Yang Chuan, Xu Jian-yuan, Liu Ai-min College of Electrical Engineering, Shenyang University of Technology • Fuzzy interpolation is the piecewise Shenyang, China linearisation of a given function B using a collection of fuzzy sets. C • The motor drive actuator which used A • K-dimensional fuzzy interpolation X in high voltage circuit breaker is 1 approximates an MISO function showed in this paper. with a collection of small hyper- • The characteristics requirement of surfaces. Rotary shift Stator Rotor Brake encoder (ei(x0), ej(x1)) VCB is analyzed and calculated. • An accurate approximation depends n SN on how fuzzy sets are defined. (ei’(x0), ej’(x1)) • The limited angle permanent magnet S N X DC motor is designed. • Fuzzy interpolation neural network 0 • The transient finite element method Structure of VCB fixed rotary allows the fuzzy set optimally Fuzzy sets modification is employed in the permanent magnet motor actuator and permanent defined. DC motor actuator. magnet brushless DC motor

HQHT-308(5) 15:20-15:40 Human-Adaptive Step Estimation Method

for Dance Partnet Robot HQHT-312 HQHT-313 HQHT-314 Daishi Nakayama, Kazuhiro Kosuge, and Yasuhisa Hirata Department of Bioengineering and Robotics in Tohoku University , Japan • A dance partner robot has been proposed as a platform for human-robot coordination. HQHT-308 • Human-adaptive step estimation method for dance partner robot is proposed. • Experimental results will show the effectiveness of the HQHT-304 HQHT-305 HQHT-306 proposed method. Partner Ballroom Dance Robot Han-Qing Conference Hall - Level 3 (东北大学汉卿会堂)

O-7 WP-3:Automatic control II

Session Chair: Donglian Qi and Shi Jin

HQHT-312 14:00-15:40 Wednesday, August 5th

HQHT-312(1) 14:00-14:20 HQHT-312(2) 14:20-14:40 Same Cycle Servo Control System Motion Control System Simulation of Cylindrical Based on Built-in NC Linear Induction Motor Used in Circuit Breaker Shuanghui Hao, Jizhu Liu ,Weifeng Zheng,Minghui Hao and Ruizheng Long Operating Mechanism School of Mechanical Engineering ,Harbin Institute of Technology Harbin,China Liu Ai-min, Li Jia-jue, Li Yong-xiang, Lin Xin School of Electrical Engineering，Shenyang University of Technology，China • The electrocircuit and control software of AC servo system are • Introduction of C-LIM. designed in this paper. • Combined with the arithmetic of vector control, the NC • Building the system load model . instruction is embedded in the control model. • Planning the model of curve based • The NC instruction, machine loop and current loop are on motor movement. successfully achieved in the same control circle, and the time • The modeling and simulation for of control circle is determined in the 60us. vector control. • The response frequency and the performance of motion • Conclusion. control are improved greatly. the simulation curves of C-LIM primary

HQHT-312(3) 14:40-15:00 HQHT-312(4) 15:00-15:20

Simulation and Design of Control System on H∞ Robust Control for VSCF Brushless High Voltage Circuit Breaker Permanent Magnet Doubly-Fed Wind Power Generator System Shi Jin, Fengge Zhang and Yongxin Li Linear Motor Drive College of Electrical Engineering, Shenyang University of Technology Wang Xiaoyu, Lin Xin, Xu Jianyuan,Li yongxiang Shenyang, China School of Electrical Engineering，Shenyang University of Technology，Shenyang，China • The operation characteristics of wind turbine is analyzed. • Design of Linear Servo Motor •AH∞ speed controller without Drive of HVCB orthogonality assumptions is • The Hardware system of servo designed for variable speed constant motor actuator frequency brushless doubly-fed wind power generation system. • Control strategy of servo system • The proposed controller enhances • Simulation of the control system system robustness to the parameters • Discussion H speed controller Structure graph of linear motor drive for Vacuum Circuit Breaker uncertainties and rapid tracking ∞ performance to the input signal to design block diagram implement the maximum wind energy tracking.

HQHT-312(5) 15:20-15:40 The Title of My Paper on Aibo Behavior

Jie Yang, Donglian Qi, Guangzhou Zhao HQHT-312 HQHT-313 HQHT-314 College of Electrical Engineering, Zhejiang University Hangzhou, China

• A synchronization control approach for unified chaotic system with uncertainties and time delay is presented using passive control theory. • Because the chaotic trajectory is limitary HQHT-308 and the nonlinear function of the system satisfies Lipschitz condition, the error system transformed by synchronization error system can be globally asymptotically stabilized by state feedback, therefore it implements the synchronization control of unified chaotic system with uncertainties HQHT-304 HQHT-305 HQHT-306 and time delay. The Error output Han-Qing Conference Hall - Level 3 • Simulation results show that the adaptive 东北大学汉卿会堂 passive synchronization scheme is efficient. ( )

O-8 WP-4:Intelligent control and automation II

Session Chair: Yuqing Chen and Fujun He

HQHT-313 14:00-15:40 Wednesday, August 5th

HQHT-313(1) 14:00-14:20 HQHT-313(2) 14:20-14:40 Key Technologies of 3D Surface Inspection for Study of PID Neural Network for Hydraulic Complex Workpiece Using OMP60 probe System Yuqing Chen, Zi Ma and Huipu Xu Automation Research Center, Dalian Maritime University Beitao Guo , Hongyi Liu, Zhong Luo and Fei Wang Dalian, China School of Mechanical Engineering and Automation,Northeastern University Shenyang, Liaoning Province, China • Probe sphere radius compensation algorithm • System model PID Controller • Hand-Eye calibration process is Based On BP Network . presented based on armed robot • PID Control and simulation. inspection system • Point-sets registration • Adaptive PID Controller • Point distribution and path planning Based on BP Neural Network. are developed and tested on realistic • Simulation Research and scenarios analysis Armed robot inspection system The Hydraulic system

HQHT-313(3)A New Fiber 14:40-15:00 Winding Precision HQHT-313(4) 15:00-15:20 Tension Control System Humanoid Detection of Indoor Dangerous Gas Source by Mobile Robot Ren Sheng-le , Lai Yi-nan , Wang Yong-zhang , and Lu Hua Fujun He, Zhijiang Du,Yingfei Sun, and Xiaolei Liu School of Mechanical & Power Engineering , Harbin University of Science and Technology State Key Laboratory of Robotics and System, Harbin Institute of Technology Harbin, China Harbin, China • The robot can detect the indoor • Introduction dangerous gas and give alarm. • Set-up of the system scheme • two kinds of searching strategy is presented. • Mathematical model of the system • The responding and recover delay • Software development of control features of gas sensors is analyzed system and indoor searching features is • Results and analysis of the discussed. experiments • The strategy imitates person’s The Fiber Tension Control Equipment searching process, which can evaluates risk with a fuzzy inferring system. The detecting robot

HQHT-313(5) 15:20-15:40 Automatic Performance Degradation Prediction

by Use of Field Data with Noise HQHT-312 HQHT-313 HQHT-314 Tao Zhang, Jian Wang and Peng Guo Department of Automation, Tsinghua University, Beijing 100084, China

Raw Data • Proposes an automatic performance 485 480

degradation prediction approach for 475

equipment by use of field data with Temperat ure 470

465 noise. 0 200 400 600 800 1000 1200 1400 1600 HQHT-308 Time • Auto-regression model is employed Raw Data of Gas Engine for making performance degradation prediction by means of field data. • Improved least square method is adopted for parameter

identification of auto-regression model. HQHT-304 HQHT-305 HQHT-306

• Field data of the temperature of gas engine are adopted to analyze Han-Qing Conference Hall - Level 3 the performance degradation of gas engine. (东北大学汉卿会堂)

O-9 WP-5:Logistics planning

Session Chair: Na Xu and Chao Hu

HQHT-314 14:00-15:40 Wednesday, August 5th

HQHT-314(1) 14:00-14:20 HQHT-314(2) 14:20-14:40 SOA - BPM Based Information System For Design on Highway Accelerated Promoting Agility Of Third Party Logistics Loading Testing Facility Wen Zhen-Hua, Huang You-Sen, Deng Zhi-Yun and Zhang Wei Hunan Technology College of Modern Logistics, Changsha, China Zhiguang Guan1,2,Mingxing Lin1, Xuguang Wang2,Jiwei Zhang2 1.School of Mechanical Engineering, Key Laboratory of High Efficiency and Clean Mechanical Manufacture, Ministry of Education, Shandong University Jinan, China Business • SOA-BPM based IT Workflow Process 2.Shandong Jiaotong University, Jinan, China modeling Create infrastructure was suggested for • The facility is 26.24 meters long, Create XML file promoting agility of 3PL Create 7.92m high and 3.48m wide. Workflow Workflow display information system Descriptor • Simulating vehicle’s real running • Tuscany and Osworkflow were conditions. used for implementation of a • Test wheel: 4,driver wheel: 2, driven Workflow Workflow prototype system Engine monitoring wheel: 2. Accelerated Loading Testing • Further research will carry out for • Effect test length: 6m. improvement and application of • Maximum load on the test wheel: the prototype system. Workflow SOA Middleware Scheduling 150kN.

HQHT-314(3) 14:40-15:00 HQHT-314(4) 15:00-15:20 Profit Allocation in Collaborative Less- Cyclic Transfer Algorithm for Single Batching Than-Truckload Carrier Alliance Machine with Total Weighted Completion Time Na Xu , Chengxin Yu,Lei Zhang and Peng Liu Daguang Feng, Lixin Tang Business School,Shandong Jianzhu University Liaoning Key Laboratory of Manufacturing System and Logistics, The Logistics Institute， Jinan, China Northeastern University, Shenyang, China • The international Financial Crisis has made the Less-than-truckload industry face with severe challenges of survival • Introduction. and development. • Based on the cooperative game theory, • Dijkstra algorithm for formulate the LTL collaboration game partitioning batches. and discuss the well-known profit allocations including proportional • Cyclic transfer algorithm allocation, Shapley value and Nucleolus. for the partitioned batches. • Propose a new allocation method Weighted Relative Savings Model • Empirical results. (WRSM) which is in the core and minimizes the maximum difference between the contribution ratio weighted Collaborative Transportation System relative savings among the participants.

HQHT-314(5) 15:20-15:40 A particle swarm optimization for the quay crane

scheduling problem with non-interference HQHT-312 HQHT-313 HQHT-314 constraints Jiao Zhao, Lixin Tang Liaoning Key Laboratory of Manufacturing System and Logistics, The Logistics Institute, Northeastern University Shenyang, China HQHT-308 z The quay crane scheduling problem discussed in this paper is to schedule a fixed number of cranes in order to load and unload containers into and from a vessel considering interference between cranes. HQHT-304 HQHT-305 HQHT-306 z A particle swarm optimization (PSO) algorithm to obtain solutions. Han-Qing Conference Hall - Level 3 The quay crane scheduling (东北大学汉卿会堂)

O-10 Thursday August 6, 2009

TM-1 Robotics III

TM-2 Modeling III

TM-3 Distribution systems

TM-4 Process automation

TM-5 Modeling and application

TM-1:Robotics III

Session Chair: Xuezhang Hou and Aiguo Ming

HQHT-304 10:20-12:00 Thursday, August 6th

HQHT-304(1) 10:20-10:40 HQHT-304(2) 10:40-11:00 Counterweight-Navigation of a Mobile Inspection An Optimal Energy Control for a Flexible Robot Working on the Ground Wires Robot Arm Jian Jin, Huajin Zhu, Guoxian Zhang Xuezhang Hou 12, Zuoquan Wang Department of Mechanical Automation, Shanghai University 1 Mathematics Department, Towson University, Baltimore, Maryland 21252-0001, USA Shanghai, China 2 School of Science, Changchun University of Science and Technology, Changchun, China In this paper, we are concerned with a flexible robot arm • An inspection robot for power formulated by partial differential equations with the initial transmission lines. and boundary conditions. An optimal energy control of the • Working on the ground wires. robot system is investigated after the system has been • Performing the obstacle navigation transformed to an abstract evolution system in an appropriate automatically. Hilbert space. The existence and uniqueness of the optimal • Governing the centroid balance by energy control are proposed, discussed and proved, and an a simple method. approximation result is obtained in terms of semi-group methods of linear operators. The Inspection Robot

HQHT-304(3) 11:00-11:20 HQHT-304(4) 11:20-11:40 Trajectory Planning for a Redundant Mobile Manipulator using Avoidance Service Robots for Hospitals: Manipulability Ze CUI Peng CAO Yan-Ming SHAO Dong-Hai Qian Xiang-Can WANG A Case Study of Transportation Tasks Department of Precision Mechanical Engineering School of Mechatronics Engineering and Automation, Shanghai University Yanchang Road, Shanghai, China, 200072 in a Hospital [email protected] Ali Gürcan Özkil, Zhun Fan, Steen Dawids, Henrik Aanæs, Jens Klæstrup Kristensen and Kim A Trajectory planning method for a mobile manipulator is proposed in this paper. The mobile Hardam Christensen manipulator studied here consists of a mobile platform and a redundant manipulator. Since the system has the redundant degree of freedom to perform some tasks, it will have infinite configurations to In this paper, the need for automated make the end-effector track the desired trajectory. The concept of AMSI and AMSIP which were used transportation systems for hospitals is to optimize the avoidance ability for redundant manipulator is extended to carry out the trajectory investigated. Among other alternatives, planning in this paper. We firstly predefine the trajectory of the mobile platform and stop it at a mobile robots stand out as the most certain position defined by AMSI where allows the manipulator to reach all the desired points on the prominent means of automation of trajectory, then optimize the configuration of the manipulator according to AMSI and AMSIP to transportation tasks in hospitals. realize the whole trajectory planning task. The experiments are performed by the mobile manipulator in real environment. Existing transportation routines of a hospital are analyzed in order to verify the need for automation and identify possible areas of improvement. The analysis shows that most of the existing transportation is carried out manually, and hospitals can greatly benefit from automated transportation. Based on the results of the analysis, three alternatives are derived for implementing mobile service robots for transportation tasks in hospitals. Screen shots when perform the rotational motion planning Screen shots when perform the transitional motion planning

HQHT-304(5) 11:40-12:00 Home Service by a Mobile Manipulator System

-Mobile Manipulation of Chairs- HQHT-312 HQHT-313 HQHT-314 Manabu Yamashiro, Zhaoxian Xie, Hisashi Yamaguchi, Aiguo Ming and Makoto Shimojo The University of Electro-Communications Tokyo, JAPAN • As a practical and versatile platform for home service, a

mobile manipulator system has HQHT-308 been developed. • A RFID-based sensor fusion system is constructed to realize efficient and detailed recognition. • Mobile manipulations of various chairs based on the detailed HQHT-304 HQHT-305 HQHT-306 information by the sensor fusion Mobile manipulation of a chair Han-Qing Conference Hall - Level 3 system have been performed (东北大学汉卿会堂) successfully.

O-11 TM-2:Modeling III

Session Chair: Sam Ge and Simon Yang

HQHT-308 10:20-12:00 Thursday, August 6th

HQHT-308(1) 10:20-10:40 HQHT-308(2) 10:40-11:00 Dynamic Analysis and Safety Monitoring of Tank Truck An Improved Semantic Entropy Zhao Jianwen, Bo Huang, Zhu Feibai, Lu Shizhou Maximum Model and Its Applications Robotics Institute, Harbin Institute of Technology at Weihai Munan LI Weihai, China School of Business Administration, South China University of Technology Guangzhou, China • Tank trucks transporting dangerous gas • Brings forward a novel semantic discovery approach based on the often have serious traffic accidents SEM (Semantic Entropy Maximum). • build full-vehicle model and found the • This approach applies the entropy maximum algorithm to the relationship between tank truck dynamic matching and the recognizing of the semantic service, and parameters and its working state meanwhile, it compares this algorithm with other related • simulations validate model’s handling algorithms. directional stability. • The results of those experiments show that the SEM (semantic • road test validate the tank truck model entropy maximum) model has some preferable performances. • Simulations under different conditions in • In a practical enterprise application of Chinese semantic order to study the performance of Tank Truck Prototype recognition, the SEM model is proved to be useful and valuable dynamic parameters. in semantic recognition of Chinese.

HQHT-308(3) 11:00-11:20 HQHT-308(4) 11:20-11:40 Design and Finite Element Analysis on a Novel PMSM with Anti-rotation Dual Rotor Applying Multifractal Spectrum Combined Jinhua Chen, Fengge Zhang, Guangwei Liu and Zhaohe Meng with Fractal Discrete Brownian Motion College of Electrical Engineering, Shenyang University of Technology Shenyang, China Model to Wood Defects Recognition for • the principles of operation, configurations, Sawing and features are discussed. • Both the equivalent magnetic circuit Windings method and Finite Element Analysis are PMs Lei Yu and Dawei Qi adopted to analyse the electromagnetic coupling of the machine. Northeast Forestry University • Design methodology is given. • the combination of field with circuit Structure Scheme of should be done to design the machine. Dual Rotor PM Machine

HQHT-308(5) 11:40-12:00 Storage Planning for Outbound Container

on Maritime Container Terminals HQHT-312 HQHT-313 HQHT-314 Yanwei Zhang School of Logistics Engineering, Wuhan University of Technology, Wuhan , China

• Storage locations pre-plan for outbound containers is important and complex.

• Practical rules for storage planning for HQHT-308 outbound containers are collected from the actual container terminals • Rules are treated and described in the three levels mathematic model with rolling-horizon planning idea.

• Simulation scenarios are designed for HQHT-304 HQHT-305 HQHT-306 validating the rules and principles The Yard described in mathematic models. Han-Qing Conference Hall - Level 3 in a Container Terminal (东北大学汉卿会堂)

O-12 TM-3:Distribution systems

Session Chair: Dengpan Zhang and Xiaotong Du

HQHT-312 10:20-12:00 Thursday, August 6th

HQHT-312(1) 10:20-10:40 HQHT-312(2) 10:40-11:00 Determination of Effective Energy in A New Approach for Task Scheduling in Distributed Systems Using Learning Automata Buildings M. Jahanshahi Xiaotong Du Xiaomei Qi Cundong Wang Department of Computer Engineering Islamic Azad University, Central Tehran Branch School of Control Science and Engineering , Shandong University Jinan, China Tehran, Iran

• The phenomenon of low M. R. Meybodi and M. Dehghan 90 efficiency and waste is 80 Department of Computer Engineering and Information Technology 70 prevalent in public buildings. 60 Amirkabir University of Technology, Tehran, Iran 50 temperature • This paper presents a concept 40 dT(t) 30 sensation of energy efficiency. One part 20 • Tasks scheduling problem is a key factor for a distributed system in order to achieve better 10 efficiency. The problem of tasks scheduling in a distributed system can be stated as if effective energy, the other 0 -10 allocating tasks to processor of each computer. The objective of this problem is minimizing one is ineffective energy. Makespan and communication cost while maximizing CPU utilization. Scheduling problem is known as NP-complete. Hence, many genetic algorithms have been proposed to search • The ratio of both times can optimal solutions from entire solution space. However, the existing approaches are going to represent energy validity scan the entire solution space without consideration to techniques that can reduce the 2007-7-19 pm 01:29:44 2007-7-19 pm 01:49:53 2007-7-19 pm 02:10:02 2007-7-19 pm 02:30:11 2007-7-19 pm 02:50:20 2007-7-19 pm 03:10:42 2007-7-19 pm 03:30:51 2007-7-19 pm 03:51:00 2007-7-19 pm 04:11:09 2007-7-19 pm 04:31:17 2007-7-19 pm 04:51:27 2007-7-19 pm 05:11:35 2007-7-19 pm 05:31:44 2007-7-19 pm 05:51:53 2007-7-19 pm 06:12:02 complexity of the optimization. In other words, the main weakness of these methods is to indirectly. spend much time doing scheduling and hence need to exhaustive time. In this paper we use • It would be a quantitative temperatures, sensation Learning algorithm to cope with the weakness of GA based method. In fact we use the indicator to obtain reasonable curve and dT(t) Learning automata as local search in the memetic algorithm. energy consumption of building.

HQHT-312(3) 11:00-11:20 HQHT-312(4) 11:20-11:40 Wavefront Array Based Distributed Collaborative Design of Three-dimensional Guidance Law Measurement Pattern Zhang Dengpan, Cui Shuxin, and He Lingsong Based on Extended State Observer for Hit-to- School of Mechanical and Power Engineering, Henan Polytechnic University Jiaozuo, Henan Province, China Kill Interceptors Yuan Tian, Jian Chen and Zhang Ren • The Architecture of Wavefront School of Automation Science and Electrical Engineering, Beihang University Array Based Measurement. Beijing, China • The Virtual Bus of the • 3D high-speed engagement. Wavefront Array Pattern. • Approximate miss distance formula. • The wavefront array model of System Architecture • Extended State Observer. the data exchange way. • Feedback linearization. • The collaborative measurement pattern for the distributed • Decoupling 3D nonlinear kinematics into two one-order subsystems. environment. • Simulation and results. The Hit-to-Kill Interceptor Measurement Pattern

HQHT-312(5) 11:40-12:00 A Hybrid Algorithm of Tabu Search

and Benders Decomposition for Multi-product HQHT-312 HQHT-313 HQHT-314 Production Distribution Network Design Wei Jiang, Lixin Tang and Shixu Xue Liaoning Key Laboratory of Manufacturing System and Logistics,The Logistics Institute Northeastern University Shenyang, China • The paper presents a hybrid HQHT-308 algorithm of Tabu search and Benders decomposition to solve a multi-product two-stage production distribution network design problem.

• The computational experiments HQHT-304 HQHT-305 HQHT-306 show the hybrid algorithm saves tremendous time compared with Han-Qing Conference Hall - Level 3 (东北大学汉卿会堂) conventional benders decomposition. The flow chart

O-13 TM-4:Process automation

Session Chair: Chao Hu and Hong Wang

HQHT-313 10:20-12:00 Thursday, August 6th

HQHT-313(1) 10:20-10:40 HQHT-313(2) 10:40-11:00 A Method to Coordination for Decentralized A Pragmatic Approach for Selecting Weight Matrix Multi- MPC Controllers in Integrated Plant Coefficients in Model Predictive Control Algorithm and Its Aimin An 1, Xiaohong Hao1, Hongye Su2, Guangtao Yang3, Yongwei Ma1 Application 1Lanzhou University of Technology, Lanzhou, China 2Zhejiang University, Hangzhou, China Aimin An1 ,Xiaohong Hao1, Chao Zhao2 and Hongye Su2 3Lanzhou Petroleum Chemical Engineering Company , Lanzhou, China 1Institute of Electrical and information Engineering, Lanzhou Universityof Thechnologu 2Institute of Cyber-Systems and Control, Zhejiang University • A method to coordinate multi-decentralized • To Select appropriate coefficients often requires large model predictive controllers (DMMPC) is proposed. expensive computation duty in MPC. • It is a strategy to control the integrated plant through decomposing the total system model into different functional sub- • The controllability and the observability Gramians of a models, stable controlled system are needed to calculate. • The main aim is to relieve the expensive computation load and • The controllability and observability Gramians singular for improving the flexibility and reliability for operation. value matrices are calculated. • A stochastic probability density functions(PDFs) coordination • Gramians singular values of the system are used as a rule is applied to coordinate optimizer with decentralized MPC controllers in the hierarchical control structure. benchmark to select the weight coefficients corresponding to different variable in the objective function of MPC.

HQHT-313(3) 11:00-11:20 HQHT-313(4) 11:20-11:40 A FCM-Weighted Markov Model for Remaining Probabilistic Quality Control in non-Gaussian Life Prediction Process Control Applications Jihong Yan and Chaozhong Guo Puya Afshar*, Amin Nobakhti, Hong Wang* Department of Industrial Engineering, Harbin Institute of Technology Control Systems Centre, The University of Manchester, Manchester, UK Harbin, China • A multi-objective process/quality control • Basic Markov model method for non-Gaussian stochastic • Weighted Markov model batch processes (CSTR) is proposed • Process tracking error and tracking error • Fuzzy C-Means (FCM)- entropy as well as dwelling time error weighted Markov model for Bently Rk4 unbalance test-bed and its entropy are minimised

35 remaining life prediction Precdiction life A c tu a l life 30 • In stage1, Evolutionary Algorithms are 25 • Application of FCM-weighted 20 used to find the best batch parameters 15

Markov model for rotor life Remaining life 10 • In stage2, PI control input is calculated 5 and applied to plant 0 prediction to a Bently RK4 65 70 75 80 85 90 Prediction cycle Remaining life prediction result • In Stage3, product is accepted/rejected unbalance test-bed of the rotor based on quality monitoring results

HQHT-313(5) 11:40-12:00 Application of Robust Right Coprime Factorization Approach to A Distributed Process HQHT-312 HQHT-313 HQHT-314 Control System

Shengjun. Wen, Mingcong. Deng, Akira. Inoue Graduate School of Natural Science and Technology, Okayama University Okayama, Japan • A distributed process experimental

system model is shown which includes HQHT-308 a water level process model and a water flow process model. • Robust right coprime factorization based nonlinear control scheme of the distributed process is proposed. • From the industrial application viewpoint, implementation issues of HQHT-304 HQHT-305 HQHT-306 designed controllers are investigated. • Experimental results show the Han-Qing Conference Hall - Level 3 effectiveness of the proposed approach. Experimental System (东北大学汉卿会堂)

O-14 TM-5:Modeling and application

Session Chair: Ming Zhong and Xiaohua Yu

HQHT-314 10:20-12:00 Thursday, August 6th

HQHT-314(1) 10:20-10:40 HQHT-314(2) 10:40-11:00 Ship Oscillation Simulation System A Dynamic Vessel Fleet Planning Model in Based on Direction Spectrum Uncertain Circumstances Bo Huang Jianwen Zhao Yanyu Su Shizhou Lu Zhong Ming and Min De quan, Liu Jian qiu Ship Engineering College, Harbin Institute of Technology Transportation Management College, Dalian Maritime University Weihai, China Liaoning, China • Oscillating motion of ship makes a • This paper aims to establish a dynamic vessel fleet planning model in uncertain circumstance, which offers the shipping companies the great impact on the safety of ocean appropriate development strategy. shipping of LNG tank • We establish a forecast model to predict the future shipping demand • construct several groups of simulation by virtue of Grey—Markov chain. environment and an algorithm based • We bring the shipping demand as an uncertainty into the fleet on direction spectrum to describe planning, further, a dynamic fleet planning model in uncertain ocean wave spectrum according to circumstances is established from the view of minimizing the total Seakeeping theory operating costs. • ship oscillation simulation system is • Finally, we use an example to verify the model and the algorithm built by VC++ and DirectX. presented. • the simulation results are close to the Ship oscillation Simulation ship’s experimental data.

HQHT-314(3) 11:00-11:20 HQHT-314(4) 11:20-11:40 Application of Model-following Adaptive Neural Electrocardiogram (ECG) Signal Modeling and Network Control Theory in Gust Load Alleviation Noise Reduction Using Wavelet Neural Network Rui Nie,Weiguo Zhang,Guangwen Li,Xiaoxiong Liu Suranai Poungponsri, Xiao-Hua Yu College of Automation Northwestern Polytechnical University Department of Electrical Engineering Xi’an, China California Polytechnic State University San Luis Obispo, CA 93407, USA

• Based on the model-following gust responce 0.4 with RCS adaptive neural network control 0.3 without RCS • Electrocardiogram (ECG) signal has been widely used in cardiac theory, a gust load alleviation 0.2 pathology to detect heart disease controller for civil airplanes by 0.1 0 • Wavelet neural networks (WNN) combine the multi-resolution Nz/g direct force control is proposed -0.1 nature of wavelets and the adaptive learning ability of artificial • Simulation results show that the -0.2 neural networks -0.3

-0.4 GLA control can obtain good robust 0 10 20 30 40 50 60 • In this paper, WNN is studied for ECG signal modeling and noise t/s stability and the capability of reduction. The WNN is trained by Adaptive Diversity Learning Aircraft’s overload response restraining gust turbulence and Particle Swarm Optimization (ADLPSO) and the gradient decent impacted by turbulence measurement noises can be obtained algorithm. Satisfactory Computer simulation results are obtained by using the method in this paper

HQHT-314(5) 11:40-12:00

Toward Bacteria Swarm for Environmental

Monitoring HQHT-312 HQHT-313 HQHT-314 John O Oyekan and Huosheng Hu University of Essex, United Kingdom

¾For optimal coverage of an environmental variable for

pollution monitoring. HQHT-308

¾Control of real and simulated agents using a combination of bacteria controller and flocking controller.

HQHT-304 HQHT-305 HQHT-306 ¾Architecture and system Han-Qing Conference Hall - Level 3 design for the System. (东北大学汉卿会堂)

O-15

Friday August 7, 2009

FM-1 Optimization I

FM-2 Simulation I

FM-3 Image Processing

FM-4 Network-based systems I

FM-5 Planning and scheduling

FP-1 Optimization II

FP-2 Simulation II

FP-3 Intelligent transportation system

FP-4 Network-based systems II

FM-1:Optimization I

Session Chair: Xiaohua Yu and Dayou Li

HQHT-304 10:20-12:00 Friday, August 7th

HQHT-304(1) 10:20-10:40 HQHT-304(2) 10:40-11:00 F-separation law inference and law mining Design and Realization of the Huang Shunliang, Gao Houli Economical Operation Monitoring System for School of Management, Shandong University of Technology Zibo, China Traction Transformers • Present concepts of f- Huige Lai1, Xuejun Zhu1, and Jinshou Yu 2 1 − School of Mechanical Engineering , Ningxia University; Yinchuan, China separation law, F - separation p()x s 2 law, f-separation law Research Institute of Automation, East China University of Science andTechnology; Shanghai , China dependence and F- separation • Security and economical operation of TT is a significant guarantee for electric railway law dependence. with growing apace. • Propose some theorems of f- • TT economical load factor must be judged separation law, F - separation 12 3 45 6 by the critical load factor; The critical load law. synthesis losses are key index; The power economized TT is selected optimally on the • Propose criteria of law F - separation law new or upgrade electric railway.

separation dependence- 0 ZA − PP 0 ZC S LZ 2 = P KZC P KZA discovery. 2 − 2 S NC S NA • Give an example of law mining. • TT monitoring system can be embedded in a digital protection system of power sub- Running Window of Monitoring System station.

HQHT-304(3) 11:00-11:20 HQHT-304(4) 11:20-11:40 Vehicle Scheduling Problem of Hazardous Application of Reinforcement Learning in Dynamic Pricing Algorithms Materials Based on Risk Wang Jintian Zhou Lei Department of Computer Science and Technology, Hefei University of Technology Jianghua Zhang1, Qiang Han 2, Jin Li 3, and Daoli Zhu4 Hefei, China School of Management, Shandong University Jinan, China • dynamic pricing problems • This paper studies the vehicle of a duopoly case in scheduling problem of Hazardous Risk Comparison of Typical Paths under the Different Scenarios electronic retail markets 80 materials based on risks which vary 70 • simulated annealing Q- with time, taking into account the cost 60 learning (SA-Q) and win-or- waiting is not allowed constraints and allowing for waiting at 50 Risk learn-fast policy hill the nodes of the network. waiting is allowed 40 climbing algorithm (WoLF- • A mixed integer programming model is 30 PHC) proposed, a novel heuristic algorithm is 20 12345678 designed. The computational Number complexity of the algorithm is also analyzed. a numerical example is Risk Comparison Two sellers monopolizing presented to show its validity and of Typical Paths electronic retail markets model feasibility of this algorithm.

HQHT-304(5) 11:40-12:00 Genetic and Simulated Annealing Algorithm based on Chaos Variables HQHT-312 HQHT-313 HQHT-314

Jing Jiang, Boxue Tan and Lidong Meng School of Electrical and Electronic Engineering , Shandong University of Technology Zibo, China

• Procedures of genetic algorithm HQHT-308 • Combination of genetic algorithm with simulated annealing • Using chaos vaviables to search neighborhood

HQHT-304 HQHT-305 HQHT-306

Han-Qing Conference Hall - Level 3 (东北大学汉卿会堂)

O-16 FM-2:Simulation I

Session Chair: Ming Zhong and Zaiwen Liu

HQHT-308 10:20-12:00 Friday, August 7th

HQHT-308(1) 10:20-10:40 HQHT-308(2) 10:40-11:00 A Model of Fleet Planning for Land-Island CODP Position of leagile supply chain based on polychromatic sets theory Transport Dongmei Liu, Wen Wang, and Weiping Fu Zhong Ming, Shen Jiang, Li Li School of Mechanical and Precision Instrument Engineering Transportation Management College, Dalian Maritime University Xi’an University of Technology, Xi’an, 710048, P. R. China Liaoning, China • According to the resource capacity of supply • The paper’s purpose is to establish a model of fleet planning of chain members, a kind of decision-making the resource The supply model represented by CODP on supply chain capacity of supply chain model land-island transport. are established. chain members • By using the linear programming method, the model’s object • Using the polychromatic sets to the positions of customer order decoupling point on leagile Decision system The Polychromatic Graphs function is to minimize the annual operating cost of the fleet and the model of of the relationships supply chain. CODP in supply chain between supply chain members constraints are defined on the basis of the characteristics of land- • a case studies including two manufacturers island transport. enterprise on supply chain network is used to illustrate the proposed model and approach. The contour portfolio matrix • Take the land-island routes between Pikou port and the islands of • Though information of different levels in the of the decision system model Changhai in Dalian PRC as an example to test the model and obtain model corresponding with the uniform color and the individual color in the polychromatic the result of the following five years. The contour portfolio matrix sets, the reasoning relational matrix of the of the decoupling point influencing factors about members and on the supply chain • The model is fit for solving the problem of the fleet planning of separation point positioning indicators in the land-island transport. supply chains. Then the reasoning on the positions of decoupling points has been The CODP positioning Model achieved.

HQHT-308(3) 11:00-11:20 HQHT-308(4) 11:20-11:40 Research of Virtual Flight Testing System Prediction Technique for Water-Bloom in Lakes Li Liu, Jinglei Li and Rongzhen Jia Based on Elman Network School of Automation Science and Electrical Engineering, Beihang University Beijing, China Zaiwen Liu, Xiaoyi Wang, Jiping Xu, Lifeng Cui, Xiaofeng Lian College of Computer and Information Engineering, • VFT can replace part of the flight test Beijing Technology and Business University, China to validate the real aircraft flying • Structure and feature of qualities and performance. --Actual Value • The VFT system includes flight Elman network . ●Predict Value management & control console, data • Main factor analysis and acquisition, processing and evaluation modeling method for console, digital aircraft, database. occurrence of water-bloom • The VFT system is based on aircraft in rivers and lakes virtual prototyping or flight simulators. • Test of water-bloom short- • By testing on takeoff performance, term prediction model research indicates that VFT is an Composition of VFT System • Conclusion . efficient method of flight testing.

HQHT-308(5) 11:40-12:00 Numerical Simulation of Induction Heating of

Steel Bar with Nonlinear Material Properties HQHT-312 HQHT-313 HQHT-314 Sun Yafei1, Sun Jing2 and Niu Dongjie3 1 Guangdong Institute of Science and Technology, Zhuhai, Guangdong, China 2 Northeastern University, Shenyang, Liaoning, China 3 Tongji University, Shanghai, China • The simulation of PC steel bar heating process with different carbon content is presented. HQHT-308 • The surface temperature of PC steel bar in heating below Curie temperature is depended on the carbon content.

• The tempering temperature of the PC HQHT-304 HQHT-305 HQHT-306 steel bar in quenched-and-tempered Evolution of Surface Temperature Han-Qing Conference Hall - Level 3 process should be adjusted according (东北大学汉卿会堂) to the carbon content. and Inner Temperature

O-17 FM-3:Image Processing

Session Chair: Yangmin Li and Fei Lu

HQHT-312 10:20-12:00 Friday, August 7th

HQHT-312(1) 10:20-10:40 HQHT-312(2) 10:40-11:00 A Multi-stage Fingerprints Matching Algorithm An Emotion based 3D Spatial Design Honglei Wei , Danni Liu School of Mechanical Engineering, Dalian Fisheries University Considering Effective Factors and Evaluation Dalian, China Yong-Tae Won, Nam-Yee Kim and Hoon-Sung Kwak This paper divides fingerprint Research Laboratory for Image Processing, National University of chonbuk matching process into three Jeonju-City, Korea stages: • Emotional spatial design ¾ Local orientation structures matching. – Coloration modeling ¾ Local minutia structure – Lighting matching – Illumination ¾ Global minutia structure – 3D Real-time rendering dictionary matching visualization The final similarity was gotten – Evaluation through database according to the matching server build-up scores calculated from above Emotion based 3D Spatial Design three stages. • Emotional spatial design coloration local orientation structure • Evaluation of emotion

HQHT-312(3) 11:00-11:20 HQHT-312(4) 11:20-11:40 Particle Swarm Mixel Decomposition for Remote Sensing Images A multi-camera system for precise pose Dong Wang1,2+, Xiangbin Wu2, Dongmei Lin3 estimation in industrial applications 1Department of Computer Science and Technology, Foshan University, China 2College of Geosciences and Environmental Engineering, Central South University, China Fangwu Shu, Jianwei Zhang, Youfu Li 3Center of Information and Education Technology, Foshan University, China

Department of Informatics, University of Hamburg, Germany Algorithm – Framework of particle swarm thematic ground object removing algorithm • Directly utilizing Input – Characteristic image Y=[X1,X2,…,XN] Output – New image • A multi-camera system is used to determine [Begin] intelligence particle Initialization the poses of the large work objects. Iteration=0; /* Iterative times */ without pre-selecting Setting up other relative parameters; Reading remote sensing image; • A practical approach to camera calibration Putting stochastically particles on image; training samples. while (Iteration< Maximum iterative time) { for( i=0; i<1 000; i++ ) { /* The maximum iteration times is 1000 */ is proposed for robot vision applications. If( Satisfying to congeneric classification rules ) { • Needn't know or assume Adjusting gravitation of current pixel according to (3); Searching 8 neighboring pixels and reaching next pixel; • Pattern compensation, pattern weight, zero } else { the distribution of samples. Reaching stochastically to next pixel; } measurement and camera pose recalibration } • Utilizing adequately Iteration++; } are introduced to solve the practical issues. Labeling all of pixels according to the value of self gravitation; Normalizing all abundance value of mixels; neighboring information of Decomposing all mixels according to normalized result according to • The proposed approaches have been proved (3); Removing the ground object according to (4); mixels in image. Saving the new image; to be able to improve the performances of a Computing residual distribution; Saving results of abundance and residual; multi-camera system in practice. • Robustness to operating [End] environments. The Improved Algorithm

HQHT-312(5) 11:40-12:00 Eyes localization in an infrared image Tang Jin, Chen Shouming and Xie Xiuzhen, Jason Gu HQHT-312 HQHT-313 HQHT-314 School of Information science and Engineering, Central South University Changsha, China

HQHT-308

HQHT-304 HQHT-305 HQHT-306

The result of Eye localization Han-Qing Conference Hall - Level 3 (东北大学汉卿会堂)

O-18 FM-4:Network-based systems I

Session Chair: Yingwei Zhang and Chunyi Su

HQHT-313 10:20-12:00 Friday, August 7th

HQHT-313(1) 10:20-10:40 HQHT-313(2) 10:40-11:00 Approximate Dynamic Programming of Temporal Characteristic Constraints between Continuous Annealing process SP and PG of SAE J1939 in Passenger Car

Yingwei Zhang Chao Guo and Xue Chen Yongdong Teng Information Integrated Control System Key Laboratory of Integrated Automation of Process Industry, Ministry of Education , Jian HU*, Gangyan LI* and Yeqiong SONG** Northeastern University *School of Mechanical & Electronic Engineering, Wuhan University of Technology, China Et() Shenyang , China c ** LORIA - Equipe TRIO, Villers Les Nancy, France

LL LL spk ƒ The production line of Continuous • SAE J1939 and Its Application i t Annealing. Layer .

LL LL • Temporal Characteristics Problem pg k t ƒ Approximate Dynamic • • • • Offset ，TT )( Offset ，TT )( Offset ，TT )( Offset ，TT )( of PG of SAE J1939. ik ik max ik ik Programming. • Deadline Calculation Method of Time offset between the Ut() ƒ Design and Implementation of c PG of SAE J1939 . production time of the data k Controller Based on ADHDP • Temporal Characteristic of spi and the first The structure chart of the Constraints of SPs Deriving from production time of pg k ADHDP method the Temporal Characteristic of PG containing this data .

HQHT-313(3) 11:00-11:20 HQHT-313(4) 11:20-11:40 Component-Based ECU Design Method of UAVs Task Allocation Passenger Car Information Integrated Using Multiple Colonies of Ants Control System Wang Zhenhua, Zhang Weiguo, Li Guangwen College of Automation Jian HU, Gangyan LI and Jun XU Northwestern Polytechnical University Xi’an, China School of Mechanical & Electronic Engineering, Wuhan University of Technology Wuhan, Hubei, China • PRMs is used to construct routs • ECU Function Analysis and Classification of for UAVs, and a new sampling Passenger Car Information Integrated Control System. method for PRMs is introduced • Component-Based ECU Design Method of • A new method using Multiple Passenger Car Information Integrated Control System . colonies of ants is introduced to • Component-Based ECU Implementation Experimental installation allocate the targets to the UAVs Technology of Passenger Car Information of WG6100ENH model • Compared with Genetic Integrated Control System . natural gas city bus • Component-Based ECU Design of WG6100ENH algorithm, our method shows Model Natural Gas City Bus Information information integrated better performance Integrated Control System control system Allocation Results

HQHT-313(5) 11:40-12:00 A Design of a WSN System for Pollution

Monitoring and Positioning Based on GAF HQHT-312 HQHT-313 HQHT-314 Routing Algorithms LI Xiao-fang, XIA Wei, MAO Ying-chi, YU Xian ResearchCollege of Computer and Information Engineering , Hohai University

Nanjing, China Command center

• Proposed an architecture of a M obile monitoring station hydrologic telemetry and pollution tracking and positioning system based WAN HQHT-308 Public mobile on the public mobile communications network layer

network. BS BS • Designed a network protocols based on BS the GAF routing algorithm. • Emphasizes：the network’s lifecycle, V Test region pollution source positioning, limited HQHT-304 HQHT-305 HQHT-306 Sensor node energy, network quality and etc. data Acquisition when the GAF algorithm is used in layer Han-Qing Conference Hall - Level 3 practical application of the system. :Leader node of each virtual cluster (东北大学汉卿会堂) The System Architecture

O-19 FM-5:Planning and scheduling

Session Chair: Yingzi Wei and Ze Tao

HQHT-314 10:20-12:00 Friday, August 7th

HQHT-314(1) 10:20-10:40 HQHT-314(2) 10:40-11:00 Pattern Driven Dynamic Scheduling Approach A Backward Algorithm for Multi-Processor using Reinforcement Learning Scheduling Problem with Unequal Release Dates WEI Yingzi, JIANG Xinli, HAO Pingbo, GU Kanfeng Shenyang Ligong University, Shenyang, China Kai Li and Shu-Chu Zhang School of Management, Hefei University of Technology, • The paper presents an adaptive iterative scheduling algorithm Hefei, China that operates dynamically to schedule the job in the dynamic makespan=119 • The multi-processor scheduling problem with unequal job-shop. 10 16 14 7 2 9 13 18 5 4 6 1 10 3 8 12 11 17 15 release dates to minimize total completion times is 9 1 8 18 14 7 4 15 16 2 3 11 13 17 6 9 5 10 12 • The reinforcement learning 8 2 9 7 10 8 6 11 13 3 14 5 16 12 4 17 1 18 15 1 job considered. 7 2 job 3 job system is done with the phased 1 5 1315 18 2 10 12 16 3 8 9 17 14 6 4 7 11 6 4 job 5 job 12 16 14 3 17 9 8 11 2 10 5 7 1 13 15 6 18 4 • This problem is NP-hard in the strong sense. Q-learning by defining the 5 6 job 7 job Machine 14 15 4 12 7 11 1 3 10 9 17 8 2 6 16 5 18 13 8 job intermediate state pattern. 4 9 job 10 job 1 13 11 14 17 7 3 15 2 9 16 10 12 18 4 8 6 5 • The localization and blindness of the traditional 3 11 job 12 job 7 6 15 1 10 13 18 14 4 9 17 3 5 12 2 11 16 8 • We use five heuristic rules, 2 13 job forward heuristics are analyzed. 14 job 2 4 161810 3 17 5 6 9 14 12 11 7 8 13 1 15 15 job 1 CNP-CR, CNP-FCFS, CNP- 16 job 17 job 0 18 job • A backward algorithm BA is proposed. 0 20 40 60 80 100 120 140 EFT, CNP-EDD and CNP-SPT, t as actions and the scheduling objective: minimization of Scheduling Gantt chart maximum completion time.

HQHT-314(3) 11:00-11:20 HQHT-314(4) 11:20-11:40

Ant Colony Optimization Based Scheduling for a A Stable Scheduling for Single Machine under Semiconductor Wafer Fabrication Facility with Uncertainty Bottleneck Stations Qiaoyun Li and Bing Wang Li Li, Fei Qiao, Xiaoyu Tian and Qidi Wu School of Mechanical & Electrical Engineering, Shandong University at Weihai School of Electronics and Information Engineering,Tongji University Weihai, China Shanghai, China • Our Results/Contribution • Introduction. – Create a production environment applicable ACO based scheduling • The classification of uncertainties and its modelling approaches. method for a wafer fab with bottleneck stations. • Problem formulation. – The proposed method is superior to the common rule (FIFO, EDD, CR,SRPT) with better performance on the movements of the jobs and the • Modified Optimized Surrogate Measure Heuristic (MOSMH). machine utilities. • Computational results and its analysis. •Summary • Conclusion. – More higher workload of a wafer fab is, the improvements of the proposed method are better.

HQHT-314(5) 11:40-12:00 Dynamic Classified JSP Scheduling Based on

Petri Net and GASA HQHT-312 HQHT-313 HQHT-314 Ze Tao, Changzhong Hao Department of Mechanical Engineering, University of Shenyang Ligong Shenyang, China • Controller design when all transitions are controllable, and Petri net model

with uncontrollable transitions. HQHT-308 L • Flexible scheduling algorithm. r • Dynamic classified scheduling based De on machine repairing time, and worker D leaving time, and task of order c cancelling.

• Simulation based on some job shop HQHT-304 HQHT-305 HQHT-306 scheduling problems. Han-Qing Conference Hall - Level 3 controller design (东北大学汉卿会堂)

O-20 FP-1:Optimization II

Session Chair: Jing Bai and Shuli Sun

HQHT-304 14:00-15:20 Friday, August 7th

HQHT-304(1) 14:00-14:20 HQHT-304(2) 14:20-14:40

Application of Neighbourhood Topology Particle A New Start Method for Rolling Piston Swarm Optimization to Cylinder Linear Induction Compressor Based on DSP to Improve the Load Motor Design Performance Liu Ai-min, Zhang Xiao-ling , Lin Xin and Li Yong-xiang JilongQi,YantaoTian, YiminGong, Cheng Zhu and XuepingZhao School of Electrical Engineering, Shenyang university of Technology School of Communication Engineering , College of Physics, Jilin University Shenyang, China Changchun, China • This paper introduces the PSO and • PMSM traditional starting schemes improved PSO algorithms. are introduced. • Neighborhood Topology Particle • The new starting method is proposed. Swarm Optimization (NTPSO) is • Hardware and the startup process of introduced. compressor are described. • Based on NTPSO, Optimization • Simulation experiments and hardware design of C-LIM is executed. experiments results are provided. Four bouquet topology Rotor Speed And Position • The optimization results analyze. Neighbourhood model

HQHT-304(3) 14:40-15:00 HQHT-304(4) 15:00-15:20

Global Convergence of Variable Step Size LMS Optimal and Suboptimal Prior Filters with Adaptive Filtering Algorithm Based on Variable Bounded Multiple Packet Dropouts Region Shuli Sun, Lihua Xie, and Wendong Xiao BAI Jing1,2, YIN Yi-xin1, HAO Zhi-hong1 , XUE Jian2 Department of Automation, Heilongjiang University, Harbin, China, [email protected] 1) Information Engineering School, UST Beijing, China, 100083; School of EEE, Nanyang Technological University,Singapore, [email protected] 2) Information Engineering School, Beihua University, China, 132021. Media Processing Department, Institute for Infocomm Research, Singapore, [email protected] • A new variable step size LMS adaptive filtering algorithm is proposed star.edu.sg based on variable region. • Discrete-time stochastic linear system with bounded multiple • The step size μ(k) in the algorithm varies with variation of the packet dropouts is considered. deviation e(k) region to ensure the optimization of the three performance objectives, which are initial convergent speed, tracking • An optimal prior filter is developed in linear unbiased ability of the time-varying system and stable state convergence minimum variance sense. precision. • The optimal solution depends on the recursion of a Riccati • The convergence of the algorithm is respectively proved by equation and a Lyapunov equation. Lyapunov function and ODE with random noise and non-random noise. • To reduce the computational cost, a suboptimal prior filter is presented.

HQHT-312 HQHT-313 HQHT-314

HQHT-308

HQHT-304 HQHT-305 HQHT-306

Han-Qing Conference Hall - Level 3 (东北大学汉卿会堂)

O-21 FP-2:Simulation II

Session Chair: Yunong Zhang and Yanjun Shen

HQHT-308 14:00-15:20 Friday, August 7th

HQHT-308(1) 14:00-14:20 HQHT-308(2) 14:20-14:40 Static Characteristics of Aerodynamic Tilting- Further Studies on Zhang Neural-Dynamics and Gradient Dynamics for Online Nonlinear Pad Journal Bearings Equations Solving Yang Lihua Qi Shemiao Geng Haipeng and Yu Lie Yunong Zhang1, Peng Xu1,and Ning Tan2 Institute of Mechatronics and Information Systems, 1 School of Information Science and Technology, 2 School of Software, Sun Yat-Sen University Key Laboratory of Ministry of Education for Strength and Vibration Guangzhou, China Xi’an Jiaotong University, China • Zhang neural dynamics (ZD) is Fy β = 0 ϕ 1 generalized for solution of nonlinear • The compressible Reynolds z1 equation f(x)=0. equation is resolved. θ

ω • ZD is different from conventional • The static characteristics of

Fx gradient-based dynamics (GD). O δ tilting-pad gas bearing are j 3 γ γ • Further analyze, investigate and calculated. z3 β2 z 2 compare the characteristics of such β 3 • The results are compared with two dynamics ZD and GD. those of reference. Sketch of Tilting-Pad Gas bearing • Three illustrative examples show us some interesting implications and the efficacy of ZD. Convergent Performance of ZD and GD

HQHT-308(3) 14:40-15:00 HQHT-308(4) 15:00-15:20 Strategy of Object Synchronization for DIS Cyclostationarity Analysis and Diagnosis Based on RT-TCP/IP Method of Bearing Faults Wu Bin,, Wang Minjie, Luo yuegang,Feng Changjian Cheng-lie Du, Tao You and Yi-an Zhu Dalian Nationality University School of Computer College, Northwestern Polytechnical University Dalian, China Xi’an, China • Demodulation Feature Analysis of • Object-oriented modeling and Cyclic Autocorrelation Function issues of crosscutting concerns (CAF) . • Aspect-orientedmodeling based • Extraction method of Bearing on high-level attributes of UML Fault Characteristic Defect Frequency by CAF. • An modeling example of • Construction Method of Fault production line simulation system Characteristic Vector Based on • Aspect-oriented expansion of eM- HMM Model. plant based on component The relationship of the expansion components and • Model Training, Fault Diagnosis Three-Dimensional Graph of structure Experiment and Conclusion. eM-Plant system components CAF of Bearing Fault Signals

HQHT-312 HQHT-313 HQHT-314

HQHT-308

HQHT-304 HQHT-305 HQHT-306

Han-Qing Conference Hall - Level 3 (东北大学汉卿会堂)

O-22 FP-3:Intelligent transportation system

Session Chair: Chengjin Zhang and Hong Zhang

HQHT-312 14:00-15:20 Friday, August 7th

HQHT-312(1) 14:00-14:20 HQHT-312(2) 14:20-14:40 Synergetic Method of Traffic State Recognition The Design and Implement of the Decision Support Systems Based on Manifold Learning of Logistics Distributing Center Based on XML Weizhi Wang, Lei Pang and Binghan Liu Tao Wang College of Civil Engineering, Fuzhou University College of Information Technology Fuzhou, China Zhejiang University of Finance & Economics • To solve the problem on large storage of images in the synergetic recognition, Hangzhou, Zhejiang Province, China an improved synergetic recognition method based on manifold learning has [email protected] been proposed in this paper, in which advantage and feasibility of combining the two methods has been discussed. Abstract - According to the research of the theories, concepts • The geometrical structure in high dimensions can be well maintained by using and methodology of Web and DSS, this paper presents a new manifold learning to reduce dimensions while it is somehow innovative to design method of DSS based on XML technology. The paper combine synergetic recognition method and the manifold learning method. describes the developing process of the DSS of logistics • The traffic state recognition has been realized by adopting Isomap algorithm in distributing center based on XML. The problems have been this paper, and the results proves that the method has better recognition effects. solved that how data is transferred and exchanged between the DSS of logistics distributing center, logistics ERP and users and how the decision support models of logistics distributing center Test sample Pre-processing Feature extraction based on Synergetic are constructed. The analysis and design method generated manifold learning recognition from the process of the system realization offers a new method for developing different kinds of the DSS based on Web. Recognition result Learning sample Pre-processing

HQHT-312(3) 14:40-15:00 HQHT-312(4) 15:00-15:20 Study on a new safety control method AUTONOMOUS VEHICLE for a vehicle OVERTAKING- AN ONLINE SOLUTION Shirong YAN[1,2] and Zhiwei Lin[1] 1 School of Mechanical Engineering, Fuzhou University, Fujian, China Ghumman Usman, and Faraz Kunwar 2 Department of Automotive, Minjiang University, Fujian 350108, China Department of Mechatronics Engineering National University of Science and Technology ,College of Electrical and Mechanical Engineering campus, • A 14-DOF dynamic model is Peshawar Road Rawalpindi, Pakistan dopted for a car. z Φ • A novel guidance-based on-line • A new chassis control scheme trajectory-planning algorithm, capable is studied, which combines φ of handling emergency situations is ABS, AFS with DYC. x θ o presented • Some simulation work is done y to test the control method. Zfr Zrr Z • The focus has been on three primary X Zfl aspects: wfr wrr • The control method developed Zrl O wfl here can make the vehicle wrl Y – Time-optimal overtaking having a good braking stability – Safety and better trace in the braking – Passenger comfort. process. vehicle model

HQHT-312 HQHT-313 HQHT-314

HQHT-308

HQHT-304 HQHT-305 HQHT-306

Han-Qing Conference Hall - Level 3 (东北大学汉卿会堂)

O-23 FP-4:Network-based systems II

Session Chair: Aiguo Ming and Yingwei Zhang

HQHT-313 14:00-15:20 Friday, August 7th

HQHT-313(1) 14:00-14:20 HQHT-313(2) 14:20-14:40 Optimal Control of Continuous Annealing Process Grid-based Load Balancing Mechanism Using PSO for RFID Middleware Applications Jiann-Liang Chen, Nong-Kun Chen, Yi-Wei Ma, Han-Chieh Chao and Cheng-Yen Wu Chao Guo, Yingwei Zhang , Xianqiang You, Xue Chen and Yang Zhang Department of Electrical Engineering, National Taiwan University of Science and Technology, Key Laboratory of Integrated Automation of Process Industry, Ministry of Education , Northeastern Taipei, Taiwan University Shenyang , China • Buffer Management. „ The Continuous Annealing Process. • Load Balancing Management. • Priority Agent and Job Relocator. „ Particle Swarm Optimization (PSO). • Monitor.

„ Simulation of Temperature and Energy. • Load Balancing Trigger. .

Load-balancing Architecture The flowchart of the particle swarm optimization algorithm

HQHT-313(3) 14:40-15:00 HQHT-313(4) 15:00-15:20 A Communication Quality Improved Routing Fault Detection for Networked Control System with Protocol for Wireless Sensor Network Data Packet Dropout Fei Chen, Chengdong Wu, Peng Ji, and Yunzhou Zhang Information Science and Technology college, Northeastern University Zhijun Li Dehui Sun Jingjing Zhen Yuntao Shi Shenyang, China

Department of Automation, North China University of Technology • An energy effective routing protocol100 Nodes in rotate 90 The nodes with area based on Low-Energy Adaptive proper are energy 80 nearby node density cosumption Clustering Hierarchy Protocol, are reasonable to be for data 1. Introduction 70 selected as cluster tranportation heads. which is called CQIR, is proposed to60

2. Problem Description meet the QoS requirement and 50 minimize the energy consumption 40 3. The Design of Fault Observer 30

• The CQIR routing protocol 20 The package lost probly happens in the high nodes 4. Simulation Example optimizes the clustering structure by 10 density area. 0 improving clustering heads selection 0 10 20 30 40 50 60 70 80 90 100 5. Conclusion mechanism Cluster head selection

HQHT-312 HQHT-313 HQHT-314

HQHT-308

HQHT-304 HQHT-305 HQHT-306

Han-Qing Conference Hall - Level 3 (东北大学汉卿会堂)

O-24

Poster Session

Wednesday August 5, 2009

WPP Automation I

WPP Automation II

WPP Automation III

WPP Automation IV

WPP Automation V

WPP Computer vision I

WPP Computer vision II

WPP Computer vision III

WPP Evaluation and optimization I

WPP Instrumentation I

WPP Intelligent control I

WPP Transportation system I

WPP:Automation I

Session Chair: Chao Hu and Yanjun Shen

HQHT 15:40-16:40 Wednesday, August 5th

HQHT Hallway (1) 15:40-16:40 HQHT Hallway (2) 15:40-16:40 Design and Application of Auto-constant A Novel Adaptive Control for Elevator System Water Supply System with Observer and Parameter Identification LI Suling and LIU Junying Haiyan Yu, Xin Zhang, Xiaohui Dong School of Electrical Engineering, Shandong University of Technology School of Information and Science Engineering, Shenyang University of Technology Zibo, Shandong Province, China Shenyang, China 1.An automatic constant pressure closed-loop water supply and • An adaptive controller is presented. monitor system of a little type secondary pressure waterworks, which composed of Siemens S7-300 PLC, WinCC and ABB • A servo error observer and a frequency converter, is presented. disturbance rejection control is 2.The principle and construction as well as characteristics of the designed. system are introduced. That has improved the function, stability • The parameters of system are and reliability of the water supply system. identified, thus allowing the 3.The super-harmonic oscillation of waterworks pipe network is observer coefficients and the gain reduced. The purpose of saving energy and reducing consumption has been realized. values to be adapted. 4.The necessary support conditions have been provided for the entire • Simulation results have shown that production management information system of waterworks. the resulting controller achieves Simplified model of an elevator Supported by scientific fund of Shandong university of technology (No:2005KJM09) good performance with robustness.

HQHT Hallway (3) 15:40-16:40 HQHT Hallway (4) 15:40-16:40 Petri Nets Method for Service Engineering Synchronous Motion Control of Biaxial Driving Chuanxi Liu Shandong University of Science and Technology System with Linear Servo Motors Taian, China Dongmei Yu, Qing Hu and Fei Zhao School of Electrical Engineering, Shenyang University of Technology, Shenyang, China • Our research aim is to • High synchronization positioning develop techniques for accuracy for gantry moving type ‘service process control milling machining centre. and scheduling’. • Use dual linear servomotors to drive • Introduce a kind of high- gantry. level Petri Nets—Colored Petri Nets to service • The same position feedback engineering. controller is used to position servo axis independently. • Present a case study by modeling a restaurant • The acceleration feedforward control and electrical cross-coupled Block of electrical service system in the coupling controller. environment of simulation control algorithms based on fuzzy software—CPN Tools. logic are applied to dual linear motor position servo axes.

HQHT Hallway (5) 15:40-16:40 HQHT Hallway (6) 15:40-16:40 Control Synthesis for Polynomial Nonlinear Speed Regulation of PMSM Based on Systems Port-Controlled Hamiltonian Chang-Fei Tong, Feng Wang Systems and PI Control Principle Institute of Operational Research and Cybernetics, Wenzhou University Haisheng Yu, Zongwei Zou and Shanshan Yu Wenzhou, China College of Automation Engineering, Qingdao University • Global controller was Qingdao, China • A novel state feedback control scheme for designed by density function. 10 speed regulation of permanent magnet • Local controller was 5 synchronous motor (PMSM) is presented designed by the proposed 0 x1 based on energy-shaping and port-controlled algorithm . −5 Hamiltonian (PCH) systems theory. • Switching plane was −10 • Proportional integral (PI) control of speed generated by the boundary −25 −20 −15 −10 −5 0 5 10 15 20 25 x2 error is used to estimate the unknown load of attraction domain. torque. Permanent magnet • Performance was improved Switching control for polynomial • The speed controller is implemented based on synchronous motor by the switching control nonlinear systems space vector pulse-width modulation presented. (SVPWM) control rule.

P-1 WPP:Automation II

Session Chair: Chao Hu and Yanjun Shen

HQHT 15:40-16:40 Wednesday, August 5th

HQHT Hallway (7) 15:40-16:40 HQHT Hallway (8) 15:40-16:40 Friction Compensation Using Dual Observer for Research on the Hydraulic Control System of 3-Axis Turntable Servo System Diamond Wire Saw LIU Zhigang, WANG Junzheng and ZHAO Jiangbo Liwen Cao Key Laboratory of Complex System Intelligent Control and Decision, School of Automation School of Mechanical and Electrical Engineering , Beijing Institute of Technology, Beijing, 100081, China Heilongjiang University , China • LuGre dynamic friction model is adopted to • The components and work principle of manual describe the mathematic model of the pitch axis servo system of the 3-Axis turntable. hydraulic control system of diamond wire saw are introduced and the work process and operation • The dynamic LuGre model is built on three unknown parameters. Besides, system torque methods of hydraulic source and control circuit and inertia are also uncertain due to the motor system are analyzed. torque ripple or disturbance. • Then the close-loop control system of radial feed • A dual sliding mode observer is adopted to (a) Position tracking response and the velocity control model of hydraulic servo- estimate these parameters, Lyapunov method is valve system have been established based on the used to design the update laws. reasonable process parameters obtained by cutting • adaptive control law is designed to compensate experiments of the diamond wire saw. the unknown friction and load disturbance. Control valve and manifold • Simulation suggest that the proposed method • The analysis results have shown that the hydraulic blocks of hydraulic system servo system can meet the design requirements. is global asymptotically convergent, which guarantee the position and speed signals track the desired trajectories asymptotically. (b) Velocity tracking response

HQHT Hallway (9) 15:40-16:40 HQHT Hallway (10) 15:40-16:40 Static H Loop Shaping Real-time Control of a Sliding Mode Control Algorithm Based on T-S Fuzzy ∞ Model for a Class of Nonlinear Systems Double Inverted Pendulum System Ming Yan,Jinhui Yi,Yang Liu, JunXiu Wei, Xiaoling Huang Yuyan Liu1, Shiliang Zhou2 1 2 College of Light Industry, Liaoning University Department of Automation ,Department of Nuclear Science & Engineering , North China Electric Power University Beijing, China 、 1 INTRODUCTION • This paper introduces a simple technique for

2、T-S Fuzzy Model designing a static H loop shaping controller. θ ∞ Pole 2 2 3、Design of sliding mode control algorithm • Sufficient conditions are established in Linear Mass of Link θ1 matrix inequality form . Pole 1 Acceleration (1) Design of Sliding Surface • The genetic algorithms are adopted to Rail cart (2) Design of Reaching Condition optimize the loop-shaping weighing functions. x • The order of controller is equal to sum of the (3) Design of Sliding Mode Controller orders of the weighing functions. Real-time A double inverted 4、Simulation results experimental results on a double inverted pendulum system pendulum system demonstrate the validity 5、Conclusion and applicability of the proposed approach.

HQHT Hallway (11) 15:40-16:40 HQHT Hallway (12) 15:40-16:40 Nonlinear Optimal Feedback Control For Delay-Dependent Passivity Analysis of Lunar Module Soft Landing a Class of Uncertain Stochastic Jingyang Zhou, Di Zhou, Kok Lay Teo, and Guohui Zhao Recurrent Neural Networks Departmet of Control Science and Engineering, Harbin Institute of Technology Heilongjiang, China Jie Fu1, Huaguang Zhang1, Derong Liu2, Xiangpeng Xie1 The optimal control problem to 1 Scholl of Information Science and Enginerring, Northeastern University, achieve lunar module soft lan- ding with least fuel consump- Shenyang, Liaoning, P. R. China tion is considered. A closed 2 Department of Electrical and Computer Engineering, University of loop optimal control law is IllinoisChicago, Chicago, IL 60607USA designed with a parameter matrix K to be determined • Introduction about passivity. which is the solution of a • Stochastic recurrent neural network(SRNN) model formulation riccati like differential equation. and preliminaries. We present a practical method to calculate the matrix K, such Coordinate systems • Robust passivity analysis of SRNN with interval time-varying that it is avoided to solve the delay and nominal norm-bounded parameter uncetainties. complex riccati like differential equation. • A numerical example.

P-2 WPP:Automation III

Session Chair: Chao Hu and Yanjun Shen

HQHT 15:40-16:40 Wednesday, August 5th

HQHT Hallway (13) 15:40-16:40 HQHT Hallway (14) 15:40-16:40 Design and Simulation of Ship Course Controller Based Position Loop Design Method in on Auto Disturbance Rejection Control Technique Yaozhen Han1, Hairong Xiao1,Changshun Wang1 and Fengyu Zhou2 Multi Axis Motion Controller 1.Department of Information Engineering , Shandong Jiaotong University, Jinan, China 2. School of Control Science and Engineering, Shandong University, Jinan, China using Extended DDA Circuit • Theory of auto disturbances rejection control is introduced. Jun Zhang and Xu Wang Information science and Technology Institute of Northeastern University • Ship course motion system and disturbances’ mathematic models Shenyang, China are establish. • The core structure of multi axis • Based on the models of the ship course system, a new and high position motion controller. robust ship course controller was designed by using ADRC method. • Extended DDA circuit make the • The ship course control simulation experiments were done under foundation of position loop the uncertain environment. controller. • Suitable for many kinds of δ c δ c motor interfaces. • Suitable for MCU &FPGA Structure of PGU implementations. Ship Course Control Based on ADRC

HQHT Hallway (15) 15:40-16:40 HQHT Hallway (16) 15:40-16:40 An Improved Ant Colony Algorithm for Multi- New delay-dependent criteria for stability objective Flexible Job Shop Scheduling Problem and stabilization of uncertain singular systems with time delay Li Li , Keqi Wang Jinfeng Gao, Meizhen Lei, and Haipeng Pan Information and Computer Engineering College , Northeast Forestry University ,Habin,China Institute of Automation, Zhejiang Sci-Tech University, Hangzhou, China • Introduction. • The problem of delay-dependent robust stability analysis and • General structure of ACA. stabilization for uncertain singular systems with time delay are concerned. • Formulation of multi-objective • The parameter uncertainty is assumed to be norm-bounded and flexible job shop scheduling possibly time varying, while the time delay considered here is problem. assumed to be constant but unknown. • By decomposing the singular time delay system into slow and • Our improved ACA. fast subsystems and using a new Lyapunov-krasovskii functional • Data analysis. which splits the whole delay interval into two subintervals and defies a different energy function on each subinterval, some sufficient conditions are presented for the singular time-delay system to be regular, impulse free and robustly stable. Flow Diagram of Our Algorithm

HQHT Hallway (17) 15:40-16:40 HQHT Hallway (18) 15:40-16:40 Synchronization Criterions of A Kind of General H ∞ Static Output Feedback Control for Constrained Uncertain Linear Systems via Complex Networks with Time Delay Degang Xu, Guojin Liu, and Zhifang Su Moving Horizon Strategy College of Information Science and Engineering, Central South University Changsha, China Juan Wang , Guokai Xu and Jinsheng Sun Department of Electromechanica Engineering and information Dalian Nationalities University • This paper discusses the problems of Dalian Liaoning, China synchronization and pinning control for a kind of general complex • By searching the condition that guarantees the given performance index based on S-procedure technique, an H-inf dynamical networks with time delay. static output feedback control method is proposed for • Less conservative criterions for this constrained system. Moreover, the dissipative constrained kind of complex dynamical networks conditions and the elliptical set that contains output trajectory are obtained. are cast as LMIs, and introduced into the optimization • Pinning control strategies are problem, then a moving horizon H-inf control using static output feedback is presented. The robust closed-loop respectively designed to make these properties, inclusive of stability, H-inf performance and the complex dynamical networks satisfaction of the time-domain constraints are discussed. synchronized.

P-3 WPP:Automation IV

Session Chair: Chao Hu and Yanjun Shen

HQHT 15:40-16:40 Wednesday, August 5th

HQHT Hallway (19) 15:40-16:40 HQHT Hallway (20) 15:40-16:40 Robust control for Markov Jump Linear Study on the Intelligent Hybrid Control for System in fixed time interval Secondary Regulation Transmission System Yanjun Shen Kong Xiangzhen and Zang Faye College of Science, China Three Gorges University, Yichang, Hubei 443002 Shandong Jiaotong University, Jinan, China • In order to overcome nonlinearity • This paper is concerned with the problem of the secondary regulation finite-time control problem of transmission system, an intelligent continuous-time linear system that hybrid controller method is presented. posses randomly jumping parameters which can be described by finite-state • The simulation and experiments for Markov processes. power and speed control prove that the designed controller for the system has • sufficient conditions for mean square rapid response, low overshoot, finite-time stability and mean square finite-timeboundedness via state adaptability. feedback are provided. • The power control and infinite speed • A Detailed example is presented to variation of the secondary regulation illustrate the proposed methodology. automation are realized. It meets the control requirements of stability and The Simulation Diagram robustness when the system works under complex environment.

HQHT Hallway (21) 15:40-16:40 HQHT Hallway (22) 15:40-16:40 Research on solution method of GTD scheduling A Novel Control for Linear Elevator Based on algorithm for WNCS and application Reference Model Sliding Model Jun Wang, Lifeng Wei Xin Zhang,Haiyan Yu, Hao Liu Shenyang Institute of Chemical Technology School of Information and Science Engineering, Shenyang University of Technology Liaoning, China Shenyang, China • This paper introduced scheduling algorithm of GTD. • To the permanent magnet linear • This algorithm resolves real-time problem of cycle communication synchronous motor(PMLSM) for vertical hoist system, discontinuous primary task in WNCS, and eases overtime phenomenon of cycle PMLSM was adopted. communication task caused by wireless network delay partly. • In order to achieve anticipated performances and improved system • We proposed the nonlinear programming model for solving dynamics, a novel compound control scheduling timetable based on the scheduling algorithms. scheme for linear elevator driving by PMLSM with reference model sliding mode control is presented, which integrates advantages of sliding mode control and reference model control. • The simulation results show that this method can overcome external disturbances and obtain good performance compared The physical structure of with traditional PID control. the linear elevator system The scheduling process of all tasks in a macrocycle

HQHT Hallway (23) 15:40-16:40 HQHT Hallway (24) 15:40-16:40 Robust control based on observer for time delay systems Design and Simulation of Fuzzy PID with nonlinear uncertainty Controller for the Movable Beam in Gantry Hou Xiao-li Guo Xiao-li department of Mathis and computer science, Zhengzhou university of light industry, Machining Center zheng zhou, China Xia YANG, YAN ZHAO, Zhu XIAO and Jiefan CUI • the problem of robust control for time-delay system with School of Electrical Engineering, Shenyang University of Technology, Shenyang nonlinear uncertainty is considered, a state observer is exploited Liaoning, China to reconstruct the state of system. z The mathematic model of the maglev beam system is analysed and • Controller is designed by using a new theory of exponential built in this design, and the Fuzzy PID control is used for theoretical stability and a new analysis method. research and simulation analysis. • relationship between state and time-delay state can be explicitly z Cmpared with the traditional PID control, not only the Fuzzy PID given in new theory and the results is shown dependent on the control can reduces the overshooting apparently, but also the upper bound of delay. accommodation time is reduced, and the capacity of resisting • The result is more suitable for fast-varying delay . disturbance is improved apparently, and the requirement of the machined maglev stiffness is satisfied.

P-4 WPP:Automation V

Session Chair: Chao Hu and Yanjun Shen

HQHT 15:40-16:40 Wednesday, August 5th

HQHT Hallway (25) 15:40-16:40 HQHT Hallway (26) 15:40-16:40 Application Research on Enterprise Integrated Application Research on OPC Server Based on Automation Based on OPC & OPC-XML Multi-Fieldbus Qiao Jiaxin QiaoJiaxin School of Information Engineering, Anhui University of Finance & Economics Bengbu, China School of Information Engineering, Anhui University of Finance & Economics Bengbu, China • Enterprise Integrated Automation • OPC Server Based on Multi- Based on OPC & OPC-XML Fieldbus . • OPC provides the high • fieldbus connects directly with performance exchanges in LAN. industry field intelligence devices. • OPC-XML solves the remote • OPC performances data interchange interview and cross platform interview. and share of applications and software/hardware. . • The system model is applied to enterprise information integration. • The system model is applied to • the system model based on OPC enterprise information integration . and OPC-XML must be applied • OPC server based on mutli-fieldbus System Application based on OPC Server system more widely. can be applied in more fields. OPC and OPC-XML based on Multi-Fieldbus

HQHT Hallway (27) 15:40-16:40 HQHT Hallway (28) 15:40-16:40 Synchronization of Unified Chaotic System with Research and Design of Remote Real-time Unknown Parameter Supervisory System Based on OPC in Cement Keyong Shao, Ruiyun Wang, Luanjie Zhou, Yishu Zhan Enterprise School of Electric and Information Engineering, Daqing Petroleum Institute Daqing, China Shaohong Jing , Qingjin Meng 4 School of Control Science and Engineering , University of Jinan 3 Jinan, China 2 1 • A distributed real-time production 0 supervisory designing conception -1 based on OPC and Web technique e1,e2,e3 -2 is presented. -3 • The OPC technique and its -4 application in the DCS of cement

-5 0 1 2 3 4 5 6 7 8 9 10 enterprise are introduced, the t/s seamless connection between the application and plant, and the Web The synchronization errors by using multiple controllers distribution of production data are The Real-time Monitoring realized. System of a Cement Plant

HQHT Hallway (29) 15:40-16:40 HQHT Hallway (30) 15:40-16:40 The Analysis and Design of Low Velocity Optimal Flight Control Design in Irregular Estimation Based on Observer Yan Song1,2, Huibin Gao 1, Shumei Zhang1 ,Yantao Tian3 and Dejun Wang3 Sampled-Data Control Systems 1.Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences; Weikun Zhong and Zhongjian Li 2.Graduate University of Chinese Academy of Science;3. Communication School ,Jilin University School of Automation, Northwestern Polytechnical University Changchun, China Xi’an, shaanxi, China • To solve the problem of the encod- • A design method for irregular er’s finite resolution in low veloci- sampled-data control system is ty . z w researched. G • The design of observer is guided by • As the reason of the possibility the robustness of error dynamic and of data lost and wrong data in S H the reduction quantized noise. the sensors or actuators, K • The experiment results show that w- irregular sampled-data Control d hen tracking the velocity of 1°/s, is considered. the estimated error decreased dram- • An angle of attack control of Irregular Sampled-Data System atically with other three estimators Main experiment result UAV is simulated to illustrate mentioned in this paper. At the sam- the design. e time, the phase lag is also the least.

P-5 WPP:Computer vision I

Session Chair: Chao Hu and Yanjun Shen

HQHT 15:40-16:40 Wednesday, August 5th

HQHT Hallway (31) 15:40-16:40 HQHT Hallway (32) 15:40-16:40 A FPGA-based Auto Focusing System For Research on Traffic Moving Object Detection, Object Distance Estimation Tracking and Track-Generating Shanming Lin , Jun Tang, Xuewu Zhang and Yanyun Lv Wei Liu, Chao Hu and Max Q.-H. Meng Computer and Information Engineering College, Hohai University Shenzhen Institute of Advance Technology Chinese Academy of Sciences, changzhou, China Chinese University of HongKong, China • Newly auto-focusing algorithm which contains interactive focusing • Gaussian mixture model object window selection and three step detection based on RGB color searching algorithm. distribution • Object distance estimation algorithm • Mean Shift Embedded Particle by looking-up lens position table. Filter object tracking algorithm • All algorithm implemented in a • Trajectory generation module FPGA developing board. and simulation results • The best performance of 3% relative The Object Distance System • The application to traffic flux errors when the target is around 8 and The Experiment Result detection meters far away from the camera. statistical traffic data

HQHT Hallway (33) 15:40-16:40 HQHT Hallway (34) 15:40-16:40 Quality Assessment Based on the A Modified Calibration Technique of Camera for Correlation Coefficient and the 2-D 3D Laser Measurement System Discrete Wavelet Transform Huipu Xu, Zi ma, Yuqing Chen Li Junfeng, Dai Wenzhan Automation Research Center, Dalian Maritime University Department of Automatic control, Zhejiang Sci-Tech University, Hangzhou, China Liaoning, China

• the 2-D Discrete Wavelet Transform. Tsai • A novel robot 3D laser measurement modified Tsai • the Correlation Coefficient. system which is representative in re- verse engineering is designed . • Image Quality Assessment Based on the Correlation coefficient and the 2-D Discrete Wavelet Transform. • A modified Tsai algorithm for CCD parameters calibration is presented . • Experiment and Analysis. • From the results, it is shown that the modified algorithm has easier computation, faster speed and higher precision than Tsai . Error comparison between Tsai and modified Tsai

HQHT Hallway (35) 15:40-16:40 HQHT Hallway (36) 15:40-16:40 Multi-garget Tracking using Mixed Spatio- Contrast Enhancement Using Morphological Temporal Features Learning Model Scale Space Ge Yinghui Andrzej Zadorozny and Hong Zhang Faculty of Information Science and Technology, Ningbo University University of Alberta, Edmonton, Alberta, Canada Ningbo, China • A multi-target visual tracking method based on mixed spatio- • We have proposed a new contrast temporal features learning model. enhancement algorithm for images • Particle Filtering is used as algorithm framework. with multiple objects of different sizes. • An incremental • Initialize particle filter and appearance descriptor • This algorithm is multi-scale, direct appearance descriptor • Iterate over image sequences: and adaptive. update strategy to adopt – Transfer particles via transition model – Assign weights to particles using observation model with • We have shown that there exists a change of appearance. embedded with mixed multi-target spatio-temporal features learning model relationship between object size and • HSV histogram and – Localize targets’ position in image scale of enhancement in order to region covariance – Update appearance descriptor – Re-sample particles according to weights improve the segmentation result. descriptor are compared. Summary of Multi-target tracking algorithm • MMS can be used to complement any existing contrast enhancement Enhancement by MMS algorithm.

P-6 WPP:Computer vision II

Session Chair: Chao Hu and Yanjun Shen

HQHT 15:40-16:40 Wednesday, August 5th

HQHT Hallway (37) 15:40-16:40 HQHT Hallway (38) 15:40-16:40 Target Tracking Based on Mean Shift and A novel registration method that Improved Kalman Filtering Algorithm incorporates template Hongxia Chu, Kejun Wang matching and mutual information College of automation, Harbin Engineering University , Heilongjiang, China Shanming Lin, Xiaoyan Zhu, Ping Yan, Jianhua zhou • Domestic and foreign research status Computer and Information Engineering College, Hohai University for target tracking algorithm is Changzhou, China introduced. • Analyze the advantages and • Maximum Fuzzy Entropy Gaussian Clustering is used to improve disadvantages of template matching Kalman Filter algorithm. and mutual information registration. • Model foundation, target positioning • Propose a novel method that and similarity function is incorporates the advantages of represented about Mean Shift target tracking algorithm. mutual information and template • Improved Kalman Filtering matching and the experimental algorithm is combined with Mean results show good effectiveness. Shift. Mosaic image • Conclusion and analyzing. The tracking result

HQHT Hallway (39) 15:40-16:40 HQHT Hallway (40) 15:40-16:40 Algorithm Research on Location of Bottle Mouth and Design of Automatic Target-Scoring Bottom in Intelligent Empty Bottle Inspection System System of Shooting Game Sile Ma1, Bin Huang1, Junmei Guo2, and Huajie Wang1 1. School of Control Science and engineering, Shandong University, Jinan, China Based On Computer Vision 2. School of Electronic Information and Control Engineering, Shandong Institute of Light Industry , Jinan, China Xinnan Fan , Qianqian Cheng , Penghua Ding, Xuewu Zhang Computer and Information Engineering College, Hohai University The bottle mouth and bottom location Changzhou, China is one of the key technologies in the • Propose a technical scheme of empty bottle inspection system design. the Automatic Target-scoring

In this paper, we put forward different System in shooting matches location algorithms for the bottle based on computer vision. mouth and bottom by their different • Analyze and design the characters. Experiments show that this automatic target-scoring method can effectively improve the software and the experimental results show that the system precision of the location of the bottle can recognize the overlapped mouth and bottom and it has great The Intelligent Empty Bottle holes on real time and practical value. the overall structure of the Inspection System correctly. auto-scoring system

HQHT Hallway (41) 15:40-16:40 HQHT Hallway (42) 15:40-16:40 A Novel Method of Face Detection Based on The Hausdorff Distance Template Matching Algorithm Fusing YCbCr and HIS Color Space Based On Kalman Filter for Target Tracking Xue-wu ZHANG, Ling-yan LIANG, Dun-qin DUAN, Wei-liang XIA Faliang Chang, Zhenxue Chen, Wencheng Wang, Lei Wang Computer and Information Institute Hohai University School of Control Science and Engineering, Shandong University Changzhou,China Jinan, China • A face detection method for the • The Hausdorff distance template matching algorithm is color image with complex one of the most popular algorithms. background is presented, which is a • The computation of needs calculating the distance mixed skin-color segmentation (a) (b) model in both YCbCr and HIS color between each point of and each point of, and it takes a space constructed. long time to compute the Hausdorff distance. • Experimental results show that the • The method of small region filtration is to reduce the proposed approach can effectively computing time. prevent the side effects on the (c) (d) image caused by light condition, (a) test image • The experiment results show that the algorithm reduces (b) binary image of YCbCr space reduce the undetected rate, increase (c) binary image of HIS space the time greatly. the speed of the face detection. (d) the fusion mage

P-7 WPP:Computer vision III

Session Chair: Chao Hu and Yanjun Shen

HQHT 15:40-16:40 Wednesday, August 5th

HQHT Hallway (43) 15:40-16:40 HQHT Hallway (44) 15:40-16:40 A Novel FPGA-Based Wireless Vision Sensor Node Reversible image hiding algorithm based on Chao Hu1, Zhiyong Liu1, Yingzi Pan1 , Zhenxing Zeng1 and Max Q.-H. Meng1，2 1、Shenzhen Institute of Advance Technology，Chinese Academy of Sciences pixels difference 2、Department of Electronic Engineering, The Chinese University of Hongkong Honge Ren, Chunwu Chang and Jian Zhang Department of Information and Computer Engineering, Northeast Forestry University • Custom image capture IP core based on Harbin, China Avalon bus was designed and implemented • Using the relevance of two • An efficient HW/SW codesign architecture adjacent pixels to hide data by for JPEG encoder and its FPGA implementation was presented modifying pixel gray scale values. • Hiding capacity can be up to 28 • Wireless communication based on FPGA is 817-88 330b when the PSNR of implemented by using the nRF24L01 carrier image is more than 48dB. wireless RF chip. • The PSNR of carrier image after • Custom peripheral can be added into SOPC restoring is ∞ . as component and the system has been tested The System Diagram well on the Cyclone II EP2C35. Camera

HQHT Hallway (45) 15:40-16:40 HQHT Hallway (46) 15:40-16:40 Detecting Human and Judging Their An Efficient Video Copy Detection Method Pose in Video Zheng Cao, Ming Zhu Guishan Xiang The Key Lab of Network Communication System & Control, Chinese Academy of Sciences School of Information and Electronic Engineering, Zhejiang University of Science & University of Science and Technology of China Hefei, China Technology Hangzhou,,China • Image signature based on spatial

statistics of YCbCr components. Video Query • Researched human detection method Database based on background subtraction • Video signature computed by the ordinal measure of image signature CIT method. Video and image frame temporal order. Signature • Video similarity measured by the • Researched major axis localization VCD Result method based on projection distance of video signature. histogram. • Efficient search approach via Clustering Index Table based on • Researched pose judgment method Major axis localization and video signature clustering. Framework of VCD system based on projection histogram. pose judgment

HQHT Hallway (47) 15:40-16:40 HQHT Hallway (48) 15:40-16:40 A Novel Algorithm for Remote Image Detection of 3D Curve for Shoe Sole Spraying Progressive Coding Based on ROI Based on Laser Triangulation Measurement Guo Shide, Cheng Chengqi, Pu Guoliang and Deng Shujun Yuzhen Jin1,2 , Zuchao Zhu1, Lei Zhang2 and Chuanyu Wu2 RS and GIS Institute, Peking University, China 1) Zhejiang University ,Hangzhou, Zhejiang Province, China • Progressive coding technology based on 2) Zhejiang Sci-Tech University , Hangzhou, Zhejiang Province, China ROI can effectively improve network • It presents the implementation of a transmission efficiency of remote image. new apparatus for adhesive spray of • Firstly analyzes traditional ROI coding the shoe upper. method based on bit plane shift. • a method based on computer vision • Then put forward a novel ROI progressive coding algorithm. The basic Principle is with structured light including in: separating each high frequency subband median filtering, bi-level thresholding, into high frequency ROI subband and high opening and closing, mid-line of the frequency BG subband and rearranging laser stripe extraction, has been transformed wavelet subband. developed . • At last, effectiveness and superiority of this • The 3D scan curves and closed The algorithm structure of algorithm are verified by experimental contour line could be interpolated as analysis. image processing Basic principle of the algorithm B-spline curves.

P-8 WPP:Evaluation and optimization I

Session Chair: Chao Hu and Yanjun Shen

HQHT 15:40-16:40 Wednesday, August 5th

HQHT Hallway (49) 15:40-16:40 HQHT Hallway (50) 15:40-16:40 The Optimization Design Method of Throttle Research of Multi-axis Force Sensor Slice Thickness of Telescopic Shock Absorber Static Decoupling Method Based on Changcheng Zhou1, Jie Meng1, and Zhiyun Zheng2 1 School of Transport and Vehicle Engineering, Shandong University of Technology Neural Network Zibo, China 2 College of Information Engineering, Zhengzhou University, Zhengzhou, China Huibin Cao 1 , Yong Yu 1, Yunjian Ge 1 1 Institute of Intelligent MachineChinese Academy of ScienceHefei, Anhui Province ,China • Analysis of damping components. z Multi-axis force sensor is • Design mathematic model of throttle one of the most important slice thickness based on single sensors applied in the velocity point. realization of Intelligent robot. • Optimization design target function . z Coupling exists in multi-axis • Optimization design of throttle slice force sensor. thickness. z A static decoupling method • Design instance of shock absorber. based on neural network is • Performance test and verifying. The Characteristic Curve put forward. four-dimension finger force sensor

HQHT Hallway (51) 15:40-16:40 HQHT Hallway (52) 15:40-16:40 Chaotic Co-evolutionary Algorithm Based on SOC Prediction Based on Evolutionary Network Differential Evolution and Particle Swarm Bo Cheng1, Yanlu Zhou1, Jiexin Zhang1, 1. School of Construction Machinery, Chang’an University, Xi’ an, China Optimization Meng Zhang, Weiguo Zhang, Yong Sun • Based on biological immune theory, a College of Automation, Northwestern Polytechnical University Xi’an, China • A chaotic co-evolutionary new immune algorithm is presented. 2 algorithm based on DE and PSO 0 • A three-layer feed-forward neural (CCDEPSO) is proposed. network is designed to predict the -2 state of charge (SOC) of Ni-MH • An information sharing scheme , -4

batteries. chaotic initialization and chaotic fitness/log -6 perturbation based on Tent map • Partial least square regression is used -8 CCDEPSO to select input variables are introduced. -10 DE CPSO PSO -12 • Finally, under the state of dynamic • Testing experiments by six 0 500 1000 1500 iterations power cycle, the proposed approach benchmark functions show that can provide an accurate estimation of Evolution curve of the SOC EV-2 the global optimization ability of Ackley function CCDEPSO is better than DE, PSO and CPSO algorithm.

HQHT Hallway (53) 15:40-16:40 HQHT Hallway (54) 15:40-16:40 An Improved Hybrid MILP/CP Algorithm An Adaptive Weight PSO for Rolling Schedules Framework for the Job-shop Scheduling Multi-objective Optimization of Tandem Cold Rolling Huizhi Ren and Lixin Tang Liaoning Key Laboratory of Manufacturing System and Logistics, The Logistics Institute Yong Li, Jianchang Liu and Yu Wang Northeastern University, Shenyang, China Key Laboratory of Integrated Automation of Process Industry Ministry of Education, Northeastern University, Shenyang, Liaoning Provice, China • The Improvement includes Strategies I and II. • An adaptive weight particle swarm optimization is proposed for multi- 3 • Strategy I can generate effective objective optimization. cuts by analyzing the preferred • It is applied to rolling schedules 2.5

jobs on the machines. multi-objective optimization of * eval • Strategy II can generate more than tandem cold rolling. 2 AWPSO one critical cuts by analyzing the • The proposed method decreases the AWA GA 1.5

assigned and preferred jobs on values of three objective functions 0 50 100 150 200 infeasible machines. simultaneously compared to actual generation rolling schedules. Value of eval* in each generation The improved algorithm framework • It has faster convergence speed than the adaptive weight approach GA.

P-9 WPP:Instrumentation I

Session Chair: Chao Hu and Yanjun Shen

HQHT 15:40-16:40 Wednesday, August 5th

HQHT Hallway (55) 15:40-16:40 HQHT Hallway (56) 15:40-16:40 Detection of Weak Magnetic Signal for Magnetic Measuring System of A 3D Force Platform for Localization and Orientation in Capsule Endoscope Plantar Pressure Distribution Huijing Wang , Sanfeng Chen and Junyao Liu Mao Li, Shuang Song, Chao Hu, Houde Dai and Max Q.-H. Meng Department of Computer Application, Shenzhen Institute of Information Technology 1068 Xueyuan Avenue, Shenzhen University Town, Nanshan District, Shenzhen, Shenzhen, Guangdong Province, China 518055, China Y • The design of a novel 3D Platform system c r o s s w i s plantar e suitable for the measurement of human foot d i pressure r e X c t • Design for Magnetic localization platform direction of movement i o pressure distribution is proposed. n

and orientation system. • System description sensor array • Include 192 analog signal channels. – Design of plantar pressure platform – System’s basic block diagram • The range of input analog signal is – Design of sensor array Fig.1 Plantar pressure measurement platform from ±0.5μV to ±30mV. – Sensors conditioning and calibration – Signals precondition • The gain of amplifiers is digital • Extracted gait parameters programmable and from 100 to – Relative length of foot-ground contact 8,000. – Center of force versus maximum force – Foot stability parameter • Applied for the cubic measurement • Applications and future works Fig.2 Extracted gait parameters from The Hardware System footprint image space of 0.5m×0.5m×0.5m. and Block Diagram

HQHT Hallway (57) 15:40-16:40 HQHT Hallway (58) 15:40-16:40 Research on Pointer Location of Multi-pointer Meter High Precision Contactless Object-diameter measurement using IR light source Zhijia Zhang, Yuan Li and Weiqi Yuan Computer Vision Group of Shenyang University of Technology, Shenyang,110178, Liaoning Xinwen Gao, Wei Ding, Tongyue Gao, Zhenbang Gong Province, China School of Mechatronics and Automation, Shanghai University Shanghai, China • 1) Filter the effective area by edge • A novel contactless object-diameter detection and mathematical method based on the binary-value morphology processing method. circuit and Kalman filter algorithm • 2) Rules out the possibility of non- in optical measurement systems. sub-dial area through the feature that • A circuit includes the binary-value those pointers are homogeneously and the data sample is used to obtain distributed at the circle center which more important information from is also the center of dial and the root imaging sensor. of pointer is a circle. • An Kalman filter mechanism is • 3) Extract accurate pointer area by Water meter image constructed to reduce the effects of locating the center of a sub-dial. after accurate pointer measurement noise. Contactless Object-diameter location Measurement Prototype

HQHT Hallway (59) 15:40-16:40 HQHT Hallway (60) 15:40-16:40 Research of cold end Temperature Application of sub-surface imaging system for Compensation of Thermal Couple detecting defects of metal material Quansheng Duan Cheng Yu-hua, Lei Kai College of Control Science and Engineering, School of Automation Engineering University of Electronic Science and Technology of China North China Electric Power University, Beijing, China Chengdu, Sichuan 610054, China • The bi-bridge of compensation solution for thermal couple cold end temperature was recommended. a a' ' ' R R1 R 1 R2 2 E E RW • The improved compensator was c d ' ' • lntroduction Rs1 c Rs2 d ' 4 2 R R ' R composed of main bridge and Cu 3 R3 4 ' b b 5 Ut() • Principles of operation assistant. The main was designed Ut10() 20 according to the traditional method • Image processing t while the assistant was designed 0 • Conclusion and outlook equilibrium at the point of the main 3 tn

bridge perfect compensation. A B • The input-output characteristic curve t 1 of the improved compensation could Principle configuration of the be made approaching the matched Improved Compensator thermal couple’s. magneto-optic device

P-10 WPP:Intelligent control I

Session Chair: Chao Hu and Yanjun Shen

HQHT 15:40-16:40 Wednesday, August 5th

HQHT Hallway (61) 15:40-16:40 HQHT Hallway (62) 15:40-16:40 Applying Hierarchical Reinforcement Learning to Adaptive Fuzzy Control Based on Genetic Computer Games Algorithm for Vertical Electric Furnace Du Xiaoqin, Li Qinghua and Han Jianjun Hongxing Li, Changbo Pan and Fei Teng College of Computer Science, Wuhan University of Science and Engineering Automation College, Beijing Union University Wuhan, China Beijing, China • The adaptive fuzzy control • Existing problems in game design. based on the GA is used for the • Hierarchical finite state machine: HFSM. vertical electric furnace. • An improved genetic algorithm • Hierarchical reinforcement learning method: HAMs. is presented for optimizing the • HFSM vs HAMs for game design. fuzzy controller in real-time. • This algorithm is used to • Advantages: eliminate coupling between the upper and the lower temperature 9 Programming in detail is unnecessary; regions of the vertical electric 9 Easy to implement; furnace in real-time. • In the industrial control network, 9 Learning ability; the experiments of controlling The adaptive fuzzy control 9 Efficiently; the vertical electric furnace are based on the GA performed by using the proposed algorithm.

HQHT Hallway (63) 15:40-16:40 HQHT Hallway (64) 15:40-16:40 Application of MEC-Based Fuzzy PSO Algorithm-Based Schedule Optimization for Tandem Cold Mills Control in Boiler of Sludge Haijun Che, Qing Zhang, Jingming Yang and Xinyan Han Key Lab of Industrial Computer Control Engineering of Hebei Province,Yanshan University Combustion and Power Generation Hebei,China Qingsong Liu, Taibin Cao Jiaxing University，zhejiang, P.R. China • This paper presents Particle Swarm Optimization (PSO) algorithm schedule optimization procedure for • Analysis of the structure for fuzzy tandem cold rolling mills. bed-temperatures control system. • Adopt the power distribution cost • Improvements of MEC . function as objective function, make • MEC-based fuzzy control design the rolling force, rolling power, rolling torque, speed and other • Results of simulation and parameters meet certain requirment. comparison between MEC and • SUMT replaces a constrained improved GA Combustion control system optimization problem with a sequence of unconstrained The flow chart of PSO optimization problem.

HQHT Hallway (65) 15:40-16:40 HQHT Hallway (66) 15:40-16:40 Direct Adaptive Fuzzy Control for a class of Asynchronous Consensus Protocols for Multi- Nonlinear Systems with Unknown Bounds Agent Systems Yongping Pan, Daoping Huang and Zonghai Sun Qin Li and Zhong-Ping Jiang College of Automation Science and Engineering, Polytechnic Institute of New York University, Brooklyn, NY 11201, USA South China University of Technology, Guangzhou, China • Objective: design of asynchronous decentralized algorithms to • Developed a fuzzy approximation realize heading consensus for a group of multi-agents with error estimator as adaptive compensator. same speed. • it is not required that the optimal • Algorithm 1 (leaderless scenario): uc ()x approximation error (OAE) is – Pairwise communication is adopted.

ym x square-integrable or its supremum e ⎧xf()n =+()x gu ()x u f (|)x θ ⎨ – Almost global consensus is reached under certain connectivity condition is known in the stability analysis. ⎩yx= • Proved that controlled object being • Algorithm 2 (leader based scenario): θ(0) bounded is the sufficient condition θ& – One agent acts as a group leader who sticks to its original heading. of the OAE being bounded. – Pairewise communication is not required. • Achieves smooth controller output, – Global consensus is reached under certain connectivity condition. small tracking error and strong robustness.

P-11 WPP:Transportation system I

Session Chair: Chao Hu and Yanjun Shen

HQHT 15:40-16:40 Wednesday, August 5th

HQHT Hallway (67) 15:40-16:40 HQHT Hallway (68) 15:40-16:40 Virtual Detector in Adaptive Traffic Signal Control Longitudinal Distance Control for Vehicle System Based on Support Vector Machine Ciyun Lin1, Zhaosheng Yang1, Bowen Gong1, and Yongchang Zu2 Following Model Based on Tracking Differentiator 1. Dept. Traffic and Transportation, Jilin University, Jilin, China Zhigang Li and Qiuguo Bai 2. Traffic Management Research Institute, The Ministry of Public Security, Jiangsu, China Department of Automatic Engineering Northeastern University at QinhuangDao • Definition of virtual detector Qinhuangdao, China 066004

Fixed detectors D • Platoon dispersion analysis card collected ata • Vehicle following for traffic safety has been an active area of GIS Urban traffic flow flow traffic Urban Probe Cars research. Longitudinal distance automate control is key for vehicle • Detectors data coefficient Database following mode in autonomous cruise system. CCTVs analysis • Tracking-differentiator is designed and applied for the longitudinal Some bus standards (GPIB, Data measure system distance sliding mode control system in this paper. It provides the • Multi-sensor data fusion and VXI,USB) instruments relative acceleration information needed in calculating the subject regression Hardware at field PC computer as software platform vehicle desired acceleration for longitudinal distance sliding mode • A simulation program for control system. adjacent intersections signal The structure of virtual of virtual detector • Theoretical analysis and computer simulation prove that the control designed linear tracking-differentiator can track the vehicle relative velocity measured by sensor and effectively supply the relative acceleration information for distance control.

HQHT Hallway (69) 15:40-16:40 HQHT Hallway (70) 15:40-16:40 An Autonomic Traffic Model Based on Generalized Disjunctive Programming Model for Micro Behavior Information Decision Multi-periodic Continuous Process Scheduling Guiqiu Jiao1 , Chen Guo1, Yining Qin2, Xiaojun Hu2, Xunmin Ou2 Ran Ding, Qiqiang Li and Tao Liang 1 Information Science & Technology College, Dalian Maritime University, School of Control Science and Engineering, Shandong University Liaoning Dalian 116026, CN Jinan, China 2 Low Carbon Energy Lab., TsingHua University, Beijing 100084, CN • In continuous production process, there are many complex

1 Traffic Trip Dynamic Navigation equipments and special requirements such as continuity and Urban traffic has become a worldwide 2 Traffic Trip Time Decision Schedule 3 Trip Fees Budget Policy References problem with the economy developing 4 Trip Exhaust & Emissions stability, which are very difficult to be presented in classical MINLP rapidly, which synthesizes out-of-order jam, Traffic Policy Trip Demand Government formulations. So the production scheduling for continuous process energy consumption and GHG emission. An Traffic Info Organization Road Information autonomic model is proposed based on the Decision Center is more difficult than that for discrete process or batch process. In micro behavior simulation of urban traffic, Traffic Situation Traffic Dispatch continuous process, experience rules usually play very important in which an information decision factor is Trip Decision Road Plan Info. Trip Fees Trip individuals choices roles. But it is impossible to obtain the optimal solution just introduced. Thus the travel patterns and Trip Exhaust influenced by the Physical Network Trip Emission information decision factor: Traffic Running Situ. choice of the road trip have been controlled 1 Trip Decision Traffic Passenger Situ. according to the rules. A novel Generalized Disjunctive 2 Traffic Tools Selection Traffic Supply and optimized dynamically. Simulation with (Bus, Track, Taxi) Public Opinion Programming model for multi-periodic continuous process NetLogo verifies that the model can realize Traveler Network Traffic Network the optimization among traffic, energy and Resident Concerned Enterprise Concerned scheduling is proposed. In this model, some logistic constraints and greenhouse gases emission and some experience rules are represented in disjunctive forms, which make practical use for solving the existing traffic Structure of Urban Traffic Autonomic Model the model more completely. An example is used to illustrate the problems. effectiveness of this method.

HQHT Hallway (71) 15:40-16:40 HQHT Hallway (72) 15:40-16:40 Stability Analysis on Hybrid Optimal Velocity A Practical Method of Road Detection for Model with Relative Velocity Intelligent Vehicle Ruiling Yu Dezhi GAO, Wei Li, Jianming Duan, and Banggui Zheng School of Control Science and Engineering, Shandong University College of Electronic Information & Control Engineering Beijing Jinan, China University of Technology Beijing, China 0.7

• Propose a new backward looking 0.65 width stable region BLAOVM • practical method for road detection Initial detection car following model with a 0.6 high region 0.55 (,ij ) (,ij ) is implemented to detect the lane 11 22 consideration of the relative Vehicle 0.5 new model between lines lines and preceding vehicles velocity (HOVM). 0.45

0.4 Lane lines Vehicle • Analyze the previous car following • distances from host vehicle to left pressing line sensitivity (1/s) models. 0.35 and right lines and preceding target vehicles and their IDR 0.3 unstable region vehicles are estimated. • Propose HOVM which takes into 0.25 0.2 2.5 3 3.5 4 4.5 5 5.5 • algorithm is feasible for different account the forward two vehicles headway (m) road conditions and detect lane and the following one vehicle with The neutral stability lines a relative velocity lines and vehicles robustly and for BLAOVM and HOVM efficiently. vehicle dynamic coordinate • Based on HOVM, the stability of the uniform flow is strengthened

P-12 Thursday August 6, 2009

TPP System Modeling I TPP Robotics and Control I TPP Logistic management I TPP Evaluation and optimization II TPP Evaluation and optimization III TPP Instrumentation II TPP Intelligent control II TPP Intelligent control III TPP Intelligent control IV TPP Intelligent control V TPP Intelligent control VI TPP Intelligent control VII

TPP:System Modeling I

Session Chair: Guohui Tian and Xianming Li

HQHT 15:00-16:00 Thursday, August 6th

HQHT Hallway (1) 15:00-16:00 HQHT Hallway (2) 15:00-16:00 The Stability Analysis and Improvement of ICA-MLP classifier for Fault diagnosis of Rotor SPIDNN Control System System Liu Junying and Li Suling Weidong Jiao 1,2 Yongping Chang 1 School of mechanical engineering, Shandong university of technology 1. Dept. of Mechanical Engineering, Jiaxing University, Jiaxing, 314001 Zibo, China 2. Dept. of Mechanical Engineering, Zhejiang University, Hangzhou, 310027 1. The composition of SPIDNN (Single-output Proportional Integral R14 R2 R1 Derivative Neural Network) control system has been introduced. • A novel classifier is proposed for fault ┇ 2. Under the different applied condition, the improved algorithm was diagnosis of rotor system, with ICA based proposed. independent component analysis based extractor1 ┇ feature extraction and multi-layer 3. The improved algorithm includes the improvement of output function, perceptron based pattern classification. ┇ adoption of the attachment momentum method, the dynamic change ICA based • By the use of ICA, feature vectors are extractor2 of the step length and the elimination of the static error. integratedly extracted from multi- ┇ x 4. Having carried on the simulation, the result indicates that the channel vibration measurements. Thus, a robust multi-MLP classifier insensitive improved algorithm has very good effect in the speed of convergence. ICA based ┇ to the change of operation conditions is extractor3 The validity of the algorithm has been verified experimentally. constructed. ┇ 5. The improved PID neural network control system can realize the • Experimental results indicate invariable fault features embedded in vibration systematic stability. Feature: [f , f , f ] observations can be effectively 1 2 3 Supported by the nature science fund of Shandong province (Y2008F54), China. captured and different fault patterns can Data sampling of rotor system and feature be correctly classified. extraction with multiple ICA networks

HQHT Hallway (3) 15:00-16:00 HQHT Hallway (4) 15:00-16:00 A New Resolution of Workspace Problem of Parallel Research on Controllable Damping Technology Machine Tool and Simulation of Hybrid Machine Tool Chunxia Zhu, Liwen Guan, Jian Han and Liping Wang Lida Zhu Department of Precision Instruments and Mechanology Tsinghua University Northeastern University Beijing 100084, China • Aiming at 3-TPS hybrid machine tool characteristics, controllable damping • In this paper a simple and practical method to technique is adopted in control system resolve workspace of parallel machine tool is of hybrid machine tool, which not only can enhance stability of machine tool, put forward, which is used for simulating the but also effectively restrain vibration workspace of parallel machine tool by using influence in machining process. virtual prototype model. Controllable damper influenced on machine tool is analyzed in theory; • This paper is organized as follows: Firstly it then taken vibration testing in dynamic characteristic as an example, by using introduces the general solution and new ADAMS software simulation validates solution put forward in this paper; Secondly, the method which indicates adjustable the example for solving the workspace of damper control technique can enhance dynamic characteristic of 3TPT parallel machine tool is given by using machine tool. two methods put forward and compared with Workspace of parallel machine tool each other. Lastly, conclusions are given in last Section.

HQHT Hallway (5) 15:00-16:00 HQHT Hallway (6) 15:00-16:00

Large-Angle Attitude Acquisition and Tracking of A Mechanism Analysis of a Novel Four-Degree-of-Freedom Lunar Probe with On-Off Jets Wuxing Jing, Changsheng Gao Parallel Manipulator Based on Larger Workspace Harbin Institute of Technology. Harbin, China Hairong Fang, Jianghong Chen & Congzhe Wang • Rapid attitude control for the Aiming at improving the problems of traditional parallel large-angle multiaxis acquisition 1

q0 robot: covers larger area, smaller workspace. A systematic 0.8 and tracking of a time-varying method is developed to obtain bigger workplace, reference attitude of a lunar 0.6 magnification ratio of Z axis and the compact framework. q3 probe using conventional 0.4 The virtual axes are employed for reduce the constraining of 0.2 quaternions parallel robot’s workspace caused by the length of the links. reaction jets is investigated. q2 0 Based on those a 4DoF (degree-of –freedom ) parallel

-0.2 • Simulation results show the q1 mechanism was designed, which is equally replaced based on

-0.4 0 10 20 30 40 50 60 the concept of generalized pair and the principle of the equivalent kinematic chain; effectiveness of the proposed time(sec) screw theory was used to analyze the DOF; the workspace shape of the traditional and control scheme. this mechanism were defined by drawing, bigger workspace than traditional manipulator Simulation Results can see from the figure. In addition, the movement of the equalized form of the parallel mechanism is simulated with the MATLAB- SimMechanics.

P-13 TPP:Robotics and Control I

Session Chair: Guohui Tian and Xianming Li

HQHT 15:00-16:00 Thursday, August 6th

HQHT Hallway (7) 15:00-16:00 HQHT Hallway (8) 15:00-16:00 Neural Q-learning in Motion Planning for Mobile An Experimental System of Mobile Robot’s Robot Self-Localization Based on WSN Zheng Qin and Jason Gu Chao Peng, Max Q.-H. Meng and Huawei Liang Department of Electrical and Computer Engineering, Dalhousie University Institute of Intelligent Machines, Chinese Academy of Sciences Halifax, Canada Hefei, Anhui Province, China • The neural and case based Q- • This paper proposes an learning (NCQL) algorithm is experimental system which is proposed to improve the s r t +1 t +1 a used for mobile robot’s self- convergence property of t+1 st amax localization based on WSN. reinforcement learning. a t • It simply introduces some related • The reinforcement learning is work of mobile robot’s self- generalized by NN; the convergence localization. property and learning efficiency are promoted by cases. • It describes the experimental system’s structure in detail. • The proposed method is applied to Learning process of NCQL motion planning for mobile robot. • Outdoor experiments and analysis Components of experimental system are given at the end of the paper.

HQHT Hallway (9) 15:00-16:00 HQHT Hallway (10) 15:00-16:00 A Preoperative Ultrasound Planning of Robotic System Optimal Robot Path Planning Based on for Hepatic Microwave Coagulation Jing Xu1, Zhangjun Song2,and Zhenzhong Jia3 Danger Degree Map 1. Michigan State University, East Lansing, Michigan, USA Yinghua Xue, Guohui Tian, and Bin Huang 2.Shenzhen Institute of Advanced Integration Technology, Shenzhen, China 3. Department of Mechanical Engineering, University of Michigan, Michigan, USA School of Control Science and Engineering, Shandong University Jinan, China This paper proposed a needle insertion robotic system for ultrasound guided hepatic • Propose a new environment model microwave coagulation. called danger degree map (DDM). One unique contribution of the system is real-time • Analyze the advantages and prediction of the microwave thermal field distribution generated by the needle’s electrode weakness of equidistant distributed in the liver. PSO. The specific absorption rate of microwave energy (a) • Propose non-equidistant distributed can be calculated and corresponding temperature field can be visualized in the 3D PSO method based on the rate of surgery scene generated by the surgical change of barrier. planning system. Such information can be used to help the surgeon adjust the operation to • The path planning based on DDM improve the treatment effect. has a rapid convergence, can avoid The Path Planning Using Experiments for the system accuracy test (b) (c) obstacles automatically, and is demonstrate that the mean value of the error is Temperature distribution: (a) 3D more reliable and flexible than Non-equidistant PSO less than 2mm. The temperature field scene (b) radial view (c) axial distribution is verified by a 5.6% difference conventional PSO. between real and predicted coagulation volume. view

HQHT Hallway (11) 15:00-16:00 HQHT Hallway (12) 15:00-16:00 Multi-Objective Path Planning for Unrestricted Mobile Optimized Planar Grasping Synthesis Algorithm for Feng Guo, Hongrui Wang and Yantao Tian Multi-Fingered Robotic Hand Institute of Automation, School of Communication Engineering Ze Yu Jason Gu Jilin University Changchun, China Robotics Research Group, Electrical and Computer Engineering Department, Dalhousie University, Halifax, Canada • Problem of multi-objective path • Developed an optimized algorithm planning is investigated in this paper for 3-fingered robot hand’s planar for the ball and plate system. grasping configuration. • The purpose of multi-objective path • Modified the existed configuration planning is to obtain the safe and quality measure criteria, with which shortest path for the ball to track. better configuration can be found • Weights for multi-objectives are calculated by entropy method for among all possible candidates. Task of multi-objective path • Proposed algorithm is applicable for each grid node. planning for ball and plate system • Dijkstra algorithm is employed to BPS-JLU II both polygonal and curved objects solve the multi-objective path and can be expected to be used for planning problem finally. more complicated robotic hands The planar optimized grasping Configuration for 3-fingered robotic hand

P-14 TPP:Logistic management I

Session Chair: Guohui Tian and Xianming Li

HQHT 15:00-16:00 Thursday, August 6th

HQHT Hallway (13) 15:00-16:00 HQHT Hallway (14) 15:00-16:00 Ordering and Pricing decisions in a Two-echelon Supply Supply Chain Coordination for Perishable Chain with Fuzzy Demands Based on Consumer Products with Consumer Returns Policies Decisions Delivery Time Guarantee Chenglin Shen 1, Xinxin Zhang 2, Ke ma1 1 2 1 Chenglin Shen , Xinxin Zhang , Ke ma 1.School of Business,Tianjin Polytechnic University, 2.Tianjin Institute of Pharmaceutical Research 1.School of Business, Tianjin Polytechnic University, 2.Tianjin Institute of Pharmaceutical Research • In this paper, we propose an optimal • This paper considers a single-period supply technique for dealing with the fuzziness chain system consisting of one aspect of demand uncertainties. A triangular manufacturer, one retailer, which the fuzzy number is used to model the external manufacturer firstly gives the wholesale price w, and then the retailer determines demand, and optimal decisions models for product’s retailing price p, ordering quantity both an independent member in the supply Q and the refund r when the wholesale price chain and an integrated supply chain are is at hand. Finally, consumers make developed. The closed form solutions for purchasing decisions according to p and both models are achieved and optimal their valuation v on the product as well as technique is demonstrated by using a single return decisions according to v and r. Next, period supply chain problem. Based on the the optimal pricing, ordering/stocking closed form solutions, the behaviors and quantity, and refund policies under the concentrate and decentralized settings are relationships of the manufacturer and the given. Finally, several contracts are retailer are analyzed, and a cooperative Triangular Demand Model designed to coordinate supply chain and policy is proved to facilitate optimization of coordination mechanisms are further The seller’s optimal price p*, the whole supply chain. discussed. quantity Q*, and refund r*

HQHT Hallway (15) 15:00-16:00 HQHT Hallway (16) 15:00-16:00 The Empirical Research on Cluster Supply Chain The Impact of Learning on Integration Flexibility and Logistics Capacity Capability and Performance: A Weifeng Yang, Zhijun Wu, Jizi Li Knowledge-based View College of Economic and Management, Wuhan University of Science and Engineering Yong Liu, YongJi Jia and ChangJun Wang Wuhan, China Glorious Sun School of Business and Management, Donghua University Shanghai, China • What is the relationship between cluster supply chain flexibility and •The effects of learning on integration logistics capability? capability and firm performance are • What is the role of empirically examined using SEM. information sharing on •Integration capability fully mediates the the relationship between relationships between external learning cluster supply chain and firm performance, and partially The Conceptual Model flexibility and logistics mediates the relationships between internal learning and firm performance. capability? Research model

HQHT Hallway (17) 15:00-16:00 HQHT Hallway (18) 15:00-16:00 Profit Allocation Mechanism for VMI under Risk Research on Customer Order Factors Decoupling Point Positioning Model Dan WANG , Shaoqiang YU for Supply Chain Cost Optimization Dalian Maritime University Jianghua Ge, Fen Wei, Yongtao Huang, and Guoan Gao Harbin University of Science and Technology, • Introduction Harbin, Heilongjiang, China

⎧ Requirement ⎪ acquisition • Operation Process Of VMI Development ⎪ Requirement Customer • Based on the product life cycle, the ⎪ Concept design process of ⎨ analysis requirement of product product ⎪ ⎪ total cost of supply chain was divided ⎪ Concept ⎩ design • Profits Brought By VMI Design stage into five stages. ⎧ of product Process Product Design process ⎪ planning of product ⎨ design ⎪ Process ⎩ planning stage • Profit Mechanism For VMI • Three mathematic models were Production ⎧ Production arrangement Component ⎪ preparation (1) assembly Parts manufacturing (2) (general and developed. Manufacture ⎪ process of ⎨ Supplier (general and customizing pieces) product ⎪ customizing pieces) (3) (4) • Case Study ⎪ Homem ade parts General ⎪ Manufacturer • The cost and time curves of three ⎩ manufacturing assembly Delivery ⎧ merchant models were obtained through ⎪ (5) (6) Delivery ⎪ • Conclusions Retailer Transportation Packaging process of ⎨ brought each parameter into model, so product ⎪ ⎪ ⎩⎪ Customer solved to get the optimal CODP to After service Customer process of service After service minimize enterprise’s total cost. product center The supply chain system structure for product life cycle

P-15 TPP:Evaluation and optimization II

Session Chair: Guohui Tian and Xianming Li

HQHT 15:00-16:00 Thursday, August 6th

HQHT Hallway (19) 15:00-16:00 HQHT Hallway (20) 15:00-16:00 Statistical Characteristic and Parameter Study of cascade PID controller Characterization of 3D Surface Parameter Optimization based Micro-topography on Micro-EDM on IMEA for thermal system Haijuan Ding, Libin Guo and Hai Cui College of Mechanical and Electrical Engineering, Harbin Engineering University Harbin, China Qingsong Liu, Suxiang Qian Jiaxing University，zhejiang, P.R. China • The characteristic of the 3D surface micro-topography of micro-EDM • Analysis of cascade PID control plays an important role on the Main controller Secondary controller system based on restrictions of Method component function properties. The Kp1 Ki1 Kd1 Kp2 Ki2 results show that it is reasonable to stable region. PID 1.2 0.013 28.554 25 0 MEA 0.8532 0.01 15.114 14.84 0.2826 characterize the two types of micro- • Improved Mind Evolutionary IMEA 0.8137 0.011 22.862 25.365 0.0027 EDM surface micro-topography using Algorithm. the statistical functions and 3D Index PID MEA IMEA characterization parameters. A • IMEA-based cascade PID parameter tr(s) 97 132 129 conclusion has been obtained that the optimization for the main steam ts(s) 461 346 250 σ% 25 9.98 5.67 micro-EDMF surface is isotropic and temperature system has much better bearing capacity and 2D ACF of Micro-EDMF Surface • Simulation and Analysis Controller parameters wear resistance performance than the micro-WEDM surface.

HQHT Hallway (21) 15:00-16:00 HQHT Hallway (22) 15:00-16:00 Experimental Research on Identification of Oil-Film Dynamic Characterizing an Equitable Allocation of Fixed Coefficients of Large-size Journal Bearings Costs and Resources via Data Envelopment Zhang Mingshu, Yulie, Xu Zili and Xu Rongdong Institute of Mechatronics and Information Systems, Xi’an Jiaotong University Analysis Xi’an, Shanxi,China Ruiyue Lin • By using the characteristic that the support stiffness Department of Mathematic and information Science, Wenzhou University of the bearing pedestal of balancing machine are Wenzhou, China changeable. • This method does not need any additional excitation • In many applications of data envelopment analysis (DEA), and thinks about the influences of the vibration mass there is often fixed costs or resources, which is imposed on all on the results. decision-making units (DMUs). How can this cost or resource • The related structural coefficients of bearing be assigned in a reasonable way to various DMUs? pedestal can be obtained by dynamic stiffness test. • To obtain an appropriate allocation, this paper proposes a • This method is verified by the test data of a 300 Experiment on the spot MW turbo-generator and be proved to be feasible DEA based approach, which is based on two assumptions: and has much value in engineering applications efficiency invariance and minimal deviation. • For illustrating the method, numerical results for examples from the literature are presented.

HQHT Hallway (23) 15:00-16:00 HQHT Hallway (24) 15:00-16:00 A Fast Setting Method of Parameters in Unified Optimization of H-infinity - Guaranteed FCS Based on Genetic Algorithm Cost Control for Linear Systems with Time-delay LeiLiu, Guoshan Zhang, Zhiqiang Zuo Guangwen Li, Qiuling Jia, Xiaoxiong Liu School of Electrical Engineering and Automation, Tianjin University Northwestern Polytechnical University Tianjin, China

∞ TT2 • A fast setting method based on the Jztztwtwtdtzw =−γ⎡⎤() () () () • H-infinity-guaranteed cost state ∫0 ⎣⎦

genetic algorithm is proposed. the changing law with Dynamic pressure of Parameter K2 32 feedback controller design fitting point • The requirement on the parameter 30 fitting curve γ 28 • Unified optimization of H- and the flying quality are considered. ∞ 26 TT JxtQxtutRutdt=+⎡⎤()JJ () () () infinity performance index and ∫0 ⎣⎦ 24 • To validate the effect of this method, K2 a simulation of a kind of aircraft, 22 the upper bound of the cost 20 * which covers a big flying envelope, 18 function J 16 2 4 6 8 10 12 14 16 18 is made. Dynamic pressure（kPa） • Several examples to * • The result shows that parameters demonstrate the effectiveness min{γ+λJ } The changing law with dynamic gotten by the given fast method are and feasibility of the proposed satisfied. pressure of parameter K2 Unified Optimization results

P-16 TPP:Evaluation and optimization III

Session Chair: Guohui Tian and Xianming Li

HQHT 15:00-16:00 Thursday, August 6th

HQHT Hallway (25) 15:00-16:00 HQHT Hallway (26) 15:00-16:00 Dispatching Optimization and Routing Guidance The Port Effect on Liaoning Economic for Emergency Vehicles in Disaster Development Bowen Gong, Zhaosheng Yang, and Ciyun Lin Li Jing, Xiao Qing Department of Traffic and Transportation, Jilin University School of Transportation Management Dalian Maritime University , Dalian, China Changchun, China Emergency Response System Emergency vehicles Centrally dynamic route • With the model of vector auto • the Framework of Emergency guidance subsystem based on parallel computation subsystem regression (VAR), the effect of port Response System. Waiting for dispatching development on the GDP, • the Integrative Spatial-temporal Routing guidance employment and FDI of Liaoning database. province are researched. • the Dispatching Strategies of Emergency strategy subsystems Other subsystems • The research shows the ports don’t Emergency Vehicles Based on play important roles in GDP, while Hierarchical Cluster. affect the FDI and employment in detector CCTV bacon 红外检测Probe car Remote sensing Telephone reports In-vehicle guidance VMS • Centrally Dynamic Route Liaoning province importantly. The city network Guidance System Based on • Dalian Port play more important Parallel Computing Technology. role in economic development in Dalian port • Simulation Based on Changchun Emergency Response System Lianoing province. network.

HQHT Hallway (27) 15:00-16:00 HQHT Hallway (28) 15:00-16:00 Evaluation of the Logistics Cost of cities Based Modified Particle Swarm Optimization on DEA Model Based on Immune Clone Principle for Xiaofeng Zhao and Qi Tang Department of Management, Tianjin Polytechnic University Analog-Matching of Equivalent Tianjin, China Yong Sun, Weiguo Zhang, and Meng Zhang • Introduction ： Background and the College of Automation, Northwestern Polytechnical University Xi’an Shaanxi, China research condition. Decision ∗ Returns to s Bode Diagram Number ∗ ∗ -2 θ σ Making Unit scale • Low order equivalent system of High_System -4 • Introduce the DEA model to help to Equ _Sy s tem aircraft can be calculated and -6 DMU1 Tianjin 1 1 1 unchanged evaluate the logistics cost. evaluated flying qualities of high -8 Magnitude (dB) • Collect the actual data and adopt the DMU2 Shanghai 1 1 1 unchanged order system. -10 -12 DMU3 Beijing 0.72 0.765 0.94 decreasing • A modified particle swarm 45 DEA model to analyze the logistics DMU4 Suzhou 1 1 1 unchanged 0 optimization based on immune clone -45 cost. DMU5 Lianyungang 1 1 1 unchanged -90 Phase (deg) Phase principle is proposed aiming at -135 DMU6 Nanjing 1 1 1 unchanged -180 • Then according to the data and analog-matching of equivalent system. -1 0 1 10 10 10 Frequency (rad/sec) analysis results, we can analyze the THE LOGISTICS EFFICIENCY • The advanced algorithm has important causes of relative high logistics cost significance in online evaluation of Amplitude-frequency and aircraft flying qualities. and find the keys to improve. Phase-frequency Curve

HQHT Hallway (29) 15:00-16:00 HQHT Hallway (30) 15:00-16:00 Optimization of the Logistics node layout model A Fuzzy Decision Making Model to Select the based on SGA Location of the Distribution Center in Logistics Liupeng Jiang, Xuejun Feng, and Wei Wang Xuefeng Yu Xiaowei Zhang Lili Mu School of Hydraulic and Electric Engineering，Heilongjiang University ,Harbin 150086,China Transportation and logistics engineering institute, Hohai University • Seeking best logistics center location (LCL) is a challenge Jiangsu, China problem and affected by many factors, these factors consist of

Supply Logistics Demand many quantitative indicators and qualitative indicators, so the • Build an optimal logistics node layout Point Point Point model with considering the land constraints LCL belongs to the fuzzy semi-structured decision-making and scale economy of freight. problem. • Do computational analysis to solve the model by SGA. • In order to solve the problem, this paper presents the new • Use example to analyze the feasibility of optimal selection for alternative programs of logistics center above-mentioned model. using the fuzzy decision-making model (FDMM), this algorithm • The paper only considered the reasonable arrangement of logistics nodes, the based on Engineering Fuzzy Set Theory (EFST), has been classification question of logistics nodes, intensively tested in real system. multi kinds and multilevel logistics nodes The Chosen Logistics Point layout optimization question are next steps The non-chosen Logistics Point • The paper introduces the FDMM and its three characteristics, to study. Schematic diagram of problem to be solved and then takes an example to discuss the specific application of the model in LCL.

P-17 TPP:Instrumentation II

Session Chair: Guohui Tian and Xianming Li

HQHT 15:00-16:00 Thursday, August 6th

HQHT Hallway (31) 15:00-16:00 HQHT Hallway (32) 15:00-16:00

Theoretical Analysis and Experimental Investigation Technological Study on Distributed Fiber of Fiber-optic Coupler’s Strain Characteristic Jian Xu 1 Bin Ma 2,3 Xin-guo Zhou 2 Sensor Monitoring of High Voltage (1:School of Computer and Information Engineering, Shandong University of Finance,Jinan,250013,China) (2:Department of Information Science and Technology, Shandong University of Political Science and Law,Jinan,250014,China) (3:School of Control Science and Engineering, Shandong University,Jinan,250061,China) Power Cable in Seafloor Qi Jiang，Qingmei Sui • Introduction • Theorical Analysis of Fiber-optic School of Control Science and Engineering Shandong University , Jinan City, Shandong Coupler Province ,China • Experiment Research of fiber- optic coupler’s strain,temperature and transverse charactersic Temperature online monitor in cable from distributed optic fiber • Experiment analysis and sensor conclusion Picture of experimental device

HQHT Hallway (33) 15:00-16:00 HQHT Hallway (34) 15:00-16:00 Design of Intrinsically Safe Intelligent Study on Application of 3-D Laser Scanning Water-level Monitor Used in Coal Mine Technology in Forestry Resources Inventory Lifu Zhang WU Yanrong College of Mechanical and Electrical Engineering, School of Information and Electric Engineering, Shandong University Northeast Forestry University, Harbin, China Jinan, China • The study concentrates on the automatic • Intrinsically Safe Intelligent Water- Battery Voltage box Regulator detection of trees and the subsequent level Monitor Used in Coal Mine to LCD Display master the dynamic change law of determination of tree height and DBH. Real-time clock groundwater Voltage rising circuit CPU • Reliability and precision of techniques • The monitor has the functions of RS-232 Communication for automatic point cloud processing automatic observation of water level were analysed. periodically, real-time display of Signal Keyboard groundwater level, data storage and Sensor adjustment • The test shows that 3-D laser scanning communication with the computer. technology has a wide application Tree detection The monitor is consisted of CPU, prospects in Forestry Resources sensor, LCD, real-time clock, Structure of the system EEPROM, keyboard, RS232 interface Inventory. and other circuits.

HQHT Hallway (35) 15:00-16:00 HQHT Hallway (36) 15:00-16:00 The Array of Micro-reactor PCR Chips and Its Temperature Measurement System for Ceramic Control System Kilns with Wireless-Communication Based on

Xianming Li, Haifa Liu nRF9E5 School of Control Science & Engineering, Shandong University Jinan, China Yu Zhang, Shulian Yang, Lei Han and Lidan Hou

• The design of low cost micro- Interchanger PCB board Chip array School of Mechanical Engineering reactor polymerase chain reaction Shandong University of Technology, Zibo, China (PCR) chips . Circul- ating • The wireless transmission of • The temperature control plan of pump micro-reactor PCR chip array. measurement signal is realized via the nRF9E5 chip for the verification • Micro-heater and sensor array requirement on the temperature based on flexible board. measurement system of ceramic kilns. Back Four Infall water • The control system of PCR chip water switch switch • The developed system provides a tanks array based on industry control valve valve small size, high stability and high computer. reliability through well-designed • Statistics process control to the communication protocol. Structure of Wireless Temperature contact thermal resistance Schematic drawing of the chip Measurement System between PCB and PCR chip array

P-18 TPP:Intelligent control II

Session Chair: Guohui Tian and Xianming Li

HQHT 15:00-16:00 Thursday, August 6th

HQHT Hallway (37) 15:00-16:00 HQHT Hallway (38) 15:00-16:00 The Negative Pressure Control of Ball Mill Based Real-Time Controlling of Inverted Pendulum by on T-S Fuzzy Model Fuzzy Logic Xinli Fang, Tao Shen Liu Yanmei1, Chen Zhen2, Xue Dingyu2, Xu Xinhe2 Institute of Automation, Jinan University 1. Shenyang Institute of Aeronautical Engineering, Shenyang 110136 Jinan, China 2. School of Information Science & Engineering, Northeastern University,Shenyang 110004 • The negative pressure of the cement ball mill fluctuate • The basic aim of our work was to balance a badly, this article introduces the T-S fuzzy strategy to real pendulum in the position in center. deal with the problems. • The fuzzy logic controller is designed in the Matlab-Simulink environment. • From the real use in • In this paper, the inverted pendulum cement factory it shows The Material Separate Valve Status Curve mathematical model is built. MATLAB based the T-S arithmetic can Hardware in Loop simulation system is bring benefit effects of designed. A novel expert fuzzy control the negative pressure. scheme was proposed. Thin Store negative Pressure Curve • The proposed control scheme was implemented and showed good performance in the real-time fuzzy control of the inverted pendulum.

HQHT Hallway (39) 15:00-16:00 HQHT Hallway (40) 15:00-16:00 Method for Subway Operation HAM Homomorphism for State Abstraction Adjustment Based on Multi-Agent Du Xiaoqin, Li Qinghua and Han Jianjun Fei Lu, Zhihui Zhang College of Computer Science, Wuhan University of Science and Engineering School of Control Science and Engineering, Shandong University Wuhan, China Jinan, China • Introduction. • Hierarchical reinforcement learning method: HAMs. • Model of subway train operation • Existing problems in HAMs. adjustment under moving block • Homomorphism in HAMs. system. • Contributions: • Algorithm of operation adjustment 9 Providing formal definitions of homomorphism in HAMs; based on multi-agent. 9 Proving the invariance of the optimal solution for HAMs; • Simulation. • Advantages: 9 Forming useful abstractions at different levels of a hierarchy; 9 Speeding up learning;

HQHT Hallway (41) 15:00-16:00 HQHT Hallway (42) 15:00-16:00

Fuzzy Control Used in Variable Speed Wind Robust Adaptive Beamforming under Quadratic Turbine Constraint in the presence of Mismatches Xinyan Zhang, Xiaobo Zhang, Peiyi Zhou,Jing Cheng and Weiqing Wang Xin Song Jinkuan Wang Bin Wang, Yinghua Han Electrical Engineering College,Xinjiang University Optimization and Smart Antenna Institute, Northeastern University at Qinhuangdao, China Xinjiang, China • The proposed algorithm is based on explicit modeling of uncertainties in the desired • The fuzzy relations between the variables and the fuzzy rules were stated in detail. signal array response and data • Simulation model and results are covariance matrix. obtained. • By using fuzzy controller, wind turbine • To decrease computation com- generator system has better anti- plexity, the weight vector is interference property compared with the PID controller. obtained by gradient-descent • The fluctuation of the power is method. minimized, the maximum power is traced Output SINR versus N and the rated power is maintained. Simulation diagram • It provides an improved robustness against mismatches

P-19 TPP:Intelligent control III

Session Chair: Guohui Tian and Xianming Li

HQHT 15:00-16:00 Thursday, August 6th

HQHT Hallway (43) 15:00-16:00 HQHT Hallway (44) 15:00-16:00 Applications of Immune Genetic Algorithm in Controlling Study of D-STATCOM Based on Reinforcement Learning Adaptive PID Frequency Speed Control Pumping Station Xing-ping Meng Hui Wang Liang Zhao Hong Zhang Jin Wang Hui Wang Hong Zhang Xing-ping Meng Changchun Institute of Changchun Institute of Jilin Electric Power Changchun Institute of Changchun Institute of Changchun Institute of Changchun Institute of Changchun Institute of Technology Technology supply company Technology Technology Technology Technology Technology In distribution power system, these power Circulating cooling water system of high quality problems mainly arise due to the speed wire factory was in trouble for pulsed loads, which causes the degradation pumps working in higher energy of the entire system performance. consumption and lower efficiency. A novel adaptive control strategy for the The paper analyzed the characteristic of Distribution Static Compensator speed water pump. Mathematical model (DSTATCOM) is based on the was established and it simplified Reinforcement Learning Adaptive PID inequality constraint conditions. Immune (RLA-PID). With this control method, the genetic algorithm (IGA) to handle compensator can stabilize the PCC voltage constrained problems was proposed. around a prescribed value. A distributes Simulation results show that it is an power system is developed in MATLBA effective algorithm to solving optimal environment, the validity of the method problem of frequency speed control proposed is verified. pumps in water circulating system.

HQHT Hallway (45) 15:00-16:00 HQHT Hallway (46) 15:00-16:00 A New Hybrid Algorithm for Image Segmentation Fault Diagnosis of High Voltage Based on Rough Sets and Enhanced Fuzzy C- Direct Current System Based Means Clustering on Particle Filter Wei Zhang1, Yu-zhu Zhang2 , Cheng Li1 1. Qingdao University of Science and Technology; Xi-mei Liu, Chao Zhang , Hong-mi Guo 2. Luoyang Institute of Science and Technology College of Automation and Electronic Engineering Qingdao University of Science and Technology Qingdao, Shandong Province,China A new particle filter based fault • In this paper, in order to speed up the diagnosis method for nonlinear segmentation process, an improved hybrid stochastic system with non- Gaussian noise and disturbances algorithm called rough- enhanced fuzzy c- is proposed by combining means (REnFCM) algorithm is presented particle filter algorithm and fault for segmentation of brain MR images. diagnosis theory. Experimental results show the method is efficient and applicable to HVDC REnFCM result system.

Pm AndPf when phase A to ground fault

HQHT Hallway (47) 15:00-16:00 HQHT Hallway (48) 15:00-16:00 Error Analysis and Compensation of Single- Self-Tuning Fuzzy PI+PD Controller for PMSM beam Laser Triangulation Measurement Servo System Bing Li Jianlu Wang Fei Zhang and Lei Chen Xianqing Cao，Meng Bi State Key Laboratory for Manufacturing Systems Engineering, Xi’an Jiaotong University College of Information Engineering ,Shenyang Institute of Chemical Technology Xi’an, China Network Management Center, Shenyang University of Technology Shenyang, China • Error Analysis of Single-beam Laser Triangulation Measurement •In order to overcome the parameter uncertainties, enhance • Profile Measurement Model based the fast tracking performance of PMSM servo system, a self- on B-Spline Extrapolation tuning fuzzy PI+PD controller is proposed. • Error Compensation of Single-beam •An load torque observer is adopted and the observed torque Laser Triangulation Measurement is feedforward to compensate the load torque disturbance. • Experimental Results and Analysis (a) The global size view This paper realized the profile measurement through combining motion path of the probe with the change of the profile and eliminated the effect of the small depth of field on the basis of three times B-Spline curve that constructed by the probe position. The measurement accuracy can be improved by controlling the probe inclination angle with the normal direction of the interpolation curve. The influence of the depth of field in the measure process is decreased and the precision (b) The partial enlarged view of the profile measurement is enhanced through the error compensation on the range of the probe depth of field. Results before and after the measurement Compensation Simplified control system block diagram

P-20 TPP:Intelligent control IV

Session Chair: Guohui Tian and Xianming Li

HQHT 15:00-16:00 Thursday, August 6th

HQHT Hallway (49) 15:00-16:00 HQHT Hallway (50) 15:00-16:00 Robust Multi-objective and Nonlinear Control of A stable self-learning PID control based High-rise Elevator Vertical Motion System Based on the artificial immune algorithm Yang Liu on T-S Fuzzy Model Using PDC and LMIs Department of Electrical and Electronics,Shandong University of Technology Haiyan Yu, Xin Zhang, Hongxia Li Zibo City, Shandong Province, China School of Information and Science Engineering, Shenyang University of Technology Shenyang, China • The model is adapted by artificial immune clustering • An linear matrix inequality(LMI)- algorithm to learn plant based and Takagi-Sugeno(T-S) dynamic change . model-based fuzzy approach is • the PID control parameters employed. are adapted by the Lyapunov • Parallel distributed compensation method to minimize a cost function. (PDC) scheme • The artificial immune Configuration of • State feedback robust controller algorithm is effective to PID artificial immune-PID was gained based on LMI method. Elevator Physical Model tuning.

HQHT Hallway (51) 15:00-16:00 HQHT Hallway (52) 15:00-16:00 Implementation of a New Fuzzy Control Strategy Design, Control and Test of a on EHB System Magnetorheological Fluid fan Clutch Hui Xu, Chuan-xue Song, Shou-tao Li, and Chang-fu Zong Hu Hongsheng, Wang Juan, Cui Liang, and Wang Jiong College of Automobiles Engineering, Jilin University Mechanical and Electrical college, Jiaxing University Changchun, Jilin, China Jiaxing, China • Introduction. • EHB systems in which the mechanical link between the braking cylinder and • Torque transfer model of MR braking pedal of an automobile are replaced by a control system seem to clutch. offer great advantages on enhanced • Magnetic circuit optimization system performance. In this paper, an EHB control system configuration is of MR clutch. presented. The proposed system provide ideal brake force according to • Control strategy simulation and the driver’s breaking intention and a controller design. PID Controller to regulate the wheel brake cylinder pressure. Simulation Control strategy block diagram • Test of MR clutch. result shows very good performance in of EHB system The developed MR clutch the experimental braking • Conclusions.

HQHT Hallway (53) 15:00-16:00 HQHT Hallway (54) 15:00-16:00 Research of a Boundary Prolongation Method in Chaos System Filter on State-space Model and EKF Runoff Forecast Based on Wavelet Transform Chen Yong, LiuXia, HuangQi and ZhangChanghua Yong-ying Zhu, Wei Pei, Zai-long Man, and Jing Wang University of Electronic Science and Technology of China,Chengdu,China Civil Engineering department , Dalian Fisheries University Dalian, China 20 60 10 • Chaos motion is a exclusive motion form of 40 z

x 0 • A new analogical boundary nonlinearity dynamics system, it widely exists in 20 0 -10 50 20 nature. Such as physics, chemistry, biology, 0 0 y x -20 extension mode is proposed based -50 -20 0 10 20 30 40 geology, as well as technique science, social T on the similarity measurement. science and so on. The exclusive characteristic 30 50 20 40

10 chaos system is that the evolvement of system is 30

• The mode can characterize the y 0 z 20 very sensitivity to initial conditions, so that the -10 10 self-correlation of runoff sequence -20 future of chaos system is uncertain in long range. -30 0 0 10 20 30 40 0 10 20 30 40 excellently with the local • Based on sates apace theory and EKF, Chaos T T similarities on the wavelet multi- system filter on state-space model and EKF is scale spaces. proposed in the paper, the states of chaos system are forecasted by the filter, At last, it takes • The case study indicates that the Lorenz system for example, founds the state- extension mode is very effective space model of Lorenz system and effectively estimated to the three attractor of Lorenz system for the runoff sequence with The annual runoff with through the proposed filtering algorithm. obvious tendency. Simulation results on MATLAB show the different extension modes proposed filtering algorithm is a effective • The boundary extension errors are method to estimate parameters of chaos system low and the forecast precision is and filter. improved. Fig.2. Simulation result of three state parameters estimation

P-21 TPP:Intelligent control V

Session Chair: Guohui Tian and Xianming Li

HQHT 15:00-16:00 Thursday, August 6th

HQHT Hallway (55) 15:00-16:00 HQHT Hallway (56) 15:00-16:00 Advanced Suspension Systems for Improving The Title of My Paper on Digital Welding Vehicle Comfort Power Sources Jiammin Sun2, Qingmei Yang1 Bin Duan, Tongjing Sun, Zhenhua Li, and Gaoqing Mei 1. College of Automation, Beijing Union University, Beijing, China Department of Control Science and Engineering, Shandong University 2. Beijing University of Civil Engineering and Architecture, Beijing, China Jinan, China • The digital welding machine works • An advanced controller with in a complicated harsh environment rectification factor is designed. and needs a fast algorithm for the • Contrasting with common fuzzy multi-parameter coupling nonlinear algorithm, there is no membership welding process . function choice of fuzzy subset for • Artificial neural network is used to input and output of controller. search optimal combination of PID • The rectification factor is adjusted parameter. online according to LMS adaptive • Several expert rules are proposed to method. improve the performance of ANN- • For the sprung mass acceleration, the Sprung mass acceleration PID. simulation results show that LMS adaptive fuzzy control algorithm has PSD of advanced suspension • The new algorithm can avoid obvious effect on the vibration control shortcomings of BP algorithm and The Fronius Fully Digitised of suspension system. meet the online real-time control. Welding Power Source

HQHT Hallway (57) 15:00-16:00 HQHT Hallway (58) 15:00-16:00 Neural Networks with Weight Function Delay-dependent Variable Structure Control for and Application Uncertain Neutral Delay Systems Youguo He Lei Zhao, Yang Gao and Jian Zhou Zhang Niaona, Lu Xiuhe, Zhang Dejiang, Chen Fang Faculty of Information Engineering, Shenyang University Shenyang Artillery Academy College of Electrical and Electronic Engineering , Changchun University of Technology , China Shenyang, China Shenyang, China • Simple network topology constituted by • The Variable Structure Control input layer and output layer only. 2 problem of a class of uncertain 1.5 neutral delay systems is considered. • Used in establishing the Energy 1 Consumption Forecasting Model of • The paper gives a delay-dependent 0.5 sufficient condition in terms of DAGUSHAN Ore Dressing Plant. err 0 linear matrix inequalities such that • Different interpolation functions are -0.5 the closed-loop system is guaranteed

selected to be the weight functions. -1 to be asymptotically stable. -1.5 • The control method guarantees that • Simulation examples show the good 0 10 20 30 40 50 60 the trajectory of system arrives at the performance of this method that little sliding surface in finite time interval calculation work, with no local minimum and is kept here thereafter. and slow convergence problems. Model The error of Energy mentioned above has minor error and the Consumption Forecasting Model • A simulation is given to illustrate the Responses of System State effectiveness of the approach. better prediction effect is obtained. Variables (x1,x2,x3)

HQHT Hallway (59) 15:00-16:00 HQHT Hallway (60) 15:00-16:00 An Intelligent Approach for Construction Domain A Distortion Tester of Geophone Based on FPGA Changzhi Lv, Di Fan,Bo Shi and Wei Wang，and Zhihai Liu Ontology Shandong University of Science and Technology,Geophysical Exploration of Shengli Petroleum Yan Yaxin, Jiang Zhenhuan, Liu Xinhua and Zhang Xutang Qingdao China Dongying China School of Management, Harbin Institute of Technology Harbin, China • Moving coil geophone is a linear second order sensor approximately. • 32 points rectangle window is adopted in our FFT because of its low analysis error and simplified computation. • The DDS and FFT are realized on the FPGA . • Experiments results show that the Block diagram of tester measurement is sample and test Merged process knowledge ontology for a cylinder precision is high.

P-22 TPP:Intelligent control VI

Session Chair: Guohui Tian and Xianming Li

HQHT 15:00-16:00 Thursday, August 6th

HQHT Hallway (61) 15:00-16:00 HQHT Hallway (62) 15:00-16:00

Research on a Direct Track Intelligent Application of Improved Neural Network Optimizing PID Control in Hydroelectric Control Method of Large Ship Generating Unit Xiaoyun Zhang, Yuanhui Wang, Yufeng Sun and Shuying Li Shuqing Wang , Hui Liu , and Zipeng Zhang Harbin Engineering University, Harbin, China Hubei University of Technology • Calculate the track deviation. Wuhan, China • fuzzy - PID control strategy is • Design of the control system for hydroelectric used for direct track at the generating unit. ψ ()k −<θε turning point. 2 • PID controller for hydroelectric generating unit. • The track fuzzy controller is a two-dimensional controller. • Improving of neural network PID controller. • The results is simulated on the • Simulation analysis of neural network control basic of a certain large surface The theory diagram of direct ship. Fuzzy-PID ship track control

HQHT Hallway (63) 15:00-16:00 HQHT Hallway (64) 15:00-16:00 Phosphorus Endpoint prediction of AOD furnace ferroalloy melting Based on Wavelet Neural New Internal Model Control Method for Network Multivariable Coupling System with Time Delays Jin Qi bing, Quan Ling, and Yuan Qin Wang Dongmei, You Wen, Lin Xiaomei , Tian Yantao College of Information Science & Technology., Beijing University of Chemical Technology Changchun University of Technology, Jilin University Beijing, China • The network structure of wavelet 35

-3 响应 r1=1 r2=0 x 10 耦合 neural networks • Introduced the principle of decoupling 1.5 6 4 30 1 internal model contorl. 2 0.5 y1 y2

) 0

• Learning Algorithm of wavelet -5 0 -2

(*10 • The proposed IMC controller design 25 -0.5 -4 neural networks 0 200 400 600 800 1000 0 200 400 600 800 1000 Time Time method was described. r2=1 r1=0

磷含量 1.5 0.02

• The BP neural network model of 20 1 0.01

• The improved ISE index and NLJ y2 0.5 y1 0 endpoint phosphorus value 0 -0.01 15 algorithm were adopted to optimize the 0 5 10 15 20 25 30 -0.5 -0.02 0 200 400 600 800 1000 0 200 400 600 800 1000 • The wavelet neural network model 预测样本数 parameter of the filter . Time Time of endpoint phosphorus value The comparison chart of actual • Two examples are given to illustrate The simulation of • The analysis and comparison of target data and predict data the effectiveness of the proposed Unmatched models the result of simulation. method.

HQHT Hallway (65) 15:00-16:00 Prediction Model of Quality Indices Based-on RBF Neural Network in the Raw Slurry Blending Process Bai Rui1, Tong Shaocheng1, Zhang Jian1, Chai Tianyou2 1 College of Electrical Engineering, Liaoning University of Technology, Jinzhou, China, 2 Research Center of Automation, Northeastern University, Shenyang, China

• An intelligent prediction model of quality indices of raw slurry is proposed. • The prediction model of calcium ratio is comprised of 6 modules. • Long-term running results show that the intelligent prediction software can predict the value and trend of calcium ratio on-line with high accuracy Raw Slurry Blending Process

P-23 TPP:Intelligent control VII

Session Chair: Guohui Tian and Xianming Li

HQHT 15:00-16:00 Thursday, August 6th

HQHT Hallway (66) 15:00-16:00 HQHT Hallway (67) 15:00-16:00 The Recognition of Working Double Inverted Pendulum System Control Condition of Cement Rotary Kiln Strategy Based on Fuzzy Genetic Algorithm Ding Cheng-jun , Duan Ping , Zhang Ming-lu Based on Information Fusion School of Mechanical Engineering, Hebei University of Technology Tianjin China Zhugang Yuan Department of Control Science and EngineeringHarbin Institute of Technology μ (z) ΔU Harbin, Heilongjiang, China LVB LB LM LS NA RS RM RB RVB • Using Genetic Algorithm to 1 • The rotary kiln is a non-linear controlled object with multi- optimize the weighting factors, slow time-variable, big time lag and close coupling, coefficient and membership and it is difficult to establish its precise mathematical model. function parameters. z 78 • Then, using the optimizing -8 -7 -5-6 -4 -3 -2 -1 0 1 234 5 6 • In this article the tendency type of the main result to control the double Before optimizing examination parameters inverted pendulum system. After optimizing is recognized with an The membership function’ improved ART-2 cluster curve of control variable u parsing algorithm.

HQHT Hallway (68) 15:00-16:00 HQHT Hallway (69) 15:00-16:00

Cooling High-speed Chips Using Intelligent L2 Robust Control for Brushless Doubly-Fed Control Technology Wind Power Generator Baosheng Yang, Xiushui Ma Fengge Zhang, Shi Jin and Xiuping Wang Department of computer science and technology, Suzhou University College of Electrical Engineering, Shenyang University of Technology Suzhou, China Shenyang, China • In order to keep high-speed chips within good performance, some • The essence of maximum wind 11 cooling methods are necessary. energy tracking is transformed into 10 • Fans can dramatically reduce the generator speed control problem. 9 wind 8 temperature of a chip. But the v /(m/s) speed •L2 robust control is adopted to design 7

a speed controller for variable speed 6 continual noise from the full-speed 0 2 4 6 8 10 constant frequency brushless doubly- t /s cooling fan is inevitable. 1200

fed wind power generation system. 1100

• The temperature sensor IC measures ) 1000 n • The proposed robust controller i m / generator r 900

the chip temperature and adjusts fan /(

enables the generation system to be nr 800 speed accordingly. speed better in robustness against 700 600 0 2 4 6 8 10 • The effects of noise in the work Feedback-based PWM parameters uncertainties, to rapidly t /s place can be solved in this way. drives circuit of fan track wind speed, and further to track BDFG speed response plot with maximum wind energy. random wind speed changing

HQHT Hallway (70) 15:00-16:00 HQHT Hallway (71) 15:00-16:00

Robust Fuzzy Control for Permanent Magnet Fuzzy Neural Networks Pattern Recognition Synchronous Motor Chaotic Systems with Method and its Application in Ultrasonic Detection Uncertain Parameters for Bonding Defect of Thin Composite materials Yan Zhao1, Tieyan Zhang1, Qiuye Sun2, Tianlong Zhang1 1. Department of Automatic Control Engineering, Shenyang Institute of Engineering Xu Yan-hong, Zhang Ze, Liu Kun and Zhang Guan-ying Shenyang, Liaoning, China College of electronic information engineering, 2. School of Information Science and Engineering, Northeastern University Inner Mongolia University, Hohhot, China Shenyang, Liaoning, China • New fuzzy neural network pattern recognition method established. • The T–S fuzzy models of a class of • Combination of fuzzy pattern PMSM chaotic systems are derived. recognition and neural network • Based on the T-S fuzzy PMSM pattern recognition. chaotic systems, a simple fuzzy • Realization of quantification recognition and quantification stabilization controller is designed. detection. • The robust stability criterion is presented. The PMSM Chaotic Attractor The fuzzy neural network system

P-24 Friday August 7, 2009 Afternoon

FPP Transportation system II

FPP Management systems on logistics

FPP System Modeling II

FPP System Modeling III

FPP System Modeling IV

FPP System Modeling V

FPP Modeling and simulation I

FPP Modeling and simulation II

FPP Modeling and simulation III

FPP Modeling and simulation IV

FPP Modeling and simulation V

FPP Modeling and simulation VI

FPP Modeling and simulation VII

FPP:Transportation system II

Session Chair: Jason Gu and Simon Yang

HQHT 15:40-16:40 Friday, August 7th

HQHT Hallway (1) 15:40-16:40 HQHT Hallway (2) 15:40-16:40 Overlapping Clustering Routing Algorithm Analysis to Detecting Methods of Analogue Based on L-PLC Meter Reading System Based on Variable Threshold Value Neuron Zhenchao Wang, YiJin Wang and Jing Wang Jian Tang and Bo Sun College of Electronic and Information Engineering HeBei University Institute of Special Equipment , Shenyang Ligong University BaoDing, China Shenyang, China

• The algorithm is based on • Detecting nonlinear analogue, anyone detection value relates to overlapping clustering, so it can three neighbouring samples at most. So the algorithm saves time. establish multiple shortest routings. The measuring error based on piecewise variable slope training • A new rule is given that a node can method is less than that based on piecewise linearization ones. respond to clustering command • Variable threshold neuron can not only detect monotonic non- from all nodes but its sub-nodes. linear analogue but also for general non-linear analogue. The two • In order to record information of all training methods to threshold and weight are cross reference on paths, we build a routing table as operating conditions. The piecewise variable slope training method matrix, and acquire paths through to threshold and weight is much superior to piecewise linearization searching the routing table. The cluster logical structure method in the area of analogue detection.

HQHT Hallway (3) 15:40-16:40 HQHT Hallway (4) 15:40-16:40 A Novel Algorithm of License Plates Automatic Research on Road Express Freight Transportation Location Based on Texture Feature Network layout in Shandong Province Zhen-Xue Chen, Cheng-Yun Liu, Fa-Liang Chang and Jian-Guang Xu Zhang Yuan, Ma Xiang-juan, and Wu Wei-Yang School of Control Science & Engineering, Shandong University Department of Transport and Logistics, Shandong Jiaotong University Jinan, China Jinan, China • Locating the vehicle license plate plays an important • Analyze of Shandong Traffic role in the vehicle license plate automatic recognition Location Lines: two horizontal system. lines and four vertical lines. • A novel locating approach based on the texture feature • Three categories: the most is presented in this paper. important nodes, the more

• This paper makes one dimension cycle clear method for important nodes and other legend  important nodes ● more important nodes map of horizontal projection. nodes. ▲ other nodes • Put forward the Shandong • This method can effectively filter the noise and clutter Shandong Express Freight around car license plates, and accurately locate them. express freight transportation Transportation Node Map network layout.

P-25 FPP:Management systems on logistics

Session Chair: Jason Gu and Simon Yang

HQHT 15:40-16:40 Friday, August 7th

HQHT Hallway (5) 15:40-16:40 HQHT Hallway (6) 15:40-16:40 Exponential Stability of a Class of Impulsive Apparel Enterprise Application Integration Model Neural Networks with Variable Delays Based on Service-Oriented Architecture Zhang Jujian Jianfu Yang1 , Fengjian Yang1 ,Jicheng Tao2, Wei Li1 , Dongqing Wu1 Computer Information Center, Beijing Fashion Institute 1 Department of Computation Science Zhongkai University of Agriculture and Beijing, China Engineering Guangzhou, P. R China.

2 Department of Mathematics, China Jiliang University,Hangzhou, P. R China. • This paper sets up an entire frame of Flow services development platform

apparel enterprise application Management Services(System Management and Monitoring) Services Plate form Distribution Flow Enterprise Portal Portal Enterprise FlowProduction integration based on SOA Flow Purchase Distribution Information Information Order Flow Flow Order Integrated Integrated Services Services Services Services Services The main purpose of this paper is to study the globally Services • Describes the designing of the exponential stability of the equilibrium point for a class of structure model of apparel enterprise application integration impulsive neural networks with time-varying delays. Enterprise Services Bus system Services Access Services Distribution Modeling Warehouse Magement System Finance Supplier ManagementSupplier League Services shop • As an example of apparel production, Head Office Services Distribution PSS System Without assuming global Lipschitz conditions on the Application exclusive agency Services Services CRM Services CRM Transportation Other Services Services Services Services inventory, sales business activities, services activation functions, applying idea of vector Lyapunov it also discusses the methods of function, combining Halanay differential inequality with Application and Data realizing application integration. Access services delay, the sufficient conditions for globally exponential Basic Services and LDAP stability of neural networks are obtained. EAI frame based on SOA

HQHT Hallway (7) 15:40-16:40 HQHT Hallway (8) 15:40-16:40 Design of Lifting Permanent Magnet Based on Preprocessing to Weld Line Based on Roller Forming for Tailored Blank Laser Welding Neural Network Ning Ding1,2, Yunpeng Wang2, Shiwu Li2, and Dingtong Zhang System 1Mechanical Engineering College, Changchun University, Jilin, China Junyi Wang, Zhigang Xu, Limin Xin, Feng Li 2Transportation and Traffic College, Jilin University, Jilin, China Shenyang Institute of Automation,Chinese Academy of Science • The Rare Earth Lifting Permanent Magnet Lifting Permanent zreduce the width of the gap is presented, which can completely Magnet Structure between the steel sheets. eliminate the disadvantages found in Artificial Neural conventional lifting electromagnets. Finite Element Network model zroller forming process by Modeling plastically deforming the • A neural network model of magnet circuit thicker sheet. design is developed. zMathematical model of roller • A drive system is invented, which can Magnetic flux forming is founded by make Rare Earth Lifting Permanent Magnet load and unload the ferromagnetic experiment and function fitting Structure of the Tailored blank laser welding system object automatically without supplying any electricity at all. proposed method

HQHT Hallway (9) 15:40-16:40 HQHT Hallway (10) 15:40-16:40 Analysis and evaluation between Maritime industry and the Analysis of Correlation between Shandong’s socio-economic relationship in Liaoning Province Transport and GDP Jing Liang, and Zhanwei Zhao Qian Wang , Liting Ren Transportation Management College, Dalian Maritime University, Dalian, China • The current Maritime industry and the School of Economics & Management ,Yantai University regional economy development status are summarized and commented on. G = 59717Ln(L) - 557011 • Granger causality between shipping • Introduction R2 = 0.8438 service and the three significant 250000 • Time series analysis of total 200000 variables of social economic develop- G 150000 LM Test White Test transport volume ment is carried out. Model 100000 PF Pchi PF Pchi • The econometric model is established • Correlation between transport 50000 Model（4） 0.1117 0.0721 0.1746 0.1520 according to the result of Granger 0 Model（5） 0.8167 0.7873 0.5611 0.5099 volume and Shandong’s GDP 0 20000 40000 60000 80000 100000 120000 140000 -50000 causality test to analyze the impact on （ ） L the social economic relations imposed Model 6 0.1344 0.1085 0.9211 0.9065 • Conclusions by the shipping service. Correlation function of freight • Constructive suggestions upon the the results of relativity and shipping service and related industry heteroscedasticity test volume and passenger volume are put forward.

P-26 FPP:System Modeling II

Session Chair: Jason Gu and Simon Yang

HQHT 15:40-16:40 Friday, August 7th

HQHT Hallway (11) 15:40-16:40 HQHT Hallway (12) 15:40-16:40 Propagation of Stress Wave in Nickel Single Propagation of Stress Wave in Nickel Single Crystals Nanofilm with Cavity via Molecular Crystals Nanofilm via Molecular Dynamics Dynamics Jia yan, Liu heng and Yu lie Jia yan, Liu heng and Yu lie Institute of Mechatronics and Information Systems, Xi’an Jiaotong University, China Institute of Mechatronics and Information Systems, Xi’an Jiaotong University, China • Molecular dynamics is applied to • Molecular dynamics is applied to free end study the propagation of stress wave free end study the propagation of stress wave in metal single crystal nano-film in metal single crystal nano-film • The simulation result shows the stress without and with cavity transfer in the nano-film is according • The simulation result shows the to the way of wave, and there is time properties of the propagation of stress lag in the progress of stress wave wave in metal single crystal nano- fixed end propagation. The wave speed is y fixed end film without and with cavity, and the y difference as the lattice direction is 0 x effect of cavity on the propagation of 0 x difference. z stress wave z Schematic of the simulation system Schematic of the simulation system

HQHT Hallway (13) 15:40-16:40 HQHT Hallway (14) 15:40-16:40 Chirping Noise of Laser Diodes under Small Remote vs. Simulated, Virtual or Real-Time Signal Modulation Automation Laboratory Dorin Popescu1, Razvan Aguridan1, Cosmin Ionete1, Livia Popescu1, Qinghao Meng2, Adina Ionete3 ZANG Huai-quan , DUAN Hui Key Lab of Industrial Computer Control Engineering Yanshan University 1-University of Craiova, Romania, 2-Tianjin University, China, 3-Elprest SRL, Romania QinHuangdao, He Bei Province , China • This work deals with virtual, remote, • Direct intensity modulation of DFB laser real-time and simulated experiments diodes brings about chirping, which affects for a workstation (FMS-200) the performance of optical transmission systems. • Aims: to train the students or engineers in the field of automation; to offer a • Chirping characteristics of DFB laser technology for industrial developments diodes under small signal modulation are studied. • The purpose of the remote experiment is to give the users access in two • The qualitative comparison between the situations: frequency chirping of laser and IM - users enrolled in some form of response is made. Red Light Laser Diode distance learning; Real body • The theoretical relationship of the ratio - laboratory experiments on expensive between the chirping and modulation power or difficult to install equipment. feed-positioning station has been derived.

HQHT Hallway (15) 15:40-16:40 HQHT Hallway (16) 15:40-16:40 The Research of Building Logistics Cost Research and Realization of Simulation Control Forecast Based on Radial Basic Probabilistic System for Rudder and Ship with Autopilot Neural Network Hairong Xiao1,Yaozhen Han1, Changshun Wang1 and Fengyu Zhou2 1.Department of Information Engineering , Shandong Jiaotong University, Jinan, China Meijuan Gao, Jingwen Tian, and Shiru Zhou 2. School of Control Science and Engineering, Shandong University, Jinan, China College of Automation., Beijing Union University Beijing, China • Various math models are Course & track rudder autopilot established during the process of Physical simulation • Introduction ship movements. x1 Feedback signal Feedback signal Feedback signal y1 simulation of simulation of simulation of • Radial Basic Probabilistic Neural Σ • The design project and realization rudder angle rudder course ship location x Network (RBPNN) 2 approach to simulate ship - rudder control system with course & track Calculation of Calculation of Calculation of M M M rudder angle rudder course ship location • Building Logistics Cost Affect M autopilot in real navigation Rudder angle Rudder course Ship location x y n n environment are introduced which motion model motion model motion model Factors Σ combine digital simulation with • Building Logistics Cost Modeling physical simulation. Control Computer Mathematical simulation Structure of radial basic Signal acquisition Based on RBPNN • The main control software and circuit of autopilot probabilistic neural network single-chip microcomputer system • Conclusion used for the actual ship simulation The framework of system are developed. simulation system

P-27 FPP:System Modeling III

Session Chair: Jason Gu and Simon Yang

HQHT 15:40-16:40 Friday, August 7th

HQHT Hallway (17) 15:40-16:40 HQHT Hallway (18) 15:40-16:40 Globally Exponential Stability of Cellular The System Identification for Multivariable Neural Networks with Distributed Delays System in Frequency Domain and Large Impulses ZHAO Yong-guo, CAI Wen-jian, and JIA Lei Institute of Automation, Shandong Academy of Science Jianfu Yang1 , Fengjian Yang1, Jicheng Tao2, Wei Li1 , Dongqing Wu1 Jinan, China 1 Department of Computation Science Zhongkai University of Agriculture and g11 g12 Engineering Guangzhou, P. R China. • Step changes in the controller set 0 15 2 Department of Mathematics, China Jiliang University,Hangzhou, P. R China. point are performed on the -2 10 -4 5

process. -6 0 The main purpose of this paper is to study the globally -8 -5 • The process frequency-response -5 0 5 10 15 -20 -10 0 10 exponential stability of the equilibrium point for a class of g21 g22 matrix is estimated An accurate 2 15 cellular neural networks with distributed delays and large 0 10 impulses. With assuming global Lipschitz conditions on the transform-function matrix can be -2 5 identified . -4 0 activation functions, applying idea of vector Lyapunov -6 -5 -5 0 5 10 -20 -10 0 10 function, combining Halanay differential inequality with delay, • It only requires response data of the sufficient conditions for globally exponential stability of process input and output, and no Nyquist plots of real process neural networks are obtained. prior knowledge of the process and estimated frequency dynamics is needed. responses

HQHT Hallway (19) 15:40-16:40 HQHT Hallway (20) 15:40-16:40 Function Modeling and Simulation of Outer Function Modeling and Simulation of Characteristic of Telescopic Shock Absorber for Deformation of Multi-Throttle-Slices for Automobile Telescopic Shock Absorber Changcheng Zhou, Yuanyi Liu Changcheng Zhou, Leilei Zhao, and Yuanyi Liu School of Transport and Vehicle Engineering, Shandong University of Technology School of Transport and Vehicle Engineering, Shandong University of Technology Zibo, China Zibo, China

• Analysis of damping components. • Mathematic model of deformation • Velocity points at which the valves of single throttle-slice. open firstly and maximally. • Formula of slice deformation at any p radius fG rr = 3 . • mathematic modeling of shock h absorber outer characteristic. • Equivalent thickness of multi-slices hnhnhnh=+++3 33... 3 • Simulation of outer characteristic of en11 2 2 n shock absorber. • Deformation of multi-slices • Performance test. Indicator diagram simulated • Simulation certification Deformation curves of multi-slices

HQHT Hallway (21) 15:40-16:40 HQHT Hallway (22) 15:40-16:40 A Practical Maximum Power Point Tracker for Design on Radar Signal Simulator of the Photovoltaic System Automotive ACC Jiying shi, Chunling Li, Lili Zhang, and Chengshan Wang School of electrical and automation Engineering, Si Hai, Zhan WenZhang, Liu Zhao-du, He Wei and LI Jingliang Tianjin University Beijing Automotive Technology Center, Beijing Institute of Technology Tianjin, China • A practical technique combining a • The valid target data can be outputted from loss-free resister with an incremental radar simulator. conductance tracking algorithm to • The effect of ACC control algorithm is well efficiently track the maximum validated after revising the wheel velocity power point of a solar array has been measurement and improving integral presented. algorithm and can be applied to ACC • The proposed tracker introduces a control. variable searching steps strategy Power (W) based on feature of the converter • The radar signals can meet the ACC and the tracking method. controller and accordingly the safety and efficiency of the exploitation of the ACC Time (s) • Both of the results of the analysis system and the real vehicle experiment can and simulation manifest that the The output power be ensured. The host vehicle velocity input conversion efficiency is improved obviously.

P-28 FPP:System Modeling IV

Session Chair: Jason Gu and Simon Yang

HQHT 15:40-16:40 Friday, August 7th

HQHT Hallway (23) 15:40-16:40 HQHT Hallway (24) 15:40-16:40 Boxed Milk Metamorphism Detecting Method Identification of PMSM Based on EKF Base on wavelet and Hilbert Transform and Elman Neural Network Xianjiang Shi, Xinjie Zhao, and Guanghui Xiao Wang Song1,2,Shi Shuang-shuang2 ,Chen Chao3,Yang Gang1, and Qu Zhi-jian1 Department of Mechanical and Engineering, Harbin University of Science and Technology 1.School of Electrical Engineering, Beijing Jiaotong University ,Beijing, China Harbin, China 2.School of Mechanical&Electrical Engineering, Shandong University at Weihai, Weihai,China Liquid milk box 3.Institute of Information Engineering Technology,Academy of Navy Equipment, Beijing,China

• Identifying parameters R s , ψ d and ψ q • Wavelet de-noising method is used Spring of PMSM is very important. to eliminate signal noise. USB2015 Pin Computer USB • Use modified EKF to estimate • Hilbert transform is used to achieve Interface motor parameters and supply CLK signal envelope curve and the 1 Angle OUT Sensor 1 Step training data for Elman NN. system damping ratio. . Motor • Construct a Elman NN ,through • The results showed the method GATE0 Driver greatly improved system damping OUT0 training and simulating ,obtain the ratio of detection accuracy. A/D CH0+ Amplifer identified values of parameters. The detecting device • Validate the generation ability of Comparison of Identificated our Elman NN with another case. and Estimated Parameters

HQHT Hallway (25) 15:40-16:40 HQHT Hallway (26) 15:40-16:40 Calculation Force Control Based on TBM Motion Simulation based on VR System Linearization Feedback Model of Jianyu Yang, Hu Li, Jun Cheng, and Wanshan Wang School of Mechanical Engineering and Automation, Northeastern University SMISMO Control System Shenyang, China Yingjie Liu, Bing Xu, Huayong Yang, and Dingrong Zeng • A practical method for numerical State key laboratory of fluid power transmission and control, Zhejiang University prototype motion simulation based on Hangzhou, China 10 9 OpenGL Performer was proposed.

AB • A main simulation procedure was • Simulation model of SMISMO 8 P T designed for processing event trigger, control system. P V 7 and a motion simulation of TBM key Program Running Instance P ABV D component was developed . • Proposed control strategy for V trajectory control of piston in the PT • A motion analyzing method was V P expressed, based on the TBM erector, cylinder. Controller and a realizing method based on • Controller is tested by simulation M OpenGL Performer toolkit was and experiment, good results is expatiated. achieved, it is promising for 1 2345 6 • Introduced a model transforming and industrial application . Schematic of SMISMO system processing method with Deep Exploration and Multigen Creator. Hardware Components

HQHT Hallway (27) 15:40-16:40 HQHT Hallway (28) 15:40-16:40 Design of BPNN Speed PI Controller based on GA Research on Machining Simulation of Turn- Yinquan Hu, Heping Liu and Wencui Yi State Key Laboratory of Power Transmission Equipment & System Security and New milling Center based on DVR Technology, Chongqing University Weili Liang, Tianyu You, Yan Yu, Lida Zhu, Wanshan Wang Shapingba District Chongqing,China School of Mechanical Engineering and Automation, Northeastern University Shenyang, China Following the performance and disturbance rejection performance of motor is directly related to the application of DC speed control system in DC motor control. DC speed control system uses regular speed, current double closed-loop speed control, the inner ring and the outer ring generally use PI regulator. However, the rotational speed will produce inevitably overshoot and disturbance rejection performance is not good in double-loop speed control system. This paper describes design method of a BP neural network speed PI controller based on genetic algorithm, which eliminate the DC speed control system rotational speed of the overshoot, and improve the system dynamic performance, and use Matlab simulation software to verify its advantages. Turn-milling center Milling “NEU” on a cylindrical surface

P-29 FPP:System Modeling V

Session Chair: Jason Gu and Simon Yang

HQHT 15:40-16:40 Friday, August 7th

HQHT Hallway (29) 15:40-16:40 HQHT Hallway (30) 15:40-16:40 Pattern recognition of chatter gestation based DTC Scheme Based-on SVM with Sliding-Mode on SVM - HMM method Observer for Induction Motors Shao Qiang ,Guan LiNa and Shao Cheng Xianqing Cao，Liping Fan Institute of Advanced control technology , Dalian university of Technology College of Information Engineering ,Shenyang Institute of Chemical Technology Dalian, Liaoning Province , China Shenyang, China Training set O • I. Introduction. This paper introduces •SVM-DTC not only preserve DTC transient merits, but also the HMM - SVM method to identification Initial HMM of nonstationary time series of the λ produce better quality steady-state performance in a wide 0 vibration signal of cutting process. Viterbi algorithm for q speed range. • II. SVM and its output probability. The t feature vector can be transformed into •The SVM-DTC system output probability by Gauss model as SVM probability for bj(k) based on input-output following: ˆ PC(|)i xx= NV ((),,)μΣ λ d − 1 Baulm-Wechl Re-estimate linearization technique for = 2π 2 Σ 2

induction machine drives 1 T exp[−− (()VVxxμ )Σ (() −μ )] N is developed. 2 Satisfaction? • III. DESIGN of identification of pattern λ and HYBRID SVM-HMM architecture Y •A sliding-mode observer Final HMM for estimating stator flux is • IV. Experiments and results. Block diagram of the DTC-SVM system • V. Conclusions. Hybrid SVM-HMM training introduced.

HQHT Hallway (31) 15:40-16:40 HQHT Hallway (32) 15:40-16:40 A Novel 5-Dimensional Indoor Localization Algorithm Based on RF Signal Strength Bond Graph Modelling of a Wastewater Longqiang Tian, Chao Hu, Lujia Wang , Dongmei Chen,Max Q.-H. Meng Biodegradation Bioprocess Chinese Academy of Sciences/The Chinese University of Hong Kong Monica Roman, Dan Selisteanu, Eugen Bobasu, Emil Petre and Dorin Sendrescu Shenzhen Institute of Advanced Integration Technology Department of Automatic Control, University of Craiova, Craiova, Romania Shenzhen, 518055, China MR • The work deals with the Bond Graph C C • This paper proposes a new localization modeling of nonlinear bioprocesses 2 6 9 1 10 Sf 0 TF 1 TF 0 R algorithm for the localization of an 4 5 7 8 • Pseudo Bond Graph modeling rules 3 MR R R object in the limited area, like indoor 11 are applied to design the model of a R 17 and in-vivo scenarios. 14 20 complex wastewater treatment process 18 21 RF signal radiation pattern TF 0 TF1 TF 0 C • The prototype of our localization model 12 15 16 19 with biomethanation 13 is based on the RF Signal Strength. 22 MSf • The procedure represents a valuable C 26 23 24 • The method can realize 5-Dimensional illustration of the power of Bond TF 0 C (including position and orientation ) Graph technique, and can be used in 25 R localization and tracking for the target order to obtain models of bioprocesses Bond Graph model of The simulation of moving with high level of complexity the bioprocess target tracking

HQHT Hallway (33) 15:40-16:40 HQHT Hallway (34) 15:40-16:40 Noise Reduction for Oil Pipeline Pressure Time Time-frequency Algorithm for Octave Analysis Series Based on Wavelet Filtering Technology of Nonstationary Interior Vehicle Noises Jinhai Liu, Xiaolan Li, Huaguang Zhang, and Derong Liu Yansong Wang, Lihua Chen, and Xuewen Chen Institute of Electric Automation, Northeastern University School of Automotive Engineering, Shanghai University of Engineering Science Shenyang, China Shanghai, China • Introduce the wavelet theory. • Use DWT, lifting WT and • A new time-frequency algorithm for UWT in denoising the sine sound octave analysis, so-called noisy signal respectively. DWT-OA method, is proposed. • Use DWT, lifting WT and • Daubechies wavelet is determined UWT in denoising the oil and applied for OA of nonstationary pipeline pressure time series vehicle noises. with noise. • The DWT-OA algorithm is accurate • Signal denoising methods for octave analysis of vehicle noises. based on lifting WT and UWT • The newly proposed DWT-OA are much more effective than algorithm may be extended to dealing that on DWT in locating the Noisy Signal and Denoised Signal with other types of sound signals in Using DWT and Lifting WT singular point. engineering. DWT Algorithm and SQE of Vehicle Noise

P-30 FPP:Modeling and simulation I

Session Chair: Jason Gu and Simon Yang

HQHT 15:40-16:40 Friday, August 7th

HQHT Hallway (35) 15:40-16:40 HQHT Hallway (36) 15:40-16:40 Linear Motor Control Based on Resistance Research on ramp optimization metering strategy in Identification with Wavelet Transform urban expressway Cui Jiefan, Fu Yue, Zhao Nan, Xing Feng and Yang Xia Xuewen Chen, Yansong Wang School of Electrical Engineering College of Automobile and Traffic Engineering, Liaoning University of Technology, Jinzhou, China Shenyang University of Technology College of Automobile Engineering, Shanghai University of Engineering Science, Shanghai China Shenyang,China

Start Mainline Bullet points D e te rm in e d e c o m p o sitio n Bridge scale J

~ P re p a rin g filte r () j a n d ~ ( j ) {h k } ∈ zk {g k } ∈ zk • The mathematic model of direct • Introduction. thrust control system. S ta rt sim p le : n = 1 Surface Street Intersection G iven voltage, late timing, from A /D C onversion, 0 • On-ramp metering method. gather the nth digital point ( ) (nC ) Controller

M a k in g th e n th d ig ita l p o in t(0 ) ()nC w a v e le t tra n sfo rm a tio n . T h e • The phase adjustment method.

• Wavelet transform using for ( j ) ()j nD j = ,1 J re su lts a re o b ta in e d () nC a n d () fro m d iffe re n t sc a le J ( L ) • The congestion dissipation resistance identification. A n a ly sis re su lts iSection − 1 iSection iSection +1 control. Y C o n tin u a tio n nn += 1 S S mi −+ 1 sim ple S i−1 Si qi +1 mi −+ 2

N • Wavelet analysis used in resistance S a v e re su lts q i … identification of direct thrust control End r s r s i i r mi −+ 2 mi −+ 1 mi −+ 1 di d d system. The flowchart of the mi −+ 2 mi −+ 1 wavelet transform Fig.1 Sketch map of urban expressway control system

HQHT Hallway (37) 15:40-16:40 HQHT Hallway (38) 15:40-16:40 The Implement of High Speed Data Acquisition A Novel Parameter Adaptive Nonlinear Model of System Based-on SOPC Technology Proportional Valve ZhengJun Wang, Junzheng Wang, Wei Shen, and Shoukun Wang Chuanfei Qiu, Quanzhi Zhou, Changjun Wei and Fei Xu School of Automation, Beijing Institute of Technology Wuhan Mechanical Technology College Beijing, China Wuhan, China • Comes up with the solution that • Aiming at controlling the nonlinear RS232 Keyboard LCD SDRAM Flash implementing high speed data Memory Memory coupled variables force and velocity Altera EP2S60 FPGA acquisition system with SOPC CFI precisely in the electro-hydraulic Nios II UART SDRAM Tri-state proportional loading system. CPU PIO PIO technology. Core Controller bridge • An open-loop control method based on • Embeds a Nios II CPU in FPGA, Avalon switch fabric a nonlinear model, achieves high and mounts customized user logic precision and adapts the parameters Clock PIO PIO Frequency and SPI Core FIFO FIFO generate periodic measure on Avalon switch fabric. Controller change online. • Successfully implements the high • The experimental results show that the ADC ADC Analog Serial DAC channel speed data acquisition system. Trigger signal maximum error of force and velocity • Includes functions of measuring the Analog input are less than 4.38% and 5.47%, respectively. parameters of input signal, Framework of the system • Statistically evaluating with the waveform rebuilding and drawing. experimental data proves the precise Electro-hydraulic loading system and efficient of the model. and experiment curves

HQHT Hallway (39) 15:40-16:40 HQHT Hallway (40) 15:40-16:40 Simulation of Soft Tissue Using Mass-Spring Modeling and Simulation of Loading and Model with Simulated Annealing Optimization Unloading Scheduling System Based on Agent Shaoping Xu, Xiaoping Liu , Hua Zhang and Petri Net School of Mechatronics Engineering , Nanchang University, Nanchang,China • A new tensor-mass model (TMM) Lipeng Ren1,2,Wen Wang1,WeipingFu1,Wei Yang1 and Can Li1 incorporating visco-elasticity was 1 School of Machinery and Precision Instrument Engineering, Xi’an University developed. of Technology,Xi’an,710048,P.R.China 2 CNPC LanZhou chemical company, Lanzhou,730000,P.R.China • Deformation calculated by the TMM is used as a reference to obtain the • the model established by the integrated method and its run rules are parameters for the Spring-Mass defined. Model (SMM)model. • the scheduling elements of the system are classified, and the object class relation are obtained. • We propose a simulated annealing identification method and it can tune • the APN (Agent-Petri Net) model of the scheduling system is set up based on the corresponding relations between each object sub-net the parameters automatically until the deformation of the mass-spring module and each function agent, which consists of five different Liver indentation simulation function agents. The system model is realized under JADE platform. model is close enough to the one defined by the new modified TMM.

P-31 FPP:Modeling and simulation II

Session Chair: Jason Gu and Simon Yang

HQHT 15:40-16:40 Friday, August 7th

HQHT Hallway (41) 15:40-16:40 HQHT Hallway (42) 15:40-16:40

Development of Ant Group and immunity Product Line System Modeling of the Cold-Rolled Algorithm for Multiple Round Scheduling Mill Based on the Hierarchy Colored Petri Nets in Cold Rolling Enterprise Yong Zhang,Jing Zhu Yong Zhang,Yukun Wang,Jiesheng Wang University of Science and Technology Liaoning University of Science and Technology Liaoning anshan, China anshan, China • The Petri Nets description ability with • The restraint combination coloring, what made steel information, optimization question of the equipment condition and routes clearer, cold rolling production thus simplified the Petri Nets model scheduling is typical NP-hard structure, reduced the model complexity. problem • This method also could be well used in • To establish minimum value other complicated process industries. It model of cold rolling production was shown that these designed object batch scheduling problem based classes were highly consistent with real on TSP Fig.1 Oliver30 the Best Solution of TSP production and the development Fig.2 Main Colored Petri Net In efficiency for production planning and cold-rolled Sheet Production in Colony Immune Algorithm Solution scheduling systems has been greatly improved.

HQHT Hallway (43) 15:40-16:40 HQHT Hallway (44) 15:40-16:40 Tibetan Language Continuous Speech Tool Wear Intelligence Measure in Cutting Recognition Based on Active WS-DBN Process Based on HMM Yue zhao, Yongcun Cao and Xiuqin Pan Shao Qiang ,kang Jing and Shao Cheng School of Information and Engineering , Minzu University of China , Beijing, China Institute of Advanced control technology , Dalian university of Technology Dalian, Liaoning Province , China • The research on Tibetan language speech recognition is in the initial stage, and it is time- • I. Introduction. a new method of Vibration signals acquisition consuming and costly to annotate the large vocabulary Tibetan languages corpus. tool wear intelligence measure FFT vectors extracting • Active learning can reduce annotation cost by based on DHMM is presented in sample selection. Query-by-Committee is a kind of sample selection method of active this paper. SOM vectors coding learning. In this paper, the Word-State DBN • II. tool wear intelligence measure model in recently Bilines’s speech recognition X={x ,x ,x ,…,x } research is combined with QBC for Tibetan theory based on hmm . 1 2 t T language continuous speech recognition. • III. Basic theory of HMM and λ λ … λc • The active WS-DBN algorithm for Tibetan 1 2 … language speech recognition. dhmm of tool wear & recognition P(X| λ ) P(X| λ ) P(X| λ ) • Experimental results show that active WS- method . 1 2 c DBN model can reach the same recognition rate as WS-DBN, and is higher than HMM • IV. experiment of tool wear model. At the same time DBN-based Tibetan i = max( P(X| λ ) ), ic speech recognition model has better distinguishing . i performance of noise suppression. • V. conclusion. Flowchart of wear state recognition

HQHT Hallway (45) 15:40-16:40 Extraction of Partial Discharge Signals in Power Transformer Using Complex Wavelet

Guihong Feng, Feng Ji, Bingyi Zhang

Department of Electrical Engineering, Shenyang University of Technology, China

• The generation of complex wavelet. Db series wavelet Similarity • The complex information used to Db2 0.9471 extract the feature of the partial Db3 0.2641 discharge. Db4 0.3751 Db5 0.2808 • Selection of the optimal basic Db10 2.39e-5 wavelet based the similarity degree Db20 0.7438 between wavelet and PD Similarity degree • Simulation results show the method has good performance.

P-32 FPP:Modeling and simulation III

Session Chair: Jason Gu and Simon Yang

HQHT 15:40-16:40 Friday, August 7th

HQHT Hallway (46) 15:40-16:40 HQHT Hallway (47) 15:40-16:40 A Simulating Model of Automotive Power Compare and Analysis of Passive and Active System Based on the Finite State Machine Suspensions under Random Road Excitation Hongbin Li and Dong Li Jiammin Sun2, Qingmei Yang1 Key Laboratory of Advanced Manufacturing and Automation Technology , Ludong University 1. College of Automation, Beijing Union University, Beijing, China Yantai, China 2. Beijing University of Civil Engineering and Architecture, Beijing, China • A simulating model which can

express dynamic characteristics of the z2 • LMS adaptive controller are designed in m 2 automotive power system in every two-DOF vehicle suspension model. k c u driving phase is proposed. The results show that the LMS adaptive 2 2 • Because of the driving torque and the control strategy has the better control resistant torque on each part of the result. z1 m automotive are different according • The calculation of LMS adaptive control 1 the state of the clutch and gearbox, a algorithm is much little, not only simple k simulating model is established based algorithm but also remarkably effective, 1 on the finite state machine. and the method is fit for the active z0 • Then, a hard-in-loop half-real control of the suspension system. The state simulating system is established using • The generation method of random road dSPACE, in which the simulating chart of temporal signal and the stability of LMS Suspension system of single model is working as its kernel. the engine adaptive control strategy need further wheel part study.

HQHT Hallway (48) 15:40-16:40 HQHT Hallway (49) 15:40-16:40 Study on Fractal Characteristics of Low- Modeling of Satellite Borne TDI CCD Pitching Voltage Power Line Communication Channel Imaging Image Motion Velocity Vector Liu Zhang, Shujun Li, Guang Jin and Xiubin Yang Zhenchao Wang, Yuqian Zhao, Yutao Gan and Shibing Zhang Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences College of Electronic and Information Engineering , HeBei University Changchun, China Baoding, China • Pitching imaging is necessary for • Hurst index of measured time three-dimensional observation series are greater than 0.5, fractal satellite with single TDI CCD dimension is greater than 1.0, L- • Imaging motion velocity vector PLC channel has obvious fractal computation model must be set up and chaotic characteristics. on image plane • Statistical moment is a nonlinear • TDI CCD pitching imaging function, multi-fractal spectrum is motion velocity vector model is Three-dimensional imaging strict symmetrical, the L-PLC derived by geometric mapping and coordinate transformation principle with single TDI CCD channel has strong multi-fractal Double logarithmic curve of method properties. R/S analysis

HQHT Hallway (50) 15:40-16:40 HQHT Hallway (51) 15:40-16:40 Research of VSCF Wind Power Generation Study on System Modeling and Analyzing of Training System Based on Matlab/LabVIEW Supply Chain Network Lv Yuegang, Xi Peiyu, Li Nailu and Fan Xiaoxu Luo Wuming1, Han Pingyang2, Gao Fei1 Research Laboratory for Wind Power, North China Electric Power University 1Institute of Military Transportation, 2Academy of Military Transportation, Tianjin, China Beijing, China General supply chain network • A variable-speed constant-frequency models cannot describe effectively wind power generation training the complexities of system. system is proposed. According to graph theory, the paper presents a multi-layer model of • In control layer, the simulation supply chain network. In the model, model in the simulink platform is the paper addresses nodes and proposed, which is based on the connections in the layers, and actual wind power generation unit. analyzes corresponding relationships • In supervisory layer, a vivid HMI in among the layers. On the basis of this model, we present analysis the LabVIEW platform is provided contents of the supply chain network to realize human-machine from four viewpoints which are interaction. VSCF wind power generation graph theory, three levels of supply The multi-layer model training system chain, simulation and flexibility.

P-33 FPP:Modeling and simulation IV

Session Chair: Jason Gu and Simon Yang

HQHT 15:40-16:40 Friday, August 7th

HQHT Hallway (52) 15:40-16:40 HQHT Hallway (53) 15:40-16:40 Comprehensive Fault Evaluation on Maglev Object Recognition Algorithm Research Based Train Basedon Ensemble Learning Algorithm on Variable Illumination Zhiqiang Long, Lianchun Wang and Ying Cai 1 1,2 2 College of Mechaeronics and Engineering and Automation REN Honge , ZHONG Qiubo and KANG Junfeng Information and Computer Engineering College, Northeast Forestry University National University of Defense Technology Changsha Hunan Province, China School of Computer Science and Technology, Harbin Institute of Technology • First, the structure of the suspension Harbin, China Single Testing Correct Ensemble Correct system of maglev train is analyzed and a classifier mode rate classifier rate fault diagnosis model is built. 90.67 T1 AdaBoost 94.33% • HSV color space. • Second, ensemble learning is introduced C4.5Tree % to the train model with learning ability. M1 (1) T2 88% 97% • Build a Sparse Classifier . • Last, this method is applied to fault evaluation on maglev train suspension SMO T1 92% AdaBoost 92.67% • Using an improved look-up table system. (SVM) T2 90% M1 (2) 98% T1 94% AdaBoost 93% algorithm, only need one AND • In comparison to single and ensemble BP classification method, the emulational T2 96% M1 (3) 98% operation. results prove that the ensemble method Naïve T1 87% AdaBoost 94.67% • Using the circular geometrical works better on the problem, the Bayes M1 (4) advantage of the Ensemble Learning T2 84% 94% characteristic to implement fast T1 92.33% 92% The schematic diagram algorithm is manifested. Logistic SMO detection. • Practice has proved that this method is T2 91% 90% of circular detection competent for precision demand. Comparison of single and ensemble classifier correct rate

HQHT Hallway (54) 15:40-16:40 HQHT Hallway (55) 15:40-16:40

Studies on Risk Management of the Urban Infrastructure Analysis of Three-people Seat Underframe in Projects Based on the PPP Financing Model Motor Train Unit Based on FEA and Experiment 1,2 1, 2 3 Hanli Chen , chaoQun Ma Bo Liu , Tao Qin Dong Zhou, Hu Huang, Xintian Liu, Lihui Zhao, Wenjun Chen, Xinjie Yuan 1College of Business Administration, Hunan University,Changsha, Hunan Province, China ,[email protected] College of Automobile Engineering, Shanghai University of Science Engineering Department of Manufacture, Shanghai Jiao Yun Accuracy Pressing CO. Ltd 1、 With the acceleration of urbanized advancement in China, governments‘ high investment and low Shanghai, China efficiency in the urban infrastructure projects have already become a worldwide problem. The PPP financing model suitable for urban infrastructure projects for theirs unique advantages. • The structure strength of three- 2、The PPP financing model in urban infrastructure projects have greater risks, so their risk management is of great significance. people seat underframe is analyzed 3、The paper establishes the VAR (Value at Risk) into financing risk assessment of the urban infrastructure and the reasonableness of FEA projects, and make quantitative risk analysis to the risks; then it applies the method of expert investigation and mark and the fuzzy comprehensive evaluation to setting up the risk assessing model that is based on the model is verified by the experiment. VAR. 4、 we establish the risk allocation model of urban infrastructure projects. • Two load methods are used in the 5、 we carry empirical analysis about the risk problem of an urban infrastructure project. FEA model: U-shaped and gasket- shaped. FEA Model of Three-people TABLE I RISK STATE • The result of FEA is fairly closed to Seat Underframe in Motor Train risk states lower low medium high higher that of experiment, which embodies R =VAR /value （0,0.2） [0.2,0.4） [0.4,0.6） [0.6,0.8） >0.8 ij ij good uniformity.

HQHT Hallway (56) 15:40-16:40 HQHT Hallway (57) 15:40-16:40

High-Gain Observers for Estimation of Kinetics Study of L Fractal Description Method of in Biological Sequencing Batch Reactors Natural Scenery Based on Genetic Dan Selisteanu, Emil Petre, Dorin Sendrescu, Monica Roman, and Dorin Popescu Department of Automatic Control, University of Craiova, Craiova, Romania Algorithm Xuezhen Cheng, Zhengfeng Sun, Zengshun Zhao, Maoyong Cao • Sequencing Batch Reactors (SBRs) -1 40 (h ) College of Information and Electrical Engineering, Shandong University of Science are used for the aerobic treatment of 30 μˆ 1 and Technology, Qingdao 266510, China. wastewater by activated sludge. 20 • A key issue for SBRs is the 10 • Put forward a kind of fractal L system μ1 0 description method based on the genetic estimation of kinetic rates. 0 0.5 1 1.5 2 2.5 Time (h) 3 algorithm for depicting natural scenery in this (h-1) 6 article. • High-gain observers for the kinetics 4 • The method can generate various kinds of shapes of the natural scenery by choosing to add in two types of SBRs are designed: μ2 2 μˆ 2 enhancement operator, crossover operator, - the treatment of wastewater from mutation operator, and insert operator etc. 0 chemical industry (toxic compounds), 0 0.5 1 1.5 2 2.5 Time (h) 3 • The experimental results indicate that the method has many obvious characteristics such as - the treatment of typical municipal Estimations of specific growth high fidelity, strong sensitivity, distinctly wastewater, polluted with organic rates (using noisy data) hierarchical and strongly stereoscopic effect. carbon and nitrogen. Fractal trees Created by method

P-34 FPP:Modeling and simulation V

Session Chair: Jason Gu and Simon Yang

HQHT 15:40-16:40 Friday, August 7th

HQHT Hallway (58) 15:40-16:40 HQHT Hallway (59) 15:40-16:40 Research on Precise Locating Active Control of the Pressure Pulsation in Control of Large Inertia Ball Mill Large-flow Mixed Gas Pipe Yudong Xie, Yanjun Liu(corresponding author) and Yong Wang Changsheng Ai, Weixing Jing, Hui Zhang, Xiangbo Ze, Fang Zhao School of Mechanical Engineering, Shandong University School of Mechanical Engineering, University of Jinan, Jinan ,Shandong provence, China Jinan, China • At the process of ball mill starting, • An active control method for the the critical parameters such as rotary attenuation of the pressure pulsation inertia and load torque can be gotten. in large-flow pipe was presented. • With the relationship between • A series of active control valves and starting and stopping, the rotary error sensors were located along the inertia and load torque and other mixed-gas pipe. inherent parameters can also be • The results shown in Fig.1 indicate acquired. that the secondary pressure • With the inherent parameters at the pulsations produced by the active process of stopping of ball mill, the The Ball Mill control valves can counteract the optimal position for power off can primary pressure pulsations of the be predicted. mixed gas effectively.

HQHT Hallway (60) 15:40-16:40 HQHT Hallway (61) 15:40-16:40 A Novel Fault Diagnosis Method Design and AQM Controller Design for TCP Vegas Networks Application for Civil Aircraft System Based on Sliding Mode Control Xiaojuan Zhang1,2 , Gang Liu2 Yongqi Chen, Zhiguang Zhong Institute of Optoelectronics Engineering,Changchun University of Science and Technology, Intelligence Science and Electromechanical Systems Laboratory, CST Changchun , China ; Department of Automation,Jilin Technology college of Electronic Ningbo University, Ningbo, China Information, Jilin , China

For estimating faults of the system with unknown nonlinear term, • A TCP Vegas model and the control theoretical based a novel fault diagnosis method based on nonlinear compensation AQM controller have been researched. term and proportional multiple-integral observer is proposed. In • The model is linearized to derive a state space this method, the nonlinear compensation term is constructed by support vector machines(SVM), which can reduce the influence representation. of unknown nonlinear part. Proportional multiple-integral • The AQM controller is designed based on sliding observer based nonlinear compensation term can estimate actuator mode control. faults when there are sensor faults or output disturbance in the system. The proposed method is applied in civil aircraft system. • And the controller is robust to model uncertainties Simulation experiments are given to demonstrate the efficiency. and disturbances.

HQHT Hallway (62) 15:40-16:40 HQHT Hallway (63) 15:40-16:40 Logistics Standard Systems Based on the Research on Performance for UWB MAC Architecture of Supply Chain Logistics Systems Protocol in the FCS Application Huishu Piao, Bin Li and Chiyu Wang Jun Wang, Yujun Pang and Weiru Chen Transportation Management College, Dalian Maritime University Shenyang Institute of Chemical Technology Dalian, China Liaoning, China • Level dimension mainly reflects the Chinese logistics standard systems structure management systems of supply chain • The advances of UWB are that high General basic standards Technology standards Information standards Management standards Serv ice standards for logistics for logistics for logistics for logistics for logistics data rates, low power consumption, logistics systems, including strategic, Packet Through average Slot precise ranging capabilities, resistance Generation put delay utilization Basic standards for logistics Basic standards in logistics

tactical and operational level. nal rofessio p for ndards Sta M anagement standards in in standards anagement M easurement unitM standards Rate (Mpbs/se (secs) Sta ndards for inte grated grated inte for ndards Sta T echnology standards in in standards echnology T Sta ndards for lo gistics gistics lo for ndards Sta Service standa rds for for rds standa Service oitc nomai n atio inform logistics oitc nomai n atio inform logistics Terminology standards inform atio n service n atio inform

logistics activities activities logistics (pkts/sec) c) logistics service logistics service to multi-path fading, and penetration of M odulus standards odulus M

• The standard systems framework from …… n atio inform …… … service dense objects. 10 0.41 129 1.89 the system architecture theory will be 20 0.82 159 1.85 helpful to unify and develop the global • Owing to a lot of short packets in the 30 1.18 269 1.93 logistics standards and its systems. FCS applications and high channel 40 1.49 410 1.91 acquisition time in UWB, the delay is 50 1.73 502 1.90

• The effective operating of supply chain technique and gy echnolo T

Equipment andmachine 60 1.89 561 1.87 Performance evaluation standards in logistics standards in logistics Statistics standa rds for for rds standa Statistics Sta ndards for lo gistics gistics lo for ndards Sta Sta ndards for lo gistics gistics lo for ndards Sta tnad o ogistics lo for standards oitc aae ent managem logistics

Basic standards for larger than in other application.

logistics systems is based on scientific pro-environment 70 1.99 618 1.92 activities …… …… logistics security logistics standardization activities. • This paper studies the slot utilization of 80 2.08 711 1.95 • The logistics standard systems UWB MAC Protocol in the different 90 2.14 856 1.91 framework from theory comparing to FCS Application. Finally, we obtain 100 2.18 1175 1.89 the Chinese logistics standard systems quantitative data about the packets from practices. logistics standards systems aggregating mechanism in FCS Simulation Results application by simulation.

P-35 FPP:Modeling and simulation VI

Session Chair: Jason Gu and Simon Yang

HQHT 15:40-16:40 Friday, August 7th

HQHT Hallway (64) 15:40-16:40 HQHT Hallway (65) 15:40-16:40 Fidelity Management and Evaluation Model and Simulation of Doubly Feed in Logistics System Induction Generator Wind Turbine Fu Yu, Liu Yukun, Su Zhiyuan and Zhou Xiaoguang Xing-jia Yao,Lei Tian ,Zuo-xia Xing and Xian-bin Su Automation School, Beijing University of Posts and Telecommunications Wind Energy Institute , Shenyang Universityof Technology Beijing, China Shenyang, China • A method to manage fidelity is • The model of DFIG and a developed, and three standards converter system are proposed. according to which fidelity levels can • The vector controal method be described are presented. oriented by the stator flux is used • A Book Retrieval Module in a Book in rotor side converter ofthe Distribution Center Design Project is generator . studied, three models with different • The transient charateristics of fidelity levels are built, and a method DFIG when the power system's to calculate the differences is given. bus voltage drop 20% and 50% • A system of evaluation index is are simulated respectively in Frame of Three Standards for Simulation Fidelity developed. Some suggestions about PSCAD. fidelity management and evaluation • The DFIG used vector control SUT 1.5MW Wind Turbine are given. can improve the power system fault ride through capacity.

HQHT Hallway (66) 15:40-16:40 HQHT Hallway (67) 15:40-16:40 Numerical Simulation of Deformation and Modeling of Logistics Distribution System Microstructures Evolution for Hexagon Hollow Including Reverse Logistics Based on Logical Steels Rolling Petri Net Zhao Yuqian, Yu Enlin, and Yan Tao QinYing College of Mechanical Engineering, Yanshan University Vocational Technical College, Shanghai University of Engineering Science Northeastern University at Qinhuangdao Shanghai, China Qinhuandao, China • Logical Petri Net • Simulated the forming process of hexagon hollow steels by 3-D rigid-plastic FEM. • Description of workflow in • The strain field, stress field, temperature field common distribution centers. were calculated in given craft parameters. • Logical Petri net model of the • Established the microstructure evolution distribution system. model for 35SiMnMoV on high temperature • Nature of the Petri net model. deformation. • The variation of average grain size with rolling velocity was obtained in this paper. The picture of average grain size Logical Petri net model

HQHT Hallway (68) 15:40-16:40

The Extraction of Brain Evoked Potential Based on Fuzzy Clustering and Wavelet Transformation Lanlan Yu, Tianxing Meng, Baina He School of Electric and Electronic Engineering, Shandong University of Technology Zibo, China ƒ Meaning and actuality of BEP extraction. ƒ Principle of wavelet transformation. ƒ Principle of fuzzy clustering. ƒ Experiment work and results analysis of BEP extraction using wavelet transformation and fuzzy clustering. ƒ Conclusion and expectation. The extraction result

P-36 FPP:Modeling and simulation VII

Session Chair: Jason Gu and Simon Yang

HQHT 15:40-16:40 Friday, August 7th

HQHT Hallway (69) 15:40-16:40 HQHT Hallway (70) 15:40-16:40 On-line Detecting of Rotor Based The Automatic Temperature System on Wavelet Digital Filter and Instantaneous PowerBar Breaking With Fuzzy Self-adaptive PID control Chen Yong Bai Xiaoping School of Automation Engineering,University of Electronic Science and Technology of In Semiconductor Laser China,Chengdu,China ZANG Huai-quan , LI Qian Key Lab of Industrial Computer Control Engineering Yanshan University • Firstly,this paper gives the definition to featrue instantaneous power by blending the QinHuangdao, He Bei Province , China instantaneous power with the digital notch • Design fuzzy self-adaptive PID control filter. temperature control system to control the • And then,this paper proposes a novel on- line detecting of rotor bar breaking Fig.3.Spectrum of the instantaneous power(s=0.03,RBB) laser temperature approach based on spectrum analysis of • The sensor transmitted temperature to high- feature instantaneous power. precision amplifier by through temperature, • Finally,the analysis of theory and simulation results demonstrate that this then signal into the controller, through novel method can get rid of the interference semiconductor refrigeration controlled temperature system to the fault related characteristic temperature components from the constant weight,and in Laser thus it keeps high reliability even when the • To Adjust effect of temperature system in slip is very small. real test, select the temperature control Fig.6.Spectrum of featrue instantaneous power(s=0.01,RBB) curve at 16℃ for example.

HQHT Hallway (71) 15:40-16:40 HQHT Hallway (72) 15:40-16:40 Satellite Navigation Observation Equations and Using IACO and QPSO to Solve Spatial Clustering theirs Solution Algorithm with Obstacles Constraints Sun Xigang, Ji Yuanfa Xueping Zhang, Taogai Zhang,and Hongmei Zhang Applied Science and Technology College, Guilin University of Electronic Technology School of Computer Science & Engineering , Henan University of Technology y???????????????( ??: ?) , Guilin, China x 10-3 x ????( ??: ?) 1.6 Zhengzhou, China • The GPS positioning accuracy sometimes is poor for the big GDOP. 1.4 • This paper proposed an Improved • Two sets of measurement equations 1.2 Ant Colony Optimization (IACO)

were put forward in this paper: 1 and Quantum Particle Swarm measurement equations subject to Optimization (QPSO) method for 0.8 height constraint; hyperboloid Spatial Clustering with Obstacles measurement equations subject to 0.6 Constraints (SCOC). height constraint; 0.4 • Kalman filtering method was designed 0 20 40 60 80 100 120 140 160 • We first use IACO to obtain the for solving these two different shortest obstructed distance. equations. Positioning results • And then we develop a novel • Finally, these two methods were proved comparison with QPKSCOC based on QPSO and K- fast convergence and high precision reference point in Medoids to cluster spatial data with Clustering by IACO and QPSO through simulation tests. longitude direction obstacles.

HQHT Hallway (73) 15:40-16:40 An Overview of Multi-Modal Biometrics Based on Face and Ear Haijun Zhang, Zengxi Huang, Yibo Li, and Zhichun Mu Colledge of Automation, Shenyang Institute of Aeronautical Engineering Shenyang, China • The paper discusses a new multi-modal biometrics based on face and ear. • The research significance of multimodal biometrics based on face and ear will be represented first. • then we introduce some researchers’ achievements and inadequacies. The images used in Chang’s • We also propose further exploration and experiment research in this area.

P-37

Friday August 7, 2009

Evening

FEP Modeling and simulation VIII FEP Network-based systems III FEP Network-based systems IV FEP Planning, scheduling and coordination I FEP Planning, scheduling and coordination II FEP Process optimization FEP Robotics and Control II FEP Robotics and Control III FEP Robotics and Control IV FEP Robotics and Control V FEP Robotics and Control VI FEP Robotics and Control VII FEP Strategy on logistics

FEP:Modeling and simulation VIII

Session Chair: Yibin Li and Chengjin Zhang

HQHT 16:40-17:40 Friday, August 7th

HQHT Hallway (1) 16:40-17:40 HQHT Hallway (2) 16:40-17:40 The Pseudorange Measurement Equations and The Research of Man-machine Cooperation in Their Solution for Satellite Navigation Warehousing System Based on Agent Ji Yuanfa, Sun Xigang,Shi Huli Yang Wei, Wang Wen and Fu WeiPing Applied Science and Technology College, Guilin University of Electronic Technology School of Mechanical and Instrumental Engineering, Xi'an University of Technology Guilin, China Xi’an , China • The common pseudorange • A new model of man-machine cooperation measurement equations and their in warehousing system is put forward, in solution for satellite navigation are which each function module is an Agent, discussed . and Petri Net is used to analyze the relationship and internal data processing of • The Unscented Kalman Filter (UKF) is the Agents. presented for nonlinear equations . • The software platform is realized based on • Experimental results showUKF can Java and VC++ , and Jade is the running achieve a better level of accuracy than platform of Agent. the common measurement equation • A detailed description of how man-machine The comparison of system computed with the iterative cooperation assists AGV to get rid of the iterative calculation method . dead interval of path is given here. results with UKF • It lays the foundation for further practical The Warehousing System man-machine cooperation in the system.

HQHT Hallway (3) 16:40-17:40 HQHT Hallway (4) 16:40-17:40 Research on Battery to Ride Comfort of Electric The Model Research of Information Automation Bicycle Based on Multi-Body Dynamics Theory System Based on RFID in Logistics Business Wenhua Du, Dawei Zhang and Xingyu Zhao School of Mechanical Engineering Tianjin University Enterprise of Warehouse Tianjin, China Minai He and Xinjun Wei Department of Transportation & Logistics Engineering , Shandong Traffic University • Based on multi-body dynamics theory Jinan, China the dynamic performance of electric • Present situation and demand analysis bicycle is simulated. of logistics informationization • The influence of the mass and location of • The structure model of network battery to ride comfort was discussed. warehouse management information platform • The ride comfort is better as the mass of • The Construction of Management battery increased. Information System in Logistics • When the mass of battery is lower, the Business Enterprise of Warehouse installation location does little influence The model of based on RFID to ride comfort, but as it is bigger, the coupled system • The data management way of network warehouse information platform influence is remarkable. The application of RFID

HQHT Hallway (5) 16:40-17:40 Application of Signal Detection and Processing on Active Power Filter

Yan-diWei1, Yong Zhang1, Xin-gong Cheng1, Yong-feng Zhang1 and Xin-feng Liu2 1School of Control Science and Engineering, University of Jinan 2Computer Engineering Department ,Shandong College of Electronic Technology Jinan, China • Signal acquisition and conditioning circuits in Active Power Filter (APF). • Propose an improved harmonic current detection method with sine/cosine function Lookup table. • Design of digital low-pass filter with fast response. • Tuning curve of the generated actual signal conditioning circuit board. The Prototype of Active Power Filter

P-38 FEP:Network-based systems III

Session Chair: Yibin Li and Chengjin Zhang

HQHT 16:40-17:40 Friday, August 7th

HQHT Hallway (6) 16:40-17:40 HQHT Hallway (7) 16:40-17:40 A New Control Mechanism for Real-time Video A Communication Model on Implementing Anycast Service with Transmission Based on H.264 Jie Zhao, Xuan Liu Multicast Tree in IPv6 College of Electronic and Information Engineering, Hebei University Baoding, China Wang Xiaonan School of Computer Science and Engineering, Changshu Institute of Technology • This paper proposed a new Changshu, China transmission framework for real- time video stream. • Anycast communication model: Anycast address, • We use H.264 algorithm as the architecture of Anycast tree, building an Anycast tree, video codec. weight value of Anycast tree. • A control manage sub-system is also introduced. • Routing analysis • The congestion of the network is • Performance analysis: total response time improved by using this framework.

The Transmission Framework

HQHT Hallway (8) 16:40-17:40 HQHT Hallway (9) 16:40-17:40 Modeling and Estimation of Networked Control Research on Peer-to-Peer Network Search Systems with Packet Losses Algorithm Based on Improved Ant Colony Yuan Ge, Qigong Chen, and Ming Jiang Department of Electrical Engineering, Anhui University of Technology and Science Optimization Wuhu, Anhui, China Zhengzhong Gao, Longji Liu and Songmei Zhang • Define the network state as the College of information and Electrical Engineering, Shandong University of Science and hidden state of an underlying Technology Markov chain. Qingdao, China • Treat the packet dropping process as • Present an improved ACO a stochastic variable whose PDF algorithm. depends on the Markov chain state. • All the experiments are based on • Model such NCS as a DTHMM and Peersim simulator. estimate the network state by EM • Added pheromones regulation algorithm. mechanism in ant colony • Design the proper control law based optimization. DTHMM of NCS on the estimated network state in the • Searching efficiency can be future. improved by using the algorithm. The experiment result

HQHT Hallway (10) 16:40-17:40 HQHT Hallway (11) 16:40-17:40 Control Target Tracking-oriented Based on An Adaptive Control Mechanism for Mitigating Overlapping Clustering DDoS Attacks Dongmei Yan1,2 JinKuan Wang1,2 Li Liu1,2 Bin Wang1 Peng Xu1 Qingtao Wu, Ruijuan Zheng,Jiexin Pu and Shibao Sun College of Information Science and Engineering Northeastern University ShenYang, China Electronic Information Engineering College, Henan University of Science and Technology Engineer Optimization & Smart Antenna Institute Northeastern University at Qinhuangdao Luoyang, China Qin huang dao, China • Control charts based-on network 20 20 • It is possible to extend the lifetime Base station x -control chart x -control chart connection for monitoring 15 15 UCL x x UCL 10 of a network by controlling its 10 CL suspicious behaviours. CL 5 Gateway 5 LCL topology through clustering. LCL 0 0 t(s) 246810 nodes • Adaptive mitigation which uses self- 10 20 30 40 50 t(min) 4 4 • Overlapping Clustering find its R-control chart R-control chart 3 3 learning scheme to establish UCL UCL R R wide applications such as inter- Cluster 2 2 CL CL cluster routing, topology discovery, head threshold of test statistics. 1 1 0 t(s) 0 A 10 20 30 40 50 246810 P =0.99 t(min) node localization, and recovery Sensing • Simulation experiments to evaluate d from cluster head failure and so on. nodes the efficiency is conducted. • Utilizing cluster technology can Fig.1 Hierarchical Routing Tree • Suitability for early DDoS attacks make good use of each sensor with multiple paths Target mitigation. nodes’ limited energy and perform Tracking using cluster tasks coordinately.

P-39 FEP:Network-based systems IV

Session Chair: Yibin Li and Chengjin Zhang

HQHT 16:40-17:40 Friday, August 7th

HQHT Hallway (12) 16:40-17:40 HQHT Hallway (13) 16:40-17:40 Implementing Communication Between Design and Implementation of Multi-Source Vehicular Ad Hoc Network and IPv6 Network by Information Monitoring System in Real Time Using Anycast Yufeng Chen1,2, Zhengtao Xiang1, Wei Jian1 and Weirong Jiang1 1School of Elec. and Info. Eng., Hubei Univ. of Automotive Technology, Shiyan, China Wang Xiaonan 2College of Computer Science and Technology, Zhejiang Univ., Hangzhou, China School of Computer Science and Engineering, Changshu Institute of Technology Changshu, China • Avoiding modifying vehicles and controlling the increase of • Key technology: address auto-configuration, routing, wiring harness to decrease etc. system complexity. • Communication model: address format, revised DSR • A vehicular information real- time monitoring framework is protocol, routing without movement, and routing proposed. with movement. • A flexible and reliable • Performance analysis: data reception rate, control communication protocol is designed to exchange multi- message rate source information between System architecture modules in the framework.

HQHT Hallway (14) 16:40-17:40 HQHT Hallway (15) 16:40-17:40 A Comprehensive Security Policy Research on Study on Web Classification Mining Method Web Information System Based on Fuzzy Neural Network Fengying Wang, Caihong Li, Lei Zhao, School of Computer Science and Technology ,Shandong University of Technology Jingwen Tian, Meijuan Gao, Yang Sun Zibo, China. College of Automation., Beijing Union University Beijing, China • we discuss whole security policies on Web information • Introduction input hidden output layer WW layer layer I O system from various aspects. • Fuzzy Neural Network • Security elements on a Web • Web Mining and Mining System information system include I O Structure Model single-sign-on , cross-domain access , access control, identity • Web Classification Mining Based I O j authentication, system audit, on Fuzzy Neural Network i k trust management, identity • Conclusion Structure of BP neural authentication, database network security. Interrelationship figure among security elements

P-40 FEP:Planning, scheduling and coordination I

Session Chair: Yibin Li and Chengjin Zhang

HQHT 16:40-17:40 Friday, August 7th

HQHT Hallway (16) 16:40-17:40 HQHT Hallway (17) 16:40-17:40 Minimizing Total Weighted Completion Time on Supply chain coordination with fuzzy demand a Batch-processing Machine with Re-entrance Jie Wei, Author2, …, and AuthorN Rui Zheng, Hongyu Li General Courses Department, cademy of Military Transportation School of Management, Fudan University Tianjin, China Shanghai, China • The single batch-processing machine scheduling problem with result1/result2 Consider a supply chain with one retailer and one manufacturer, the re-entrance to minimize total 1.4 retailer orders a type of products from the manufacturer and sells weighted completion time is 1.2 considered, where the capacity of them to end consumers. We assume that consumers' demand for this 1 the machine is infinite and there are product is uncertain (is a fuzzy variable), and assume that all different processing types in the 0.8 activities occur within a single period. The manufacturer must decide same machine. 0.6 on his product's wholesale price, and the retailer makes the ordering • This problem can be transformed 0.4 quantity of the product facing uncertain market demand. In this into a model with parallel chains 0.2 precedence constraint. paper, we give a coordination mechanism through profit sharing 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 contract under fuzzy environment, and assume that the manufacturer, • A polynomial time heuristic algorithm for the problem is given. as a Stackelberg leader, offers a profit sharing contract to the retailer Experimentation results show that Experimentation Results who faces fuzzy consumers' demand. the algorithm is efficient.

HQHT Hallway (18) 16:40-17:40 HQHT Hallway (19) 16:40-17:40 Material Balance Model and Scheduling Study on the Relationship between Port Cluster Algorithm of Cold Rolling Production Line and Regional Economic Development with SD Jun Zhao, Wei Wang, Sujie Tian Research Center of Information and Control, Dalian University of Technology Li Zhaoliang ，Zhao Xu , Xu ye-er ,Zhu Baoyi Dalian, China Transportation Management College , Dalian Maritime University Dalian, China • This paper first studies this interactive relationship qualitatively, then chooses system dynamics model, analyzing the characteristics of their coordinated development quantitatively in Zhejiang Province. • By changing three system parameters-the investment ratio in port, the • A total cost objective function, considering a great number of collaboration coefficient and ROR in port cluster, we do the simulation manufacturing units and a diversity of product categories in cold between the economic development and port cluster in Zhejiang Province, in rolling production line, is established in this paper. order to make port cluster develop more rapidly. Finally, we put forward the corresponding proposals and countermeasures. • A class of correlative genetic algorithm is used to obtain a better • It improves the whole operation affiance of port cluster t, and better serve the solution of this function. regional economic development, and finally it realizes the harmonious • The practical production data from a cold mill of Shanghai development. Baosteel Co., Ltd are adopted to verify the proposed method.

HQHT Hallway (20) 16:40-17:40 HQHT Hallway (21) 16:40-17:40 Scheduling optimization of Model Constrains based on Polychromatic Sets refinery operations based on Theory with Genetic Algorithm for Sequencing production continuity Ming Li and Qiqiang Li Machining Step on a Machining Center School of Control Science and Engineering, Shandong University Yunlong Li, Zongbin Li, and Enfu Liu Jinan, China State Key laboratory for Manufacturing System Engineering, Xi’an Jiaotong University • Production continuity is an Xi’an, China important process characteristic which receives significant attentions • The approach of sequencing machin- in actual refinery process. ing step on a machining center was • The core of this paper is to deal with proposed. the scheduling optimization problem • Constraints model was established of refinery operations based on based on polychromatic sets theory. production continuity. Simplified flow sheet of a refinery • Genetic Algorithm was used to • A generalized disjunctive programming model is proposed to sequence the machining step. tackle such scheduling problem. • A box part example demonstrated the • Case study shows that the optimal schedule obtained from the approach. Box part proposed model is more feasible and practical.

P-41 FEP:Planning, scheduling and coordination II

Session Chair: Yibin Li and Chengjin Zhang

HQHT 16:40-17:40 Friday, August 7th

HQHT Hallway (22) 16:40-17:40 HQHT Hallway (23) 16:40-17:40 Coordination Contracts for a Kind of Reverse The Study on Dynamic Scheduling of Mobile Supply Chain Based on Retailer’s Manipulator Based on MAS and Neural Network Delivery Time Guarantee Liu Shu-ying , Duan Ping , Ding Cheng-jun , Zhang Ming-lu Chenglin Shen 1, Xinxin Zhang 2, Ke ma1 School of Mechanical Engineering, Hebei University of Technology 1. School of Business, Tianjin Polytechnic University , 2. Tianjin Institute of Pharmaceutical Research Tianjin, China • In this paper, we consider a two-echelon reverse supply chain system with one manufacturer, one retailer and a group of • A distributed intelligent consumers, which offers consumers with decision and management perishable products, and consumers are allowed to return products to the retailer after system is erected. purchasing. Joint decision models of delivery time and ordering quantity as well as refund • Use Neural network as the was established separately in decentralized intelligent controller of the decision mode and centralized decision mode under the random market demand and hierarchical intelligent system delivery time. Further, we also analyze the optimal decisions under the two modes. Retailer’s Optimal Ordering Quantity Results show that centralized decisions not The converse kinetics only shorten the delivery time and increase the total profit of the supply chain but increase procedure of 5DOF manipulator the ordering quantity.

HQHT Hallway (24) 16:40-17:40 HQHT Hallway (25) 16:40-17:40 The Relationship between Location of New Time Control Based on Three- Central Business District and Traffic Flow Shiyan Xu1, Yuyao He 2, Xue Li 3 Dimensional Dynamic Path 1, 3 School of Electronic and Control Engineering, Chang’an University, Xi’an, China 2. College of Marine Engineering , Northwestern Polytechnical University, Xi’an, China Ji Jun, Ma Pei-bei, Lei Dan and Zhang Xiao-jie Department of Commanding , Naval Aeronautical and Astronautical University Yantai,China • A traffic equilibrium model is proposed to research the relationship between central • This paper proposes a new time business district (CBD) location and control method based three- traffic flow in the city. dimensional dynamic path planning . Fig.1 The initial contour plot of the flow intensity in the • To make the model more realistic, monocentric city • The proposed method can lead A vehicles to maneuver turning Δψ i+1 travelers can either travel directly to the i Bi+1 CBD on the continuum (surface streets), according to the signal of time-error. Z1 Y1 or they can travel on the continuum and • According to the time error, the yaw Ci X1 B A then exchange at an interchange (ramp) maneuvering command is generated Ai−1 i i before moving to the CBD on the discrete and three-dimensional dynamic path (a) (b) network. Fig.2 (a) The contour plot of the flow intensity choosing planning on the basis of the the facility0 and 1 as the CBD Sketch for turning segment • Finite element method is used to solve the (b) The contour plot of the flow intensity choosing prescribed path is done to control the facility0 and 4 as the CBD model. impact time.

HQHT Hallway (26) 16:40-17:40 Risk Identification, Evaluation and Mitigation in Spare Parts Purchasing NING Zhong，XU Xun School of Management, Fudan University Shanghai, China

• Introduction and Literature Review • Identification of Spare Parts Purchasing Risk of the Purchasing Department of Bao Maintenance Co. LTD • Evaluation and Further Evaluation of Spare Parts Purchasing Risk • Control of Spare Parts Purchasing Risk Distribution Diagram of Cluster • Conclusion and Outlook Combine Analysis Results

P-42 FEP:Process optimization

Session Chair: Yibin Li and Chengjin Zhang

HQHT 16:40-17:40 Friday, August 7th

HQHT Hallway (27) 16:40-17:40 HQHT Hallway (28) 16:40-17:40 Edge Grinding Process and Edge Checking Roller Bearings Fault Diagnosis Control for Screw Cutter Based on LS-SVM Guang-yao MENG, Jiwen TAN & Yucheng DING Wentao Sui1, Dan Zhang2, Wilson wang 3 Department of mechanical engineering Qingdao Technological University, School of Mechanical Engineering, Shandong University of Technology Zibo, China Qingdao (266033) China • A fault diagnosis method of roller bearings in rotating machinery • I. Introduction based on LS-SVM is presented. ⎧ dR/ dZ • II. Calculation of theory edge ZZr′ = ⎪ m 2 • LS-SVM is used as fault ⎪ 1/+（）dR dZ curve and tool-path. ⎨ 1 classifier. • III. Grinding force model and ⎪ RRr′ =± 2 fuzzy control ⎩⎪ 1/+（）dR dZ • A a feature selection method based on simulated annealing • VI. Edge checking and envelop error of edge curve (SA) is presented. fFΔ 1 • V. Processes of edge curve ii++11=−[(1 )α ⋅ ]β grinding and cutter re- faipii+Δ F sharpening simulated annealing algorithm • IV. Conclusion

HQHT Hallway (29) 16:40-17:40 HQHT Hallway (30) 16:40-17:40 Valve Control Command Based Fouling Differential Evolution Algorithm for Hot Rolling Detection of Heat Exchanger in an Aircraft Process Optimization Engine Bleed Air Temperature Control System Li Chen, Lixin Tang and Rui Luo Lan Shang and Guangjun Liu Liaoning Key Laboratory of Manufacturing System and Logistics, The Logistics Institute, Department of Aerospace Engineering, Ryerson University Northeastern University, Shenyang, China Toronto, Canada • A nonlinear model for hot rolling • Fouling of heat exchanger reduces the process optimization is designed. heat transfer efficiency and introduces additional resistance to flow. • An improved differential evolution • A fouling detection method is proposed algorithm by adjusting the mutation Command signal for bypass valve 13.0 average based on the valve control command of 12.5 factor and crossover rate is proposed. 12.3 12.0 11.8 the normal bleed air temperature 11.5 11.3 • The proposed differential evolution 11.0 10.8 regulation system in operation. 10.5 algorithm is used to solved the hot 10.0 Command signal (mA) Command • No additional sensor measurement is 9.5 9.4 rolling process optimization problem. 9.0 required for the heat exchanger fouling 8.5 250 300 350 400 450 Time (sec) detection. Experimental facility The process of improved and results DE algorithm

HQHT Hallway (31) 16:40-17:40 Design and Distributed Control of Discrete Event Robotic Manufacturing Systems using Petri Nets Gen’ichi Yasuda Nagasaki Institute of Applied Science Nagasaki, Japan

• A methodology of decomposition Petri net engine Station and coordination is presented for Global model controller hierarchical and distributed control Coordinator Determination of firing of robotic manufacturing systems global transitions Firability test result of Firability test result of • The overall Petri net is decomposed global transitions in global transitions in machine 1 model machine 2 model and the constituent subnets are Petri net engine Petri net engine Machine assigned to local Petri net based Machine 1 model Machine N model controller controllers. Machine 1 • The cooperation of each controller is Machine N implemented so that the behavior of the overall system is the same as the Structure of Hierarchical and centralized control system. Distributed Control

P-43 FEP:Robotics and Control II

Session Chair: Yibin Li and Chengjin Zhang

HQHT 16:40-17:40 Friday, August 7th

HQHT Hallway (32) 16:40-17:40 HQHT Hallway (33) 16:40-17:40 Detecting Abnormal State of Elderly for Service Robot with H-FCM Haitao Li,Lingfu Kong and Peiliang Wu College of Information Science and Engineering Yanshan University,Qinhuangdao, China A New Fusion Rule with Dynamic • The abnormal state of elder is dangerous. y Decision Threshold for • We can find some relation Heterogeneous Field Gripper between the state and the (x,y) Sensory System: Part III location information of elder. Debanik Roy • Home map is necessary. θ Board of Research in Nuclear Sciences Department of Atomic Energy, INDIA • H-FCM algorithm can be the E-mail: [email protected] effective method in detection. Home map built by robot

HQHT Hallway (34) 16:40-17:40 HQHT Hallway (35) 16:40-17:40 A Robust Adaptive Fuzzy Sliding Mode Control Brain-Computer Interface Based Camera Carrier in Aerospace Genghuang Yang, Li Zhao,Shigang Cui, Yuliang Liu, Longteng Xiao, Xuelian Xu Method of 3-RRRT Parallel Manipulator School of Automation & Electrical Engineering, Tianjin University of Technology & Education Yanbin Liu, Xinjie Jia, Jianhai Han, and Xinhua Zhao Tianjin, China School of Mechatronical Engineering, Henan University of Science and Technology • An on-line BCI system based on Luoyang, China

event-related potentials P300 is • On the base of Lyapunov’s stability Base platform designed to control self-designed theory, using fuzzy logic system with robot for photo-taking in simulated adaptive technology, a direct adaptive aerospace enviroment. Fuzzy sliding mode control strategy is • The BCI system is The robot can Actuated move forward and backward, go up put forward, and then control system joints and down, turn to left and right. numerical simulation is completed. Also the robot can grasp and release • simulation results show that in case of a shuttle of a camera. periodic interferential signal, • The EEG is sampled by a DSP based Motion platform device and the robot is controlled to employing direct adaptive Fuzzy move or operate by the subject’s sliding mode control strategy, the 3-RRRT Parallel Manipulator EEG with wire or wireless The Subject and the Manipulator parallel manipulator attain higher path communication. tracking precision.

HQHT Hallway (36) 16:40-17:40 HQHT Hallway (37) 16:40-17:40 Location for Audio signals Based Research on Motion Control System on Empirical Mode ecomposition of Mine Rescue Robot MA Hongwei Tian Hao LI Xiaopeng —Wu Xiao , State Key Laboratory of School of Mechanical Engineering, Xi’an University of Science and Technology, Xi’an China Precision Measuring Technology & x1 ()t x2 ()t Instruments , Tianjin University ,China EMD EMD • Real-time of robot running is IMF energy IMF energy distributing 1.Signals is separated into several distributing improved by hardware architecture intrinsic mode functions (IMFs) using Main IMFs Main IMFs basing on upper and lower empirical mode decomposition (EMD). selection selection processor of multi-cpu. 2.We compute normal energy of each Restructure Restructure IMFs and denoise IMFs according to signal signal • Reliability and maintainability of normal energy. motion control system are 3.The signals is restructured only Time dely improved by modularization of using main IMFs . estimate 4. Time delay is estimated and hardware and software’s design. location is completed by solving the Hardware structure of Target • Accuracy of motion control system geometry equation. location motion control system is improved by Closed-loop system. Fig. The flow of method for loction in paper

P-44 FEP:Robotics and Control III

Session Chair: Yibin Li and Chengjin Zhang

HQHT 16:40-17:40 Friday, August 7th

HQHT Hallway (38) 16:40-17:40 HQHT Hallway (39) 16:40-17:40 Global Path Planning for Mobile Robot Based on An automated method to Robot calibration Improved Artificial Potential Function using line-structure-light vision sensor Pu Shi , Yiwen Zhao Wenbiao Wang, Aiguo Li, Zi Ma, Huipu Xu , Yuan Tian Department of Automation Engineering , North-eastern University at Qinhuangdao Automation Research Center, Dalian Maritime University Dalian, China Qinhuangdao, China • The formation of local minima is a • An industrial robot calibration algorithm major problem with traditional is presented, which uses a line-structure- artificial potential field method. light sensor. • The kinematic parameters of the robot are • An improved potential field estimated due to the coordinates of the virtue point is invariable in robot base approach ensures the target is the frame. global minimum of the artificial • The proposed method uses a line- potential function. structure-light vision sensor on the robot’s end-effector to measure a fixed • Experimental results show that the point and the whole calibration process is separated into two stages: hand-to-eye improved approach has much higher calibration and robot calibration. capacity of global optimization than Mobile Robot, Obstacle traditional method. and Target

HQHT Hallway (40) 16:40-17:40 HQHT Hallway (41) 16:40-17:40 A Robotic Surgery Navigation System for Hepatic Microwave The Piecewise Monte Carlo Localization System Coagulation Therapy Jing Xu1, Zhangjun Song2,and Zhenzhong Jia3 for a Humanoid Soccer Robot 1. Michigan State University, East Lansing, Michigan, USA Wei Hong and Yantao Tian 2.Shenzhen Institute of Advanced Integration Technology, Shenzhen, China College of Communication Engineering, Jilin University, China 3. Department of Mechanical Engineering, University of Michigan, Michigan, USA Changjiu Zhou Advanced Robotics and Intelligent Control Centre, School of Electrical • The system includes a needle and Electronic Engineering, Singapore Polytechnic, Singapore insertion robot, a surgical planning subsystem, a magnetic tracking • Piecewise MCL method can change device and a stereo vision system. the size of sample set and resampling • First, a patient body fixation technique was used to reduce unintentional rules dynamically. . patient movement. Then, stereo vision • Two feature variables are defined to technique was used to track the divide the state of filter into two markers on the patient’s body surface. phases: global localization and local • By combining the fixation and stereo vision, we can trace the exact motion tracking. . of the organ, minimizing influence of • Physical walking toward target respiratory and cardiac motion. The planned trajectory experiments and simulative kidnap Snapshots of Robot local • Accuracy experiment on 3D ultrasound and target problem experiments were performed tracking in the playing field phantom was carried out to evaluate the navigation system. on the humanoid soccer robot. .

HQHT Hallway (42) 16:40-17:40 HQHT Hallway (43) 16:40-17:40 Application of Fuzzy Reasoning on an Research on the Omni-directional Mobile Autonomous Robot Manipulator Motion Planning based on Qingmei Yang1 Jiammin Sun2, Improved Genetic Algorithm 1. College of Automation, Beijing Union University, Beijing, China Lin Jiang1,Baiyan Liu1,Liangcai Zeng1, Xinyuan Chen1 , Jie Zhao2 and Jihong Yan2 2. Beijing University of Civil Engineering and Architecture, Beijing, China Wuhan University of Science and Technology1,HIT2 Wuhan, China

10 1 9 • Move-in-mud robot is a new kind 1 Ultrasonic obstacle • The research was on the omni- Digital compass of underwater robot, which is avoidance module Ultrasonic transmitter module used in wreck salvage. Kinematic directional mobile manipulator model of move-in-mud robot is D (ODMM) motion planning based on set up based on the mechanical the scope of species self-adjust GA. structure. 132 4 5678 1. Move-in-mud head 2. ripple pipe 3. turning mechanism • The speed and efficiency of • To control the filter exhale, Fuzzy 4. creepage feed mechanism 5. flexible connection Kalman filter is designed. The 6. control parts 7. sensors 8. circuit board redundant robot motion planning was simulation results show that the 9. former sustaining mechanism improved by the improved GA. precision of Fuzzy Kalman filter 10. back sustaining mechanism 11. complex cable • The omni-directional mobile platform Manipulator Omnidirectional Driving wheel Wireless module is better than the precision of mobile platform Steering motors Kalman filter. Project design of move-in-mud robot and manipulator cooperation drawing circle was performed Omni-directional-mobile manipulator

P-45 FEP:Robotics and Control IV

Session Chair: Yibin Li and Chengjin Zhang

HQHT 16:40-17:40 Friday, August 7th

HQHT Hallway (44) 16:40-17:40 HQHT Hallway (45) 16:40-17:40 Self-Adaptive Monte Carlo for Single-Robot and Constrained Motion Planning for Robot Manipulators Multi-Robot Localization Lei Zhang, René Zapata and Pascal Lépinay G. Antonelli, S. Chiaverini C. Curatella A. Marino LIRMM, Université Montpellier II, France DAEIMI, Università degli Comau S.p.A., Torino, DIFA, Università degli Studi • Propose the Self-Adaptive Monte Carlo Studi di Cassino, Italy Italy della Basilicata, Potenza, Italy Localization (SAMCL) algorithm. • SAMCL solves efficiently position • An off-line and non-iterative planning tracking, global localization and the algorithm for robot manipulators kidnapped robot problem, together. • Joint velocity, acceleration, jerk, and • SAMCL is extended to handle multi- torque limits are taken into account Comau Smart-Six and NJ130 robot cooperative localization through a Position Mapping (PM) algorithm. • Point-to-Point and fly motions in joint and cartesian space • Experiments are implemented on the robot Pioneer 3-DX in a real laboratory • Clotoid geometric blend in cartesian space environment. Experimental robot • Extensive simulations and experiments and environment A typical fly clotoid blend in cartesian space

HQHT Hallway (46) 16:40-17:40 HQHT Hallway (47) 16:40-17:40 Face Detection and Recognition with SURF Swarm of Robots Flocking via the Null-Space-based Behavioral Control for Human-Robot Interaction Shan An, Xin Ma, Rui Song, and Yibin Li Gianluca Antonelli Filippo Arrichiello Stefano Chiaverini School of Control Science and Engineering, Shandong University DAEIMI, University of Cassino, Italy Jinan, China http://webuser.unicas.it/lai/robotica • We have described a feature-based approach for face detection • Flocking problem for a swarm of robots and recognition for human-robot interaction (HRI), which is • Solved via a behavior based approach, capable of processing images rapidly and achieving high namely the Null-Space based detection and recognition rates. Behavioral control • The face region is located by employing some morphological steps after skin region detected. • Tested in presence/absence of a • SURF descriptors are used to common rendez-vous point extract the features of elliptical • Validated by numerical simulations and face region. experimental results with a team of grounded mobile robots • The query images are matched with the database images and the geometrical consistency The geometric SURF correspondence is checked. between the images

HQHT Hallway (48) 16:40-17:40 HQHT Hallway (49) 16:40-17:40

Adaptive Behavior Design Based Research and Design of Control System for on FNN for the Mobile Robot a Tracked SAR Robot Under Coal Mine Caihong Li, Yibin Li, Ping Chen Yibin Li , Caihong Li, Ping Chen Computer Science and Technology , Shandong University of Technology Research Laboratory of Robotics, Shandong University Zibo, China Jinan, China • A fuzzy neural network (FNN) has been • A search and rescue (SAR) robot, named trained off-line to memory the fuzzy LUKER-I, is introduced to replace people control rules of the adaptive behaviors for to access to the accident scene when the the local optimal path planning of mobile underground gas and dust explosion

robot. happen under coal mine. • The simulation results show that because • The control system uses two layers all training samples are from the trained architecture, namely the bottom control fuzzy rules, the output is almost the same network and IPC. as the result of the training rules. Path planned by FNN • A modular method is used to carry out a The robot shape structure detailed design.

P-46 FEP:Robotics and Control V

Session Chair: Yibin Li and Chengjin Zhang

HQHT 16:40-17:40 Friday, August 7th

HQHT Hallway (50) 16:40-17:40 HQHT Hallway (51) 16:40-17:40 Design of Linear Quadratic Optimal Controller Virtual Pheromones to Control Mobile Robots for Bicycle Robot Ioan Susnea, Grigore Vasiliu, Adrian Filipescu, Adriana Serbencu, Adrian Radaschin Department of Automation and Industrial Informatics of the “Dunarea de Jos” University Lei Guo, Qizheng Liao, Shimin Wei, and Yufeng Zhuang Galati, Romania School of Automation, Beijing University of Posts and Telecommunications Beijing, China • Aiming at the goal of balancing • Agents send queries containing their bicycle robot with high speed, a kind current position. of linear dynamic model of bicycle • The pheromone server locates the robot was presented. agent on the internal map, computes • The district time model of the the corresponding pheromone bicycle robot was presented. intensity, and sends the result back • A linear quadratic optimal controller to the agent. was designed for the linear dynamic model of bicycle robot. . • The agent adjusts its position to get • The computer simulation results closer to the existing pheromone How Virtual show the efficiency of the control trail, embedded in the map. Pheromones Operate algorithm. The Bicycle Robot • The map is dynamically modified.

HQHT Hallway (52) 16:40-17:40 HQHT Hallway (53) 16:40-17:40 Cooperative Q-learning Based on A New and Efficient Method for the Direct Kinematic Learning Automata Solution of the General Planar Stewart Platform Mao Yang, Yantao Tian and Xinyue Qi Xiguang Huang, Guangpin He School of Communication Engineering, Jilin University School of Mechanical and Electrical Engineering , North China University of Technology Changchun, Jilin,China Beijing,China, China • Object: accelerate the speed of reinforcement learning. • A new and efficient algorithm to solve • Communication architecture: a closed-form direct kinematics blackboard solution of the general 6-6 planar • Methods: Learning automata. Stewart platform is presented. • A 20th degree univariate equation and • Results of learning automata is used for sharing experiences among all 40 sets closed-form direct robots. kinematics solutions are obtained. • Finally a numerical example is given to verify the algorithm and its results Comparative Graph without extraneous roots agree with the The General Planar Stewart Platform original equations.

HQHT Hallway (54) 16:40-17:40 HQHT Hallway (55) 16:40-17:40 Symbolic-based Motion Control Method for Path-planning of Underwater Microrobot in 3-D Space Wheeled Mobile Robots Using Spiral Particle Pathway Searching Approach Shuxiang Guo*1, *2, Baofeng Gao*1 and Nan Xiao*1 1 2 2 2 Jianning Hua , Hongyi Li , Yuechao Wang and Ning Xi *1 Kagawa University, Takamatsu, Japan * 2 Harbin Engineering University, Harbin, China 1Northeastern University 2Shenyang Institute of Automation, Chinese Academy of Sciences • We have introduced research on a multi- • Great challenge can be incurred by the richness of the set of robot system including the structure and mappings from sensor to actuator signals in designing a suitable control system, task allocation, Motion control law to steer mobile robots. Planning Approaches. • A new approach called the Primary Spiral • Control procedures in the symbolic-based control method have a Particle Pathway Searching Approach was linguistic flavor. proposed to search for particles in the • Symbolic-based motion control method can be used to design pathway in the plane of parallel subspace. control laws at the level of strings and primitives. • MATLAB was used to calculate the particles in order to search the best path to the target points in 3-D space. • Finally, we determined the shortest Shortest pathway optimized pathway for a single robot, optimized by in 3-D space Framework for Symbolic-based Motion Control DPSO.

P-47 FEP:Robotics and Control VI

Session Chair: Yibin Li and Chengjin Zhang

HQHT 16:40-17:40 Friday, August 7th

HQHT Hallway (56) 16:40-17:40 HQHT Hallway (57) 16:40-17:40 Electric Control and Design for Automatic Study on the Calculation method of the Light Precision Forging Production Line Mobile Robot Motor Power GAO junyao, DUAN xingguang, YANG yigang, HE ming, WEI boyu GAO junyao, GAO xueshan, ZHU jianguo, ZHU wei, WEI boyu, WANG shilin Intelligent Robotics Institute, School of Mecha-Electronic Engineering, Beijing Institute of Intelligent Robotics Institute, School of Mecha-Electronic Engineering, Beijing Institute of Technology ,Beijing, China Technology ,Beijing, China • This paper presents a precision forging line of plate • Light mobile robot’s motor powers spares production. The line include press, robot, heating calculate and design is an furnace, convey. Main computer control each device important problem. It includes work in coordination with each other. Robots are used to weight, velocity, motor power, slope carry spares automatically. The line is automatic control angle, reduce gearbox, battery, etc. and high efficient, high quality. These factors influence each other. It is difficult to calculate precise each parameter. But it decides function and ability of mobile robot. In this paper, motor powers calculate method is advanced. Mobile robot climb slope There are many factors to consider. There are several aspects of Precision forging production line principle calculate method are analyzed.

HQHT Hallway (58) 16:40-17:40 HQHT Hallway (59) 16:40-17:40 Path Following Problem for PatrolBot Solved Research on Seam Tracking Controller with Fuzzy Control of Mobile Welding Robot Adriana Serbencu, Daniela Cristina Cernega, Adrian Emanoil Serbencu, and Ioan Susnea Ting Zhang, Kai Li, Shijie Dai, Shumei Xiao and He Huang Department of Automation and Industrial Informatics, “Dunarea de Jos” University Research Institute of Robotics and Automation, Hebei University of Galati, Romania Technology，Tianjin, China • The paper has been focused on • Dynamics model of the mobile welding implementing a fuzzy command on robot was designed. low-cost embedded devices, built • Self-turning fuzzy controller was designed around simple microcontroller to complete coordinately controlling of cross-slider and wheels. e ∑e structures. • The controller utilized self-learning and e α e e& ν 1 • The design and implementation of self-adaptive ability of the neural network / dtde 1−α to deal with the partial uncertainty and the e the fuzzy controller is described and disturbances of the parameters of the robot & analyzed model.. • It is proved that the selected control input • Tests on circular trajectory and torques make the system globally Self-turning fuzzy controller generalized two-dimensional curve The MobileRobots PatrolBot asymptotically stable based on the Lyapunov function selected out. are presented.

HQHT Hallway (60) 16:40-17:40 HQHT Hallway (61) 16:40-17:40 Navigation and Positioning Research of Service An Optimized Gait Generator Based on Fourier Robot Based on Intelligent space Series Towards Fast and Robust Biped Shouyin Lu Wei Qi Locomotion Involving Arms Swing Nima Shafii, Ali Khorsandian, Abbas Abdolmaleki, Bahram Jozi School of Information and Electrical Eng. Shandong Jianzhu University Young Researchers Club, Islamic Azad University Jinan, China cameras projecting Qazvin Branch, Iran cameras • The overall structure of the service robots system based on the • A model free approach with PC intelligent space emphasis on making robot’s walking • The navigation and positioning more stable and faster is presented. robot Particle Swarm Optimization (PSO) research based on the intelligent space has been used to optimize the signals artificial wireless produced by Truncated Fourier • Experimental results landmarks labels Series (TFS) which control joints’ angels. The role of hands is also The overall structure of considered to smooth walking and to NAO simulated robot the Service robots system increase its robustness.

P-48 FEP:Robotics and Control VII

Session Chair: Yibin Li and Chengjin Zhang

HQHT 16:40-17:40 Friday, August 7th

HQHT Hallway (62) 16:40-17:40 HQHT Hallway (63) 16:40-17:40 Design of Mobile Robot Teleoperation System Real Time Algorithm for Magnet's Based on Virtual Reality Localization in Capsule Endoscope Shuang Song, Chao Hu, Mao Li, Wanan Yang and Max Q.-H. Meng Ding Cheng-jun , Duan Ping , Zhang Ming-lu Shenzhen Institute of Advanced Technology (SIAT) Shenzhen China School of Mechanical Engineering, Hebei University of Technology Tianjin China Chinese University of HongKong Shatin,N.T. HongKong

• Sets up the model of . • Track the movement of the • Creats virtual hierarchical wireless capsule. controlenvironment, • Magnetic localization and designing video supervising orientation system, in a space of system and brings 0.5m*0.5m*0.5m. technology of virtual reality • Good accuracy, high speed and in robot teleopration system high robustness. • Nonlinear Minimization. • Localization error is 1.8mm, and The Interface of Virtual Environment orientation error is 1.54°.

HQHT Hallway (64) 16:40-17:40 HQHT Hallway (65) 16:40-17:40 Modular Robot System Configuration Design Simulation Research of Location System for Fire Detecting Robot Based on Axiomatic Design Liu Xuan, Zhang Minglu, Liu Wei, and Fan Lina Qingmei Yang1 Jiammin Sun2, School of Mechanical Engineering, Hebei University of Technology 1. College of Automation, Beijing Union University, Beijing, China Tianjin, China 2. Beijing University of Civil Engineering and Architecture, Beijing, China • AD is applied to the modular special robot configuration system. • General methods of multi-sensor • For a specific design mission, refine information fusion are analysed. design objective and design restraints, • The GPS and INS integrated measure and then complete decomposing of system of the robot is designed to FRs and DPs between the functional improving precise and reliable and physical domains. measurement information of the robot. • By adjusting design matrix, • The simulation results show that the configuration design order can be Design process as a mapping fusion location can improve the overall planned. By using the design precision of measurement system parameters, the corresponding module efficiently. y errors of fusion can be found from the module set easily.

HQHT Hallway (66) 16:40-17:40 Study on the Mobile Robot Reconfiguration Control Methods Jingli Lu, Chunguang Bu State Key Laboratory of Robotics, Shenyang Institute of Automation Chinese Academy of Sciences Shenyang, China • proposes a new reconfiguration control method for rough terrain and an auto-control method to get over channels, ridges etc.

• control method are validated experimentally on a six-leg-wheel hybrid robot of SIA. Leg-wheel hybrid robot

P-49 FEP:Strategy on logistics

Session Chair: Yibin Li and Chengjin Zhang

HQHT 16:40-17:40 Friday, August 7th

HQHT Hallway (67) 16:40-17:40 HQHT Hallway (68) 16:40-17:40 Optimal pricing decisions for a product with two Logistics Cost Optimized Election in the Manufacturing Process Based on ELECTURE-ⅡAlgorithm selling chances ——take XinFeiLin artificial wood-based board company as an example Jing Zhao, Author2, …, and AuthorN Qin Long, LI-Qing Meng , Lin-Fang Liu and Wei Chen School of Science, Tianjin Polytechnic University Faculty of Management and Economics, Southwest Forestry College, China Tianjin, China

This paper investigates the seasonal product's optimal pricing • The logistics cost optimization of manufacturing Cost enterprises is a significant link for reducing the total Y …Y Y problem in one period with two selling chances under price- cost. This paper analyzes several major logistic cost 1 2 n dependent demand. A retailer who has a chance to order a certain attributes of the production flow in such kind Scheme enterprise, checks all the alternative operating X1 Z …Z Z initial level of inventory of a product from a supplier before the programs in harmoniousness test and non- 11 12 1n harmoniousness test, and identifies their level in beginning of stage 1, which can be used to sell by the retailer at X superior relation. And it figures out the average 2 Z21 22 …Z Z2n stage 1 and stage 2. Another new product will be introduced to the sequence by means of forward sequencing and market by this supplier, which will influence the customers' demand reverse sequencing, and eventually comes out the … ………… final sequent results ranking from superiority to for the existing product. The retailer's problem is to determine the inferiority that stem from each alternative program Xm Z …Z Z initial ordering quantity and the optimal retail price at stage 1 and at for logistics cost control. The purpose of this m1 m2 mn research lies in helping decision-makers select the stage 2, considering that the product's retail price charged both at optimal program scientifically and make decisions Decision Matrix of Schemes stage 1 and at stage 2 will influence the product's demand at stage 2. rationally.

HQHT Hallway (69) 16:40-17:40 HQHT Hallway (70) 16:40-17:40 Benchmarking Model for Reverse Logistics Realizing Risk: A Shift from Classic Optimization Entry by Third-party Providers in Production Processes Shuyun Wang, Shuai Zhang School of Economics & Management, Yantai University Lei Chen, Allen Fang Yantai, China Fuzhou University, Fujian, China • Benchmarking model is theoretically • Most logistics studies concentrate on useful for a third-party logistics optimizing costs. provider to make decisions. • However, when considering production • It can help the third-party logistics processes, manufacturers may become

company in making the decision to prude and maximize their own utility function. more aggressively enter the reverse → → Demand • That utility function may include logistics business or to decline. factors such as costs, revenues, market • Chinese logistics providers should share, and also risk. utilize the benchmarking model to • We present an efficient way to A Production System make great improvements. calculate the trade-off between costs and risk. Benchmarking model

HQHT Hallway (71) 16:40-17:40 Optimal Location Planning of Logistics Park with Variable Capacity Jianxun Tang, Lixin Tang Liaoning Key Laboratory of Manufacturing System and Logistics, The Logistics Institute, Northeastern University, Shenyang, China • A mix integer programming model for location planning of logistics park is designed. • Size mode is introduced to reformulate model. Then the set partitioning model is given and the column generation is used to solve the model to get a good low bound. • Based on the lower bound, we obtain the best solution by tabu search. The flow chart of solving the problem

P-50 FEP:Vehicle routing and planning system

Session Chair: Yibin Li and Chengjin Zhang

HQHT 16:40-17:40 Friday, August 7th

HQHT Hallway (72) 16:40-17:40 HQHT Hallway (73) 16:40-17:40 Trends and Analysis of Development of Dual Mode Trailer systems for Enhancing Railroad Logistics Dynamic Model and Scheduling System for a Young Jin Lee1,Hyeun Cheol Cho2,Dong Seop Han3, Geun Jo Han3,Jin Woo Lee4, and Kwon Soon Lee4 Real-Time Pickup and Delivery Problem with 1Dept. of Aero Electrical Engineering, Korea Aviation Polytechnic College, S. Korea 2School of Electrical & Electronic Eng., Ulsan College, S. Korea 3Dept. of Mechanical Eng., Dong A University, S. Korea 4Dept. of Electrical Eng., Dong A University, S. Korea Time Windows YongJi Jia, ChangJun Wang, Yong Liu and XiaoFeng Wang • Diverse DMT(Dual mode trailer) systems are Glorious Sun School of Business and Management, Donghua University introduced and investigated in terms of working Piggyback Shanghai, China time, loading type, connectivity of Ro-Ro system, productivity, logistic Cost, maintenance cost, Bimodal • Real-Time Pickup and Delivery and etc. Mo da lo hr Problem with Time Windows is • In the result, the parallel method and wagon Dual Mode Trailer Cargo Speed Tr a n s p o r t a t i o n Tu r n i n g Typ e rotating method have more advantage because System Flexiwaggon studied. they have diagonal loading and unloading Cargo Be ame r • Framework of scheduling system mechanism. Cargo Domino • Finally, we proposed an improved transportation λ Pa ra lle l Typ e CCT Plus based on -opt algorithm is system with buffer. The basic idea of new type is based on LMTT (Linear Motor Transfer NETHS Plu s proposed. Technology). Othe rs • Simulation result shows that the • The proposed system can give many advantages scheduling system is well suitable such as reduction of working time and increasing The Dual Mode Trailer Systems productivity in logistics viewpoint. to deal with RTPDPTW in real life conditions.

HQHT Hallway (74) 16:40-17:40 HQHT Hallway (75) 16:40-17:40 A Rolling Horizon Procedure for Dynamic Pickup and Delivery Problem with Time Research on Timing-Critical Mission Based on Windows Dynamic Path Planning YongJi Jia1, ChangJun Wang1 and LiMing Wang2 Ma Pei-bei, Lei Dan ,Ji Jun and Zhang Xiao-jie 1Glorious Sun School of Business and Management, Donghua University Shanghai, China Department of Commanding , Naval Aeronautical and Astronautical University Yantai,China 2Procurement Fulfillment Dept., Huawei Technologies Co., Ltd. Shenzhen, China • With the increasing complication of the battlefield ,UCAVs are required to reach targets at times • Periodic and event-driven RHP within a certain tolerance in order to achieve the necessary method for solving dynamic PDPTW sequencing of mission goals. is discussed. • Time error signal is obtained from the real time-to-go and • RHP is a heuristic mechanism with the estimated time-to-go by adopting a simple online time- which the global scheduling is to-go estimation method. replaced by a procedure of solving a • According to the time error, the yaw maneuvering series of local sub-problems. command is generated and dynamic path planning on the • Simulation results show that RHP basis of the prescribed path is done to control impact time. method is fast and effective and its • The results of simulation show the reasonableness of the performance is satisfactory. method.

HQHT Hallway (76) 16:40-17:40 HQHT Hallway (77) 16:40-17:40 An Improved Immune Clonal Selection Study of Toll-by-load policy effects on vehicle Algorithm and its Applications for VRP type constitution of parallel network Chen Liang1, Li Qiaoru1 , Li Xiaoxiao1 and Zhi Xuejun2 Ma Jia Shi Gang Gao Liqun 1Department of Traffic Engineering, Hebei University of Technology, Tianjin, China;2 Road Shenyang Aeronautical Engineering Institute, Shenyang china and bridge engineering company, Tianjin Dagang oil field, Tianjin, China • The influences of toll-by-load policy 400 • This paper is put forward an on vehicle type constitution are improved immune clonal selection discussed. 350 algorithm (ICSA) through the • The variations of toll collecting 300 introduction of cloning operator. volume of toll gates after adopting 250

• clonal proliferation operators、super toll-by-load policy are analyzed. 200 The mi nmum general ized cost Freeway mutation operators and clonal • This paper presents the generalized Parallel road 150 type2 type3 type4 type5 selection operators. cost route choice model with Truck type • The paper applies ICSA to solving considering the time value. the vehicle routing problems and • Take Jin-ji and the parallel road Jin- Comparison of the minimum wei as an example, the minimum generalized cost on Freeway and the verifies the validity and stability. generalized costs on Freeway and parallel road under current toll rate the parallel road under current toll rate are Compared.

Flow chart of immune clonal selection algorithm

P-51

2009 IEEE ICAL SESSION CHAIR INDEX Bai, Jing...... FP-1 Chen, Yuqing...... WP-4 Du, Xiaotong...... TM-3 Ge, Sam...... WM-4 Ge, Sam...... TM-2 Gu, Jason...... WM-2 Gu, Jason...... FPP He, Fujun...... WP-4 Hou, Xuezhang...... TM-1 Hu, Chao...... TM-4 Hu, Chao...... WPP Hu, Chao...... WP-5 Jin, Shi...... WP-3 Li, Yangmin...... FM-3 Li, Yibin...... FEP Li, Yangmin...... WP-1 Li, Dayou...... WP-2 Li, Dayou...... FM-1 Li, Xianming...... TPP Liu, Zaiwen...... FM-2 Lu, Fei...... WM-5 Lu, Fei...... FM-3 Ming, Aiguo...... FP-4 Ming, Aiguo...... TM-1 Park, Jong Hyeon...... WM-1 Qi, Donglian...... WP-3 Shen, Yanjun...... FP-2 Shen, Yanjun...... WPP Su, Chunyi...... FM-4 Sun, Qiqi...... WM-5 Sun, Shuli...... FP-1 Tao, Ze...... FM-5 Tian, Guohui...... TPP Wang, Hong...... WM-3 Wang, Hong...... TM-4 Wei, Qiang...... WM-3 Wei, Yingzi...... FM-5 Wu, Lingfei...... WP-1 Xu, Na...... WP-5 Xu, Na...... WM-2 Yang, Simon...... WM-4 Yang, Simon...... TM-2 Yang, Simon...... FPP Yu, Xiaohua...... FM-1 Yu, Xiaohua...... TM-5 Zaki, Aziza...... WM-1 Zhang, Mingzhu...... WP-2 Zhang, Dengpan...... TM-3 Zhang, Yingwei...... FM-4 Zhang, Yunong...... FP-2 Zhang, Chengjin...... FP-3 Zhang, Hong...... FP-3 Zhang, Yingwei...... FP-4 Zhang, Chengjin...... FEP Zhong, Ming...... TM-5 Zhong, Ming...... FM-2

S-1

2009 IEEE ICAL AUTHOR INDEX

Chen, Jinhua...... T M - 2 . 4 Chen, Lei...... F E P . 7 0 A Chen, Lei TPP.47 Chen, Li...... F E P . 3 0 Aanaes, Henrik...... T M - 1 . 4 Chen, Liang...... F E P . 7 7 Abdolmaleki, Abbas...... F E P . 6 1 Chen, Lihua...... F P P . 3 4 Afshar, Puya...... T M - 4 . 4 Chen, Mou...... W M - 4 . 1 Aguridan, Razvan...... F P P . 1 4 Chen, Ping...... F E P . 4 8 Ai, Changsheng...... F P P . 5 8 Chen, Ping FEP.49 Ai, Maoliang...... W P - 2 . 2 Chen, Qigong...... F E P . 8 Al Janaideh, Mohammad...... W M - 4 . 4 Chen, Sanfeng...... W P P . 5 6 An, Aimin...... T M - 4 . 1 Chen, Shouming...... F M - 3 . 5 An, Aimin T M - 4 . 2 Chen, Wei...... F E P . 6 8 An, Shan...... F E P . 4 7 Chen, Weiru...... F P P . 6 3 Angerer, Andreas...... W M - 1 . 4 Chen, Xinyuan...... F E P . 4 3 Antonelli, Gianluca...... F E P . 4 5 Chen, Xue...... F M - 4 . 1 Antonelli, Gianluca F E P . 4 6 Chen, Xue FP-4.1 Arrichiello, Filippo...... F E P . 4 6 Chen, Xuewen...... F P P . 3 4 Chen, Xuewen F P P . 3 6 Chen, Yong...... F P P . 6 9 B Chen, Yong T P P . 5 4 Chen, Yongqi...... F P P . 6 0 Bai, Jing...... F P - 1 . 3 Chen, Yuchen...... W P - 4 . 1 Bai, Qiuguo...... W P P . 6 8 Chen, Yufeng...... F E P . 1 3 Bai, Rui...... T P P . 6 5 Chen, Yuqing...... W P P . 3 4 Bi, Meng...... T P P . 4 8 Chen, Zhen...... T P P . 3 8 Bobasu, Eugen...... F P P . 3 2 Chen, Zhenxue...... W P P . 4 2 Bu, Chunguang...... F E P . 6 6 Chen, Zhen-xue...... FPP.3 Cheng, Bo...... W P P . 5 1 Cheng, Chengqi...... W P P . 4 7 C Cheng, Jun...... F P P . 2 6 Cheng, Qianqian...... W P P . 4 0 Cheng, Xingong...... F E P . 5 Cai, Wen-jian...... F P P . 1 8 Cheng, Xuezhen...... F P P . 5 7 Cai, Yanhui...... F P P . 2 0 Cheng, Yu Hua...... W P P . 6 0 Cai, Ying...... F P P . 5 2 Chiaverini, Stefano...... F E P . 4 5 Cao, Huibin...... W P P . 5 0 Chiaverni, Stefano...... F E P . 4 6 Cao, Liwen...... W P P . 7 Cho, Hyeun Cheol...... F E P . 7 2 Cao, Maoyong...... F P P . 5 7 Choo, Yoo Sang...... W M - 4 . 1 Cao, Taibin...... W P P . 6 3 Christensen, Kim Hardam...... T M - 1 . 4 Cao, Xianqing...... F P P . 2 9 Chu, Hongxia...... W P P . 3 7 Cao, Xianqing T P P . 4 8 Cui, Hai...... T P P . 2 0 Cao, Yajuan...... F E P . 3 6 Cui, Jiefan...... F P P . 3 5 Cao, Yongcun...... F P P . 4 3 Cui, Jiefan W P P . 2 4 Cao, Zheng...... W P P . 4 6 Cui, Shigang...... F E P . 3 4 Cernega, Daniela Cristina...... F E P . 5 8 Cui, Ze...... T M - 1 . 3 Chai, Tianyou...... T P P . 6 5 Curatella, Cataldo...... F E P . 4 5 Chai, Tianyou W M - 3 . 5 Chang , Yongping...... T P P . 2 Chang, Chunwu...... W P P . 4 4 D Chang, Faliang...... W P P . 4 2 Chang, Fa-liang...... FPP.3 Chao, Peng...... T P P . 8 Dai, Houde...... W P P . 5 5 Che, Haijun...... W P P . 6 4 Dai, Shijie...... F E P . 5 9 Chen, Dongmei...... F P P . 3 1 Dai, Wenzhan...... W P P . 3 3 Chen, Fei...... F P - 4 . 4 Dawids, Steen...... T M - 1 . 4 Chen, Hanli...... F P P . 5 4 Deng, Mingcong...... T M - 4 . 5 Chen, Huan...... W M - 2 . 1 Deng, Shujun...... W P P . 4 7 Chen, Jian...... T M - 3 . 4 Deng, Zhiyun...... W P - 5 . 1 Chen, Jianghong...... T P P . 6 Ding, Chengjun...... F E P . 6 2 Chen, Jiann-liang...... F P - 4 . 2 Ding, Chengjun T P P . 6 6

A-1

2009 IEEE ICAL AUTHOR INDEX

Ding, Haijuan...... T P P . 2 0 Gao, Fei...... F P P . 5 1 Ding, Ning...... FPP.8 Gao, Guoan...... T P P . 1 8 Ding, Peng-hua...... W P P . 4 0 Gao, Houli...... F M - 1 . 1 Ding, Ran...... F E P . 2 0 Gao, Huibin...... W P P . 2 9 Ding, Ran W P P . 7 0 Gao, Jinfeng...... W P P . 1 6 Ding, Rui...... W M - 2 . 2 Gao, Junjun...... W M - 5 . 2 Ding, Rui W P - 1 . 1 Gao, Junyao...... F E P . 5 6 Ding, Wei...... W P P . 5 8 Gao, Junyao F E P . 5 7 Dong, Xiaohui...... W P P . 2 Gao, Meijuan...... F E P . 1 5 Du, Chenglie...... F P - 2 . 3 Gao, Meijuan F P P . 1 5 Du, Wenhua...... F E P . 3 Gao, Tongyue...... W P P . 5 8 Du, Xiaoqin...... T P P . 3 9 Gao, Wen...... W P - 1 . 2 Du, Xiaoqin W P P . 6 1 Gao, Xinwen...... W P P . 5 8 Du, Xiaotong...... T M - 3 . 1 Gao, Yang...... T P P . 5 8 Du, Xiaotong W P - 2 . 2 Gao, Zhengzhong...... F E P . 9 Du, Zhijiang...... W P - 4 . 4 Ge, Jianghua...... T P P . 1 8 Duan, Bin...... T P P . 5 6 Ge, Shuzhi Sam...... W M - 3 . 5 Duan, Dunqin...... W P P . 4 1 Ge, Shuzhi Sam W M - 4 . 1 Duan, Guang-ren...... W P - 1 . 3 Ge, Shuzhi Sam W M - 4 . 2 Duan, Hui...... F P P . 1 3 Ge, Yan-hong...... W M - 2 . 1 Duan, Ping...... F E P . 2 3 Ge, Yinghui...... W P P . 3 5 Duan, Ping FEP.62 Ge, Yuan...... F E P . 8 Duan, Ping TPP.66 Ge, Yunjian...... W P P . 5 0 Duan, Quansheng...... W P P . 5 9 Geng, Haipeng...... F P - 2 . 1 Ghumman, Usman...... F P - 3 . 4 Gong, Bowen...... T P P . 2 5 E Gong, Bowen W P P . 6 7 Gong, Yimin...... F P - 1 . 2 Elshafei, Abdel-latif...... W M - 1 . 1 Gong, Zhenbang...... W P P . 5 8 Gu, Jason...... F M - 3 . 5 Gu, Jason TPP.12 F Gu, Jason TPP.7 Gu, Kanfeng...... F M - 5 . 2 Fan , Xinnan...... W P P . 4 0 Guan, Liwen...... T P P . 3 Fan, Lina...... F E P . 6 5 Guan, Wei...... W M - 3 . 4 Fan, Liping...... F P P . 2 9 Guan, Zhiguang...... W P - 5 . 2 Fan, Xiaoxu...... F P P . 5 0 Guo, Beitao...... W P - 4 . 2 Fan, Zhun...... T M - 1 . 4 Guo, Chao...... F M - 4 . 1 Fang, Allen...... F E P . 7 0 Guo, Chao F P - 4 . 1 Fang, Hairong...... T P P . 6 Guo, Chaozhong...... T M - 4 . 3 Fang, Huijiang...... W M - 5 . 5 Guo, Chen...... W P P . 6 9 Fang, Xinli...... T P P . 3 7 Guo, Feng...... T P P . 1 1 Feng, Daguang...... W P - 5 . 4 Guo, Hongmi...... T P P . 4 6 Feng, Guihong...... F P P . 4 5 Guo, Junmei...... W P P . 3 9 Feng, Xuejun...... T P P . 2 9 Guo, Lei...... F E P . 5 1 Filipescu, Adrian...... F E P . 5 0 Guo, Libin...... T P P . 2 0 Fu, Jie...... W P P . 1 1 Guo, Peng...... W P - 4 . 5 Fu, Weiping...... F E P . 2 Guo, Qingqiang...... F E P . 2 0 Fu, Weiping F M - 2 . 2 Guo, Shide...... W P P . 4 7 Fu, Weiping F P P . 3 9 Guo, Shuxiang...... F E P . 5 5 Fu, Yu...... F P P . 6 4 Guo, Xiao-li...... W P P . 2 3 Fu, Yue...... F P P . 3 5 H G Han, Dong Seop...... F E P . 7 2 Gan, Yutao...... F P P . 4 8 Han, Geun Jo...... F E P . 7 2 Gao, Baofeng...... F E P . 5 5 Han, Jian...... T P P . 3 Gao, Changsheng...... T P P . 5 Han, Jianjun...... T P P . 3 9 Gao, Dezhi...... W P P . 7 1 Han, Jianjun W P P . 6 1

A-2

2009 IEEE ICAL AUTHOR INDEX

Han, Pingyang...... F P P . 5 1 Han, Qiang...... F M - 1 . 3 Han, Xinyan...... W P P . 6 4 J Han, Xuan...... W M - 4 . 3 Han, Yaozhen...... W P P . 1 3 Jahanshahi, Mohsen...... T M - 3 . 2 Han, Yinghua...... T P P . 4 2 Ji, Feng...... F P P . 4 5 Hao, Changzhong...... F M - 5 . 5 Ji, Peng...... F P - 4 . 4 Hao, Minghui...... W P - 3 . 1 Ji, Yuanfa...... F E P . 1 Hao, Pingbo...... F M - 5 . 2 Ji, Yuanfa F P P . 7 1 Hao, Shuanghui...... W P - 3 . 1 Jia, Lei...... F P P . 1 8 Hao, Xiaohong...... T M - 4 . 1 Jia, Qiuling...... T P P . 2 3 Hao, Xiaohong T M - 4 . 2 Jia, Rongzhen...... F M - 2 . 4 Hao, Zhihong...... F P - 1 . 3 Jia, Yan...... F P P . 1 1 He, Baina...... F P P . 6 8 Jia, Yan F P P . 1 2 He, Fujun...... W P - 4 . 4 Jia, Yongji...... F E P . 7 3 He, Minai...... F E P . 4 Jia, Yongji F E P . 7 4 He, Shan...... T P P . 4 1 Jia, Yongji T P P . 1 6 He, Youguo...... T P P . 5 8 Jia, Zhenzhong...... F E P . 4 1 He, Yuyao...... F E P . 2 4 Jia, Zhenzhong T P P . 9 Hirata, Yasuhisa...... W P - 2 . 5 Jian, Huang...... W M - 5 . 1 Hoffmann, Alwin...... W M - 1 . 4 Jian, Wei...... F E P . 1 3 Hong, Qingfeng...... W M - 3 . 2 Jiang, Jing...... F M - 1 . 5 Hong, Wei...... F E P . 4 0 Jiang, Lin...... F E P . 4 3 Hou, Xiaoli...... W P P . 2 3 Jiang, Lin F M - 1 . 5 Hou, Xuezhang...... T M - 1 . 1 Jiang, Liupeng...... T P P . 2 9 Hu, Chao...... F E P . 6 3 Jiang, Ming...... F E P . 8 Hu, Chao F P P . 3 1 Jiang, Qi...... T P P . 3 2 Hu, Chao W P P . 3 1 Jiang, Shen...... F M - 2 . 1 Hu, Chao W P P . 4 3 Jiang, Wei...... T M - 3 . 5 Hu, Chao W P P . 5 5 Jiang, Weirong...... F E P . 1 3 Hu, Chengzhong...... W M - 3 . 1 Jiang, Xingyu...... W M - 5 . 4 Hu, Hongsheng...... T P P . 5 2 Jiang, Xinli...... F M - 5 . 2 Hu, Huosheng...... T M - 5 . 5 Jiang, Zhenhuan...... T P P . 5 9 Hu, Jian...... F M - 4 . 2 Jiang, Zhong-ping...... W P P . 6 6 Hu, Jian F M - 4 . 3 Jiang, Zhuangde...... T P P . 4 7 Hu, Qing...... W P P . 4 Jiao, Guiqiu...... W P P . 6 9 Hu, Wenbin...... T M - 2 . 5 Jiao, Weidong...... T P P . 2 Hu, Xiaojun...... W P P . 6 9 Jin, Guang...... F P P . 4 9 Hu, Yinquan...... F P P . 2 7 Jin, Jian...... T M - 1 . 2 Hua, Jianning...... F E P . 5 4 Jin, Qi-bing...... T P P . 6 4 Huang, Bin...... T P P . 1 0 Jin, Shi...... T P P . 6 9 Huang, Bin W P P . 3 9 Jin, Shi W P - 3 . 4 Huang, Biqing...... W M - 5 . 3 Jin, Shijiu...... F E P . 3 6 Huang, Bo...... T M - 2 . 1 Jin, Yuzhen...... W P P . 4 8 Huang, Bo T M - 5 . 2 Jing, Shaohong...... W P P . 2 8 Huang, Daoping...... W P P . 6 5 Jing, Weixing...... F P P . 5 8 Huang, He...... F E P . 5 9 Jing, Wuxing...... T P P . 5 Huang, Shunliang...... F M - 1 . 1 Jozi, Bahram...... F E P . 6 1 Huang, Xiaoling...... W P P . 9 Jun, Ji...... F E P . 2 5 Huang, Xiguang...... F E P . 5 3 Huang, Yongtao...... T P P . 1 8 Huang, Yousen...... W P - 5 . 1 K Huang, Zengxi...... F P P . 7 3 Kang, Junfeng...... F P P . 5 3 I Kang, Ting...... W M - 5 . 2 Khorsandian, Ali...... F E P . 6 1 Kim, Jeong Jung...... W M - 1 . 2 Inoue, Akira...... T M - 4 . 5 Kim, Nam-yee...... F M - 3 . 2 Ionete, Adina...... F P P . 1 4 Kong, Lingfu...... F E P . 3 2 Ionete, Cosmin...... F P P . 1 4 Kong, Xiangzhen...... W P P . 2 0

A-3

2009 IEEE ICAL AUTHOR INDEX

Kosuge, Kazuhiro...... W P - 2 . 5 Li, Nailu...... F P P . 5 0 Kristensen, Jens...... T M - 1 . 4 Li, Qian...... F P P . 7 0 Kunwar, Faraz...... F P - 3 . 4 Li, Qiaoru...... F E P . 7 7 Kwak, Hoon-sung...... F M - 3 . 2 Li, Qiaoyun...... F M - 5 . 4 Li, Qin...... W P P . 6 6 Li, Qinghua...... T P P . 3 9 L Li, Qinghua W P P . 6 1 Li, Qiqiang...... F E P . 2 0 Lai, Huige...... F M - 1 . 2 Li, Qiqiang W P P . 7 0 Lee, Jin Woo...... F E P . 7 2 Li, Shiwu...... FPP.8 Lee, Ju Jang...... W M - 1 . 2 Li, Shou-tao...... T P P . 5 1 Lee, Kwon Soon...... F E P . 7 2 Li, Shujun...... F P P . 4 9 Lee, Sang-hun...... W M - 1 . 5 Li, Shuying...... T P P . 6 1 Lee, Tong Heng...... W M - 3 . 5 Li, Suling...... T P P . 1 Lee, Tong Heng W M - 4 . 2 Li, Suling W P P . 1 Lee, Young Jin...... F E P . 7 2 Li, Wen-feng...... W M - 2 . 1 Lei, Kai...... W P P . 6 0 Li, Xianming...... T P P . 3 5 Lei, Meizhen...... W P P . 1 6 Li, Xianming W M - 3 . 1 Lepinay, Pascal...... F E P . 4 4 Li, Xiao-fang...... F M - 4 . 5 Li, Aiguo...... F E P . 3 8 Li, Xiaolan...... F P P . 3 3 Li, Bin...... F P P . 6 2 Li, Xiaopeng...... F E P . 3 7 Li, Bin W M - 2 . 1 Li, Xiaoxiao...... F E P . 7 7 Li, Bin W P - 1 . 4 Li, Xiumei...... F E P . 1 4 Li, Bing...... T P P . 4 7 Li, Xue...... F E P . 2 4 Li, Caihong...... F E P . 1 4 Li, Yangmin...... W P - 1 . 5 Li, Caihong F E P . 4 8 Li, Yibin...... F E P . 4 7 Li, Caihong F E P . 4 9 Li, Yibin FEP.48 Li, Can...... F P P . 3 9 Li, Yibin FEP.49 Li, Cheng...... T P P . 4 5 Li, Yibo...... F P P . 7 3 Li, Chunling...... F P P . 2 1 Li, Yong...... W P P . 5 4 Li, Dayou...... W P - 2 . 4 Li, Yongxiang...... W P - 2 . 3 Li, Dong...... F P P . 4 6 Li, Yong-xiang...... F P - 1 . 1 Li, Feng...... FPP.7 Li, Yong-xiang WP-3.2 Li, Gangyan...... F M - 4 . 2 Li, Yong-xiang WP-3.3 Li, Gangyan F M - 4 . 3 Li, Yongxin...... W P - 3 . 4 Li, Guangwen...... F M - 4 . 4 Li, Youfu...... F M - 3 . 3 Li, Guangwen T M - 5 . 3 Li, Yuan...... W P P . 5 7 Li, Guangwen T P P . 2 3 Li, Yunlong...... F E P . 2 1 Li, Haitao...... F E P . 3 2 Li, Yuren...... W M - 2 . 3 Li, Hongbin...... F P P . 4 6 Li, Zhaoliang...... F E P . 1 9 Li, Hongxia...... T P P . 4 9 Li, Zhenhua...... T P P . 5 6 Li, Hongxing...... W P P . 6 2 Li, Zhigang...... W P P . 6 8 Li, Hongyi...... F E P . 5 4 Li, Zhijun...... F P - 4 . 3 Li, Hongyu...... F E P . 1 7 Li, Zhiqing...... W P - 1 . 4 Li, Hu...... F P P . 2 6 Li, Zhongjian...... W P P . 3 0 Li, Jia-jue...... W P - 3 . 2 Li, Zongbin...... F E P . 2 1 Li, Jin...... F M - 1 . 3 Liang, Huawei...... W P - 1 . 2 Li, Jing...... T P P . 2 6 Liang, Jing...... FPP.9 Li, Jinglei...... F M - 2 . 4 Liang, Lingyan...... W P P . 4 1 Li, Jizi...... T P P . 1 5 Liang, Tao...... W P P . 7 0 Li, Junfeng...... W P P . 3 3 Liang, Weili...... F P P . 2 8 Li, Kai...... F E P . 5 9 Liang, Xiao-lei...... W M - 2 . 1 Li, Kai F M - 5 . 1 Lin, Ciyun...... T P P . 2 5 Li, Li...... F M - 2 . 1 Lin, Ciyun W P P . 6 7 Li, Li F M - 5 . 3 Lin, Mingxing...... W P - 5 . 2 Li, Li W P P . 1 5 Lin, Ruiyue...... T P P . 2 2 Li, Mao...... F E P . 6 3 Lin, Shanming...... W P P . 3 2 Li, Mao WPP.55 Lin, Shanming W P P . 3 8 Li, Ming...... F E P . 2 0 Lin, Xin...... F P - 1 . 1 Li, Munan...... T M - 2 . 2 Lin, Xin WP-2.3

A-4

2009 IEEE ICAL AUTHOR INDEX

Lin, Xin WP-3.2 Liu, Yuanyi...... F P P . 1 9 Lin, Xin WP-3.3 Liu, Yukun...... F P P . 6 4 Lin, Zhiwei...... F P - 3 . 3 Liu, Yuliang...... F E P . 3 4 Liu, Ai-min...... F P - 1 . 1 Liu, Yuyan...... W P P . 1 0 Liu, Ai-min WP-2.3 Liu, Zaiwen...... F M - 2 . 3 Liu, Ai-min WP-3.2 Liu, Zhao-du...... F P P . 2 2 Liu, Baiyan...... F E P . 4 3 Liu, Zhigang...... W P P . 8 Liu, Cheng-yun...... FPP.3 Liu, Zhiyong...... W P P . 4 3 Liu, Chuanxi...... W P P . 3 Liu, Zhiyuan...... T P P . 6 7 Liu, Danni...... F M - 3 . 1 Long, Ruizheng...... W P - 3 . 1 Liu, Derong...... F P P . 3 3 Long, Zhiqiang...... F P P . 5 2 Liu, Derong WPP.11 Lu, Fei...... T P P . 4 0 Liu, Dongmei...... F M - 2 . 2 Lu, Jingli...... F E P . 6 6 Liu, Enfu...... F E P . 2 1 Lu, Shizhou...... T M - 2 . 1 Liu, Gang...... F P P . 6 1 Lu, Shizhou T M - 5 . 2 Liu, Guangjun...... F E P . 2 9 Lu, Shouyin...... F E P . 6 0 Liu, Guangwei...... T M - 2 . 4 Luo, Rui...... F E P . 3 0 Liu, Guojin...... W P P . 1 8 Luo, Wuming...... F P P . 5 1 Liu, Haifa...... T P P . 3 5 Luo, Zhong...... W P - 4 . 2 Liu, Hao...... W P P . 2 2 Lv, Changzhi...... T P P . 6 0 Liu, Hongyi...... W P - 4 . 2 Lv, Yanyun...... W P P . 3 2 Liu, Hui...... T P P . 6 2 Lv, Yuegang...... F P P . 5 0 Liu, Jianchang...... W P P . 5 4 Liu, Jianqiu...... T M - 5 . 1 Liu, Jinhai...... F P P . 3 3 M Liu, Jizhu...... W P - 3 . 1 Liu, Junyao...... W P P . 5 6 Ma, Bin...... T P P . 3 1 Liu, Junying...... T P P . 1 Ma, Hongwei...... F E P . 3 7 Liu, Junying W P P . 1 Ma, Jia...... F E P . 7 6 Liu, Kun...... T P P . 7 1 Ma, Ke...... F E P . 2 2 Liu, Lei...... T P P . 2 4 Ma, Ke TPP.13 Liu, Li...... F E P . 1 0 Ma, Ke TPP.14 Liu, Li F M - 2 . 4 Ma, Liling...... W M - 3 . 3 Liu, Linfang...... F E P . 6 8 Ma, Peibei...... F E P . 7 5 Liu, Longji...... F E P . 9 Ma, Qian...... F E P . 6 7 Liu, Peng...... W M - 2 . 5 Ma, Shugen...... W P - 1 . 4 Liu, Peng WP-5.3 Ma, Sile...... W P P . 3 9 Liu, Qing-song...... T P P . 1 9 Ma, Xiang-juan...... FPP.4 Liu, Qing-song W P P . 6 3 Ma, Xin...... F E P . 4 7 Liu, Rong...... W M - 3 . 2 Ma, Xiushui...... T P P . 6 8 Liu, Shuying...... F E P . 2 3 Ma, Yongwei...... T M - 4 . 1 Liu, Wei...... F E P . 6 5 Ma, Zi...... F E P . 3 8 Liu, Wei WPP.31 Ma, Zi WP-4.1 Liu, Weiguo...... W M - 2 . 3 Ma, Zi WPP.34 Liu, Xiaolei...... W P - 4 . 4 Mahgoub, Osama...... W M - 1 . 1 Liu, Xiaoping...... F P P . 4 0 Man, Zai-long...... T P P . 5 3 Liu, Xiaoxiong...... T M - 5 . 3 Mao, Wentao...... W M - 1 . 2 Liu, Xiaoxiong T P P . 2 3 Mao, Ying-chi...... F M - 4 . 5 Liu, Ximei...... T P P . 4 6 Maple, Carsten...... W P - 2 . 4 Liu, Xin-feng...... F E P . 5 Marino, Alessandro...... F E P . 4 5 Liu, Xinhua...... T P P . 5 9 Mei, Gaoqing...... T P P . 5 6 Liu, Xuan...... F E P . 6 5 Meng, Guang-yao...... F E P . 2 7 Liu, Xuan FEP.7 Meng, Jie...... W P P . 4 9 Liu, Yanbin...... F E P . 3 5 Meng, Lidong...... F M - 1 . 5 Liu, Yang...... W P P . 9 Meng, Liqing...... F E P . 6 8 Liu, Yanjun...... F P P . 5 9 Meng, Max Q.-h...... F P P . 3 1 Liu, Yanmei...... T P P . 3 8 Meng, Max Q.-h...... F E P . 6 3 Liu, Yingjie...... F P P . 2 5 Meng, Max Q.-h. W P - 1 . 2 Liu, Yong...... F E P . 7 3 Meng, Max Q.-h. W P P . 3 1 Liu, Yong TPP.16 Meng, Max Q.-h. W P P . 4 3

A-5

2009 IEEE ICAL AUTHOR INDEX

Meng, Max Q.-h. W P P . 5 5 Qi, Xiaomei...... T M - 3 . 1 Meng, Qinghao...... F P P . 1 4 Qi, Xiaomei W P - 2 . 2 Meng, Qingjin...... W P P . 2 8 Qi, Xinyue...... F E P . 5 2 Meng, Tianxing...... F P P . 6 8 Qi, Zhiyong...... T P P . 3 8 Meng, Xiangping...... T P P . 4 4 Qian, Suxiang...... T P P . 1 9 Meng, Xiang-ping...... T P P . 4 3 Qian, Wang...... F P P . 1 0 Meng, Zhaohe...... T M - 2 . 4 Qiao, Fei...... F M - 5 . 3 Min, Dequan...... T M - 5 . 1 Qiao, Jiaxin...... W P P . 2 5 Ming, Aiguo...... T M - 1 . 5 Qiao, Jiaxin W P P . 2 6 Ming, Zhong...... F M - 2 . 1 Qin, Long...... F E P . 6 8 Mu, Lili...... T P P . 3 0 Qin, Yining...... W P P . 6 9 Qin, Zheng...... T P P . 7 Qing, Xiao...... T P P . 2 6 N Qiu, Chuanfei...... F P P . 3 7 Qiu, Hua...... W P - 2 . 4 Nakayama, Daishi...... W P - 2 . 5 Quan, Ling...... T P P . 6 4 Nie, Rui...... T M - 5 . 3 Ning, Zhong...... F E P . 2 6 Niu, Dongjie...... F M - 2 . 5 R Nobakhti, Amin...... T M - 4 . 4 Radaschin, Adrian...... F E P . 5 0 Rakheja, Subhash...... W M - 4 . 4 O Reif, Wolfgang...... W M - 1 . 4 Ren, Beibei...... W M - 4 . 2 Ortmeier, Frank...... W M - 1 . 4 Ren, Honge...... F P P . 5 3 Ou, Xunmin...... W P P . 6 9 Ren, Honge WPP.44 Oyekan, John...... T M - 5 . 5 Ren, Huizhi...... W P P . 5 3 Ozkil, Ali Gurcan...... T M - 1 . 4 Ren, Lipeng...... F P P . 3 9 Ren, Shengle...... W P - 4 . 3 Ren, Zhang...... T M - 3 . 4 P Rocca, Jesse...... W M - 4 . 5 Roman, Monica...... F P P . 3 2 Pan, Changbo...... W P P . 6 2 Roman, Monica FPP.56 Pan, Haipeng...... W P P . 1 6 Roy, Debanik...... F E P . 3 3 Pan, Xiuqin...... F P P . 4 3 Pan, Yingzi...... W P P . 4 3 S Pan, Yongping...... W P P . 6 5 Pang, Yujun...... F P P . 6 3 Park, Jong Hyeon...... W M - 1 . 5 Schetinin, Vitaly...... W P - 2 . 4 Pei, Hailong...... W M - 4 . 2 Selisteanu, Dan...... F P P . 3 2 Pei, Run...... T P P . 6 7 Selisteanu, Dan F P P . 5 6 Pei, Wei...... T P P . 5 3 Sendrescu, Dorin...... F P P . 3 2 Petre, Emil...... F P P . 3 2 Sendrescu, Dorin F P P . 5 6 Petre, Emil F P P . 5 6 Serbencu, Adrian Emanoil...... F E P . 5 8 Piao, Huishu...... F P P . 6 2 Serbencu, Adriana...... F E P . 5 0 Popescu, Dorin...... F P P . 1 4 Serbencu, Adriana F E P . 5 8 Popescu, Dorin F P P . 5 6 Shafii, Nima...... F E P . 6 1 Popescu, Livia...... F P P . 1 4 Shan, Yong-zheng...... W P - 1 . 3 Poungponsri, Suranai...... T M - 5 . 4 Shang, Lan...... F E P . 2 9 Pu, Guoliang...... W P P . 4 7 Shao, Keyong...... W P P . 2 7 Pu, Jiexin...... F E P . 1 1 Shao, Qiang...... F P P . 3 0 Shao, Qiang F P P . 4 4 Shen, Chenglin...... F E P . 2 2 Q Shen, Chenglin T P P . 1 3 Shen, Chenglin T P P . 1 4 Qi, Dawei...... T M - 2 . 3 Shen, Tao...... T P P . 3 7 Qi, Donglian...... W P - 3 . 5 Shen, Wei...... F P P . 3 8 Qi, Jilong...... F P - 1 . 2 Shen, Yanjun...... W P P . 1 9 Qi, Shemiao...... F P - 2 . 1 Shi, Dawei...... W M - 3 . 3 Qi, Wei...... F E P . 6 0 Shi, Jiying...... F P P . 2 1

A-6

2009 IEEE ICAL AUTHOR INDEX

Shi, Pu...... F E P . 3 9 Tang, Jun...... W P P . 3 2 Shi, Shuangshuang...... F P P . 2 4 Tang, Lixin...... F E P . 3 0 Shi, Xianjiang...... F P P . 2 3 Tang, Lixin F E P . 7 1 Shi, Yuntao...... F P - 4 . 3 Tang, Lixin T M - 3 . 5 Shimojo, Makoto...... T M - 1 . 5 Tang, Lixin W P - 5 . 4 Shu, Fangwu...... F M - 3 . 3 Tang, Lixin W P - 5 . 5 Si, Hai...... F P P . 2 2 Tang, Lixin W P P . 5 3 Soliman, Ayman...... W M - 1 . 1 Tang, Qi...... T P P . 2 7 Son, Seung-woo...... W M - 1 . 5 Tao, Ze...... F M - 5 . 5 Song, Chuan-xue...... T P P . 5 1 Teng, Fei...... W P P . 6 2 Song, Rui...... F E P . 4 7 Teng, Yongdong...... F M - 4 . 1 Song, Shuang...... F E P . 6 3 Teo, Kok Lay...... W P P . 1 2 Song, Shuang W P P . 5 5 Tian, Guanglai...... W M - 2 . 3 Song, Xin...... T P P . 4 2 Tian, Guohui...... T P P . 1 0 Song, Yan...... W P P . 2 9 Tian, Hao...... F E P . 3 7 Song, Yeqiong...... F M - 4 . 2 Tian, Jingwen...... F E P . 1 5 Song, Zhangjun...... F E P . 4 1 Tian, Jingwen F P P . 1 5 Song, Zhangjun T P P . 9 Tian, Lei...... F P P . 6 5 Su, Chun-yi...... W M - 4 . 4 Tian, Longqiang...... F P P . 3 1 Su, Hongye...... T M - 4 . 1 Tian, Sujie...... F E P . 1 8 Su, Hongye T M - 4 . 2 Tian, Xiaoyu...... F M - 5 . 3 Su, Yanyu...... T M - 5 . 2 Tian, Yantao...... F E P . 4 0 Su, Zhifang...... W P P . 1 8 Tian, Yantao FEP.52 Su, Zhiyuan...... F P P . 6 4 Tian, Yantao F P - 1 . 2 Sui, Qingmei...... T P P . 3 2 Tian, Yantao T P P . 1 1 Sui, Wentao...... F E P . 2 8 Tian, Yantao W P P . 2 9 Sun, Dehui...... F P - 4 . 3 Tian, Yuan...... F E P . 3 8 Sun, Jianmin...... F E P . 4 2 Tian, Yuan T M - 3 . 4 Sun, Jianmin F E P . 6 4 Tong, Chang-fei...... W P P . 5 Sun, Jianmin F P P . 4 7 Tong, Shaocheng...... T P P . 6 5 Sun, Jianmin T P P . 5 5 Sun, Jing...... F M - 2 . 5 Sun, Limin...... F E P . 1 6 V Sun, Qiqi...... W M - 5 . 1 Sun, Qiuye...... T P P . 7 0 Vasiliu, Grigore...... F E P . 5 0 Sun, Shibao...... F E P . 1 1 Vistein, Michael...... W M - 1 . 4 Sun, Shuli...... F P - 1 . 4 Sun, Tongjing...... T P P . 5 6 Sun, Xigang...... F P P . 7 1 W Sun, Yafei...... F M - 2 . 5 Sun, Yang...... F E P . 1 5 Wang, Bin...... F E P . 1 0 Sun, Yingfei...... W P - 4 . 4 Wang, Bin TPP.42 Sun, Yong...... T P P . 2 8 Wang, Bing...... F M - 5 . 4 Sun, Yufeng...... T P P . 6 1 Wang, Changjun...... F E P . 7 3 Sun, Zhendong...... W M - 4 . 2 Wang, Changjun F E P . 7 4 Sun, Zhengfeng...... F P P . 5 7 Wang, Changjun T P P . 1 6 Sun, Zonghai...... W P P . 6 5 Wang, Chengshan...... F P P . 2 1 Susnea, Ioan...... F E P . 5 0 Wang, Chiyu...... F P P . 6 2 Susnea, Ioan F E P . 5 8 Wang, Congzhe...... T P P . 6 Wang, Cundong...... T M - 3 . 1 T Wang, Dan...... T P P . 1 7 Wang, Dejun...... W P P . 2 9 Wang, Dong...... F M - 3 . 4 Tan, Boxue...... F M - 1 . 5 Wang, Dongmei...... T P P . 6 3 Tan, Min...... W M - 2 . 2 Wang, Fei...... W P - 4 . 2 Tan, Min WP-1.1 Wang, Feng...... W P P . 5 Tan, Ning...... F P - 2 . 2 Wang, Fengying...... F E P . 1 4 Tang, Jian...... FPP.2 Wang, Hong...... T M - 4 . 4 Tang, Jianxun...... F E P . 7 1 Wang, Hongrui...... T P P . 1 1 Tang, Jin...... F M - 3 . 5 Wang, Huajie...... W P P . 3 9

A-7

2009 IEEE ICAL AUTHOR INDEX

Wang, Hui...... T P P . 4 3 Wang, Yuechao...... F E P . 5 4 Wang, Hui TPP.44 Wang, Yuechao W P - 1 . 4 Wang, Huijing...... W P P . 5 6 Wang, Yukun...... F P P . 4 1 Wang, Jian...... W P - 4 . 5 Wang, Yunpeng...... FPP.8 Wang, Jianlu...... T P P . 4 7 Wang, Zhenchao...... FPP.1 Wang, Jiesheng...... F P P . 4 1 Wang, Zhenchao F P P . 4 8 Wang, Jin...... T P P . 4 3 Wang, Zhengjun...... F P P . 3 8 Wang, Jing...... FPP.1 Wang, Zhenhua...... F M - 4 . 4 Wang, Jing T P P . 5 3 Wang, Zuoquan...... T M - 1 . 1 Wang, Jingguo...... W P - 1 . 5 Wei, Fen...... T P P . 1 8 Wang, Jinkuan...... F E P . 1 0 Wei, Hanbing...... W M - 2 . 3 Wang, Jinkuan T P P . 4 2 Wei, Honglei...... F M - 3 . 1 Wang, Jintian...... F M - 1 . 4 Wei, Jie...... F E P . 1 6 Wang, Juan...... W P P . 1 7 Wei, Jie FEP.67 Wang, Jun...... F P P . 6 3 Wei, Junxiu...... W P P . 9 Wang, Jun W P P . 2 1 Wei, Lifeng...... W P P . 2 1 Wang, Junyi...... FPP.7 Wei, Qiang...... W M - 3 . 1 Wang, Junzheng...... F P P . 3 8 Wei, Yan-di...... F E P . 5 Wang, Junzheng W M - 3 . 3 Wei, Yingzi...... F M - 5 . 2 Wang, Junzheng W P P . 8 Wen, Shengjun...... T M - 4 . 5 Wang, Kejun...... W P P . 3 7 Wen, Zhen Hua...... W P - 5 . 1 Wang, Keqi...... W P P . 1 5 Won, Yong-tae...... F M - 3 . 2 Wang, Lei...... W P P . 4 2 Wu, Bin...... F P - 2 . 4 Wang, Lianchun...... F P P . 5 2 Wu, Chengdong...... F P - 4 . 4 Wang, Liming...... F E P . 7 4 Wu, Chuanyu...... W P P . 4 8 Wang, Liping...... T P P . 3 Wu, Lingfei...... W P - 1 . 2 Wang, Lujia...... F P P . 3 1 Wu, Peiliang...... F E P . 3 2 Wang, Minghui...... W P - 1 . 4 Wu, Qidi...... F M - 5 . 3 Wang, Ruiyun...... W P P . 2 7 Wu, Qingtao...... F E P . 1 1 Wang, Shoukun...... F P P . 3 8 Wu, Wei-yang...... FPP.4 Wang, Shuqing...... T P P . 6 2 Wu, Xiao...... F E P . 3 6 Wang, Shuyun...... F E P . 6 9 Wu, Yanrong...... T P P . 3 3 Wang, Song...... F P P . 2 4 Wu, Yaohua...... W M - 2 . 5 Wang, Tao...... F P - 3 . 1 Wu, Zhijun...... T P P . 1 5 Wang, Wanshan...... F P P . 2 6 Wang, Wei...... F E P . 1 8 Wang, Wei T P P . 2 9 X Wang, Weibing...... W P - 1 . 1 Wang, Weiqing...... T P P . 4 1 Xi, Ning...... F E P . 5 4 Wang, Weizhi...... F P - 3 . 2 Xi, Peiyu...... F P P . 5 0 Wang, Wen...... F E P . 2 Xia, Junjian...... F E P . 1 6 Wang, Wen F M - 2 . 2 Xia, Wei...... F M - 4 . 5 Wang, Wen F P P . 3 9 Xia, Weiliang...... W P P . 4 1 Wang, Wenbiao...... F E P . 3 8 Xiang, Guishan...... W P P . 4 5 Wang, Wencheng...... W P P . 4 2 Xiang, Zhengtao...... F E P . 1 3 Wang, Wilson...... F E P . 2 8 Xiao, Guanghui...... F P P . 2 3 Wang, Xiaofeng...... F E P . 7 3 Xiao, Hairong...... F P P . 1 6 Wang, Xiaonan...... F E P . 1 2 Xiao, Hairong W P P . 1 3 Wang, Xiaonan F E P . 6 Xiao, Longteng...... F E P . 3 4 Wang, Xiaoyi...... F M - 2 . 3 Xiao, Nan...... F E P . 5 5 Wang, Xiao-yu...... W P - 3 . 3 Xiao, Shuomei...... F E P . 5 9 Wang, Xingping...... W M - 3 . 4 Xiao, Tianyuan...... W M - 5 . 3 Wang, Xiuping...... T P P . 6 9 Xiao, Wendong...... F P - 1 . 4 Wang, Xu...... W P P . 1 4 Xiao, Yunkui...... F E P . 3 6 Wang, Xuguang...... W P - 5 . 2 Xiao, Zu...... W P P . 2 4 Wang, Yansong...... F P P . 3 4 Xie, Lihua...... F P - 1 . 4 Wang, Yijin...... FPP.1 Xie, Wen-fang...... W M - 4 . 3 Wang, Yong...... F P P . 5 9 Xie, Xiangpeng...... W P P . 1 1 Wang, Yu...... W P P . 5 4 Xie, Xiuzhen...... F M - 3 . 5 Wang, Yuanhui...... T P P . 6 1 Xie, Yudong...... F P P . 5 9

A-8

2009 IEEE ICAL AUTHOR INDEX

Xie, Zhaoxian...... T M - 1 . 5 Yang, Guangtao...... T M - 4 . 1 Xin, Liming...... FPP.7 Yang, Hao...... W M - 3 . 2 Xin, Tian...... F E P . 2 0 Yang, Huayong...... F P P . 2 5 Xing, Feng...... F P P . 3 5 Yang, Jianfu...... F P P . 1 7 Xu, Bing...... F P P . 2 5 Yang, Jianfu FPP.5 Xu, Changxin...... W M - 5 . 1 Yang, Jianyu...... F P P . 2 6 Xu, Degang...... W P P . 1 8 Yang, Jie...... W P - 3 . 5 Xu, Hui...... T P P . 5 1 Yang, Jingming...... W P P . 6 4 Xu, Huipu...... F E P . 3 8 Yang, Lihua...... F P - 2 . 1 Xu, Huipu W P - 4 . 1 Yang, Liu...... T P P . 5 0 Xu, Huipu W P P . 3 4 Yang, Mao...... F E P . 5 2 Xu, Jian...... T P P . 3 1 Yang, Qinghai...... W M - 2 . 2 Xu, Jian-guang...... FPP.3 Yang, Qingmei...... F E P . 4 2 Xu, Jianyuan...... W P - 2 . 3 Yang, Qingmei FEP.64 Xu, Jian-yuan...... W P - 3 . 3 Yang, Qingmei FPP.47 Xu, Jing...... F E P . 4 1 Yang, Qingmei TPP.55 Xu, Jing T P P . 9 Yang, Simon X...... W M - 4 . 5 Xu, Jun...... F M - 4 . 3 Yang, Wanan...... F E P . 6 3 Xu, Liyou...... W P - 2 . 1 Yang, Wei...... F E P . 2 Xu, Na...... W M - 2 . 5 Yang, Wei FPP.39 Xu, Na WP-5.3 Yang, Weifeng...... T P P . 1 5 Xu, Peng...... F E P . 1 0 Yang, Xia...... F P P . 3 5 Xu, Peng FP-2.2 Yang, Xia WPP.24 Xu, Rongdong...... T P P . 2 1 Yang, Zhaosheng...... T P P . 2 5 Xu, Shaoping...... F P P . 4 0 Yang, Zhaosheng W P P . 6 7 Xu, Shiyan...... F E P . 2 4 Yasuda, Gen'ichi...... F E P . 3 1 Xu, Xiaona...... F P P . 4 3 Yeon, Je Sung...... W M - 1 . 5 Xu, Xinhe...... T P P . 3 8 Yi, Jinhui...... W P P . 9 Xu, Xuelian...... F E P . 3 4 Yin, Yixin...... F P - 1 . 3 Xu, Xun...... F E P . 2 6 Ying, Qin...... F P P . 6 7 Xu, Yanhong...... T P P . 7 1 You, Tao...... F P - 2 . 3 Xu, Ye-er...... F E P . 1 9 You, Tianyu...... F P P . 2 8 Xu, Zhigang...... FPP.7 You, Xianqiang...... F P - 4 . 1 Xu, Zili...... T P P . 2 1 Yu, Chengxin...... W P - 5 . 3 Xue, Dingyu...... T P P . 3 8 Yu, Dongmei...... W P P . 4 Xue, Jian...... F P - 1 . 3 Yu, Enlin...... F P P . 6 6 Xue, Jing...... W M - 2 . 3 Yu, Haisheng...... W P P . 6 Xue, Shixu...... T M - 3 . 5 Yu, Haiyan...... T P P . 4 9 Xue, Xiao...... W M - 5 . 3 Yu, Haiyan W P P . 2 Xue, Yinghua...... T P P . 1 0 Yu, Hiayan...... W P P . 2 2 Yu, Hongliang...... T P P . 6 7 Yu, Jianjun...... W P P . 3 5 Y Yu, Jinshou...... F M - 1 . 2 Yu, Junzhi...... W M - 2 . 2 Yamaguchi, Hisashi...... T M - 1 . 5 Yu, Junzhi W P - 1 . 1 Yamashiro, Manabu...... T M - 1 . 5 Yu, Lanlan...... F P P . 6 8 Yan, Dongmei...... F E P . 1 0 Yu, Lei...... T M - 2 . 3 Yan, Jihong...... F E P . 4 3 Yu, Lie...... F P - 2 . 1 Yan, Jihong T M - 4 . 3 Yu, Lie TPP.21 Yan, Ming...... W P P . 9 Yu, Ruiling...... W P P . 7 2 Yan, Ping...... W M - 2 . 4 Yu, Shanshan...... W P P . 6 Yan, Ping W P P . 3 8 Yu, Shaoqiang...... T P P . 1 7 Yan, Shirong...... F P - 3 . 3 Yu, Xian...... F M - 4 . 5 Yan, Tao...... F P P . 6 6 Yu, Xiao-hua...... T M - 5 . 4 Yan, Wang...... W M - 5 . 1 Yu, Xuefeng...... T P P . 3 0 Yan, Yaxin...... T P P . 5 9 Yu, Yan...... F P P . 2 8 Yang, Baosheng...... T P P . 6 8 Yu, Yong...... W P P . 5 0 Yang, Chenguang...... W M - 3 . 5 Yu, Ze...... T P P . 1 2 Yang, Chuan...... W P - 2 . 3 Yuan, Qin...... T P P . 6 4 Yang, Genghuang...... F E P . 3 4 Yuan, Weiqi...... W P P . 5 7

A-9

2009 IEEE ICAL AUTHOR INDEX

Yuan, Zhugang...... T P P . 6 7 Zhang, Mingshu...... T P P . 2 1 Yue, Yong...... W P - 2 . 4 Zhang, Mingzhu...... W P - 2 . 1 Zhang, Niaona...... T P P . 5 7 Zhang, Qing...... W P P . 6 4 Z Zhang, Shibing...... F P P . 4 8 Zhang, Shu-chu...... F M - 5 . 1 Zadorozny, Andrzej...... W P P . 3 6 Zhang, Shumei...... W P P . 2 9 Zaki, Aziza...... W M - 1 . 1 Zhang, Songmei...... F E P . 9 Zang, Faye...... W P P . 2 0 Zhang, Tao...... W P - 4 . 5 Zang, Huaiquan...... F P P . 1 3 Zhang, Taogai...... F P P . 7 2 Zang, Huai-quan...... F P P . 7 0 Zhang, Tianlong...... T P P . 7 0 Zapata, Rene...... F E P . 4 4 Zhang, Tieyan...... T P P . 7 0 Ze, Xiangbo...... F P P . 5 8 Zhang, Ting...... F E P . 5 9 Zeng, Dingrong...... F P P . 2 5 Zhang, Wei...... T P P . 4 5 Zeng, Liangcai...... F E P . 4 3 Zhang, Weiguo...... F M - 4 . 4 Zeng, Zenxing...... W P P . 4 3 Zhang, Weiguo T M - 5 . 3 Zeng, Zhoumo...... F E P . 3 6 Zhang, Weiguo T P P . 2 8 Zhai, Lianfei...... W M - 3 . 5 Zhang, Xianku...... W M - 3 . 4 Zhai, Xudong...... W M - 3 . 2 Zhang, Xiaojuan...... F P P . 6 1 Zhan, Wenzhang...... F P P . 2 2 Zhang, Xiao-ling...... F P - 1 . 1 Zhang, Bingyi...... F P P . 4 5 Zhang, Xiaowei...... T P P . 3 0 Zhang, Chao...... T P P . 4 6 Zhang, Xiaoyun...... T P P . 6 1 Zhang, Dan...... F E P . 2 8 Zhang, Xin...... T P P . 4 9 Zhang, Dawei...... F E P . 3 Zhang, Xin W P P . 2 Zhang, Dengpan...... T M - 3 . 3 Zhang, Xin W P P . 2 2 Zhang, Dingtong...... FPP.8 Zhang, Xinxin...... F E P . 2 2 Zhang, Dong...... W M - 3 . 1 Zhang, Xinxin T P P . 1 3 Zhang, Fei...... T P P . 4 7 Zhang, Xinxin T P P . 1 4 Zhang, Feng...... W P - 1 . 3 Zhang, Xinyan...... T P P . 4 1 Zhang, Fengge...... T M - 2 . 4 Zhang, Xueping...... F P P . 7 2 Zhang, Fengge T P P . 6 9 Zhang, Xuewu...... W P P . 3 2 Zhang, Fengge W P - 3 . 4 Zhang, Xuewu W P P . 4 0 Zhang, Guanying...... T P P . 7 1 Zhang, Xue-wu...... W P P . 4 1 Zhang, Guoshan...... T P P . 2 4 Zhang, Xutang...... T P P . 5 9 Zhang, Guoxian...... T M - 1 . 2 Zhang, Yang...... F P - 4 . 1 Zhang, Haijun...... F P P . 7 3 Zhang, Yanwei...... T M - 2 . 5 Zhang, Hong...... T P P . 4 3 Zhang, Yingwei...... F M - 4 . 1 Zhang, Hong T P P . 4 4 Zhang, Yingwei F P - 4 . 1 Zhang, Hong W P P . 3 6 Zhang, Yishu...... W P P . 2 7 Zhang, Hongmei...... F P P . 7 2 Zhang, Yong...... F E P . 5 Zhang, Hua...... F P P . 4 0 Zhang, Yong F P P . 4 1 Zhang, Huaguang...... F P P . 3 3 Zhang, Yong F P P . 4 2 Zhang, Huaguang W P P . 1 1 Zhang, Yong-feng...... F E P . 5 Zhang, Hui...... F P P . 5 8 Zhang, Yu...... T P P . 3 6 Zhang, Jian...... T P P . 6 5 Zhang, Yu W M - 2 . 1 Zhang, Jian W P P . 4 4 Zhang, Yuan...... FPP.4 Zhang, Jianghua...... F M - 1 . 3 Zhang, Yunong...... F P - 2 . 2 Zhang, Jianwei...... F M - 3 . 3 Zhang, Yunzhou...... F P - 4 . 4 Zhang, Jiwei...... W P - 5 . 2 Zhang, Yu-zhu...... T P P . 4 5 Zhang, Jujian...... FPP.6 Zhang, Ze ...... T P P . 7 1 Zhang, Jun...... W P P . 1 4 Zhang, Zhihuan...... W M - 1 . 3 Zhang, Lei...... F E P . 4 4 Zhang, Zhihui...... T P P . 4 0 Zhang, Lei W P - 5 . 3 Zhang, Zhijia...... W P P . 5 7 Zhang, Lei W P P . 4 8 Zhang, Zipeng...... T P P . 6 2 Zhang, Lifu...... T P P . 3 4 Zhao, Chao...... T M - 4 . 2 Zhang, Lili...... F P P . 2 1 Zhao, Fang...... F P P . 5 8 Zhang, Liu...... F P P . 4 9 Zhao, Fei...... W P P . 4 Zhang, Meng...... T P P . 2 8 Zhao, Guangzhou...... W P - 3 . 5 Zhang, Meng W P P . 5 2 Zhao, Guohui...... W P P . 1 2 Zhang, Minglu...... F E P . 6 5 Zhao, Jiangbo...... W M - 3 . 3

A-10

2009 IEEE ICAL AUTHOR INDEX

Zhao, Jiangbo W P P . 8 Zhu, Chunxia...... T P P . 3 Zhao, Jianwen...... T M - 2 . 1 Zhu, Chunxia T P P . 4 Zhao, Jianwen T M - 5 . 2 Zhu, Daoli...... F M - 1 . 3 Zhao, Jiao...... W P - 5 . 5 Zhu, Feibai...... T M - 2 . 1 Zhao, Jie...... F E P . 4 3 Zhu, Huajin...... T M - 1 . 2 Zhao, Jie F E P . 7 Zhu, Jing...... F P P . 4 2 Zhao, Jing...... F E P . 6 7 Zhu, Lida...... T P P . 4 Zhao, Jun...... F E P . 1 8 Zhu, Ming...... W P P . 4 6 Zhao, Lei...... F E P . 1 4 Zhu, Xiaoyan...... W P P . 3 8 Zhao, Lei T P P . 5 8 Zhu, Xuejun...... F M - 1 . 2 Zhao, Leilei...... F P P . 2 0 Zhu, Yi-an...... F P - 2 . 3 Zhao, Li...... F E P . 3 4 Zhu, Yongying...... T P P . 5 3 Zhao, Liang...... T P P . 4 4 Zhu, Zongming...... T M - 2 . 5 Zhao, Nan...... F P P . 3 5 Zhu, Zuchao...... W P P . 4 8 Zhao, Xiaofeng...... T P P . 2 7 Zong, Chang-fu...... T P P . 5 1 Zhao, Xingyu...... F E P . 3 Zou, Zongwei...... W P P . 6 Zhao, Xinhua...... F E P . 3 5 Zu, Yongchang...... W P P . 6 7 Zhao, Xinjie...... F P P . 2 3 Zuo, Zhiqiang...... T P P . 2 4 Zhao, Xu...... F E P . 1 9 Zhao, Xueping...... F P - 1 . 2 Zhao, Yan...... T P P . 7 0 Zhao, Yan W P P . 2 4 Zhao, Yiwen...... F E P . 3 9 Zhao, Yong-guo...... F P P . 1 8 Zhao, Yue...... F P P . 4 3 Zhao, Yuqian...... F P P . 4 8 Zhao, Yuqian F P P . 6 6 Zhao, Zengshun...... F P P . 5 7 Zhao, Zhanwei...... FPP.9 Zhen, Jingjing...... F P - 4 . 3 Zheng, Rui...... F E P . 1 7 Zheng, Ruijuan...... F E P . 1 1 Zheng, Weifeng...... W P - 3 . 1 Zheng, Zhiyun...... W P P . 4 9 Zhi, Xuejun...... F E P . 7 7 Zhong, Ming...... T M - 5 . 1 Zhong, Qiubo...... F P P . 5 3 Zhong, Weikun...... W P P . 3 0 Zhong, Zhiguang...... F P P . 6 0 Zhou, Changcheng...... F P P . 1 9 Zhou, Changcheng F P P . 2 0 Zhou, Changcheng W P P . 4 9 Zhou, Changjiu...... F E P . 4 0 Zhou, Di...... W P P . 1 2 Zhou, Dong...... F P P . 5 5 Zhou, Fengyu...... F P P . 1 6 Zhou, Fengyu W P P . 1 3 Zhou, Jian...... T P P . 5 8 Zhou, Jianhua...... W P P . 3 8 Zhou, Jingyang...... W P P . 1 2 Zhou, Lei...... F M - 1 . 4 Zhou, Luanjie...... W P P . 2 7 Zhou, Qiang...... T M - 2 . 5 Zhou, Shiru...... F P P . 1 5 Zhou, Xiaoguang...... F P P . 6 4 Zhou, Xiaoqiang...... W M - 5 . 5 Zhou, Xin-guo...... T P P . 3 1 Zhou, Zhili...... W P - 2 . 1 Zhu, Bao-yi...... F E P . 1 9 Zhu, Cheng...... F P - 1 . 2

A-11

IEEE ICAL 2009 Program at a Glance Time \ Room HQHT-304 HQHT-308 HQHT-312 HQHT-313 HQHT-314 Tuesday, August 4, 2009 14:00 – 20:00 Registration at Sheraton Hotel Shenyang Wednesday, August 5, 2009 08:30 – 09:00 Opening Ceremony at HQHT Auditorium – Chair: Tzyh-Jong Tarn Plenary Talk I (HQHT Auditorium) – Chair: Toshio Fukuda 09:00 – 10:00 Resource Control in Stochastic Processing Networks David D. Yao, Columbia University, USA 10:00 – 10:20 Tea and Coffee Break WM-1 WM-2 WM-3 WM-4 WM-5 10:20 – 12:00 Robotics I Modeling I Automatic Control I Intelligent Control Supply Chain and Automation I Management 12:00 – 14:00 Lunch Break (University Town Restaurant buffet lunch) WP-1 WP-2 WP-3 WP-4 WP-5 14:00 – 15:40 Robotics II Modeling II Automatic Control II Intelligent Control Logistics Planning and Automation II 15:40 – 16:00 Tea and Coffee Break 15:40 – 16:40 Poster Session WPP (HQHT): Automatic Control I Plenary Talk II (HQHT Auditorium) – Chair: Max Q.-H. Meng 16:40 – 17:40 iCART: Intelligent Cooperative Autonomous Robot Transporters Kazuhiro Kosuge, Tohoku University, Japan 18:00 – 20:30 Welcome Dinner at Xin Hong Ji Restaurant (新洪记饺子馆) Thursday, August 6, 2009 Plenary Talk III (HQHT Auditorium) – Chair: Tianyou Chai 09:00 – 10:00 Supernetwork and its Applications Zhongtuo Wang, Dalian University of Science and Technology, China 10:00 – 10:20 Tea and Coffee Break TM-1 TM-2 TM-3 TM-4 TM-5 10:20 – 12:00 Robotics III Modeling III Distribution Systems Process Automation Modeling and Applications 12:00 – 14:00 Lunch Break (University Town Restaurant buffet lunch) Plenary Talk IV (HQHT Auditorium) – Chair: Derong Liu 14:00 – 15:00 Approximate Dynamic Programming for High-Dimensional Resource Allocation Problems Warren B. Powell, Princeton University, USA 15:00 – 16:00 Poster Session TPP (HQHT) 15:40 – 16:00 Tea and Coffee Break 16:00 – 17:00 NEU 111 Project Panel Automation and Logistics Forum – Chairs: Tianyou Chai and Tzyh-Jong Tarn Plenary Talk V (HQHT Auditorium) – Chair: Huaguang Zhang 17:00 – 18:00 Automatic Fabrication of Ultraprecision Microparts with Complicated Shape Yoshimi Takeuchi, Osaka University, Japan 18:00 – 21:00 Award Banquet with Local Performance in Banquet Hall of Liaoning Tower (辽宁大厦宴会厅) Friday, August 7, 2009 Plenary Talk VI (HQHT Auditorium) – Chair: Jason J. Gu 09:00 – 10:00 Modeling, Scheduling and Control of Cluster Tools in Semiconductor Manufacturing Automation Mengchu Zhou, New Jersey Institute of Technology, USA 10:00 – 10:20 Tea and Coffee Break FM-1 FM-2 FM-3 FM-4 FM-5 10:20 – 12:00 Optimization I Simulation I Image Processing Network-based Planning and Systems I Scheduling 12:00 – 14:00 Lunch Break (University Town Restaurant buffet lunch) FP-1 FP-2 FP-3 FP-4 14:00 – 15:20 Optimization II Simulation II Intelligent Network-based - Transportation Systems Systems II 15:20 – 15:40 Tea and Coffee Break 15:40 – 16:40 Poster Session FPP (HQHT): Intelligent Transportation System 16:40 – 17:40 Poster Session FEP (HQHT): Modeling, Identification and Simulation 18:00 – 20:30 Farewell Party at Sheraton Hotel Shenyang (沈阳丽都喜来登大酒店) Saturday, August 8, 2009 09:00 – 17:00 Local tour of Shenyang attractions