INTELLIGENT TRANSPORTATION SYSTEMS http://www.ieee.org/its

IEEE ITS SOCIETY NEWSLETTER Editor: Prof. Bart van Arem, [email protected] Vol. 7, No. 3, September 2005 ITSC Executive Committee ————– President: Charles J. Herget, [email protected] President-Elect 2005: Fei-Yue Wang, In This Issue ... [email protected] Society News 3 Vice President Conferences: Paul Kostek, . . . . [email protected] From the Editor ...... 3 Vice President Publications: Message from the IEEE ITS Society President ...... 3 Emily Sopensky, [email protected] Calendar of Society Events ...... 4 Vice President Finance: Message VP Conferences ...... 5 William Scherer, [email protected] Message VP Publications ...... 7 Vice President Technical Activities: Last words from a VP Technical Activities ...... 9 Daniel Zeng, Message VP Technical Activities ...... 10 ... [email protected] 2005 IEEE Annual Election ...... 11 Vice President for Adminstrative Ac- tivities: Bookreview ...... 14 Daniel J. Dailey, . [email protected] IEEE Trans. on ITS Report ...... 15 ————– IEEE Transactions on ITS - Index ...... 17 Transactions Editor: Alberto Broggi, . . [email protected] Conference Reports 20 Newsletter Editor: Bart van Arem, Technical Contributions 28 ...... [email protected]

Non-Society ITS News 36 A Glimpse on the Web ...... 36 Information for contributors ————– Results of AHS demonstration experiment ...... 37 Announcements, feature articles, books Upcoming Conferences, Workshops and Symposia ..... 44 and meetings reviews, opinions, letters to the editor, professional activities, ab- stracts of reports, and other material Web Archive of interest to the ITS community is so- All past issues of this Newsletter can be reached through the licited. Society’s Official Web Site at: http://www.ieee.org/its Please submit electronic material for consideration in any of the fol- lowing formats: LATEX, plain ASCII, Electronic Newsletter Subscription PDF, or Word, to the Editor at To obtain a free short announcement in your e-mail as soon as [email protected] at least 1 month the next Newsletter issue is available, please sign in through the prior to the newsletter’s distribution: Society Web Site at: http://www.ieee.org/its Issue Due date March February 1st June May 1st September August 1st December November 1st

Permission to copy without fee all or part of any material without a copyright notice is granted provided that the copies are not made or distributed for direct commercial advantage, and the title of the publication and its date appear on each copy. To copy material with a copyright notice requires specific permission. Please direct all inquiries or requests to IEEE Copyrights Office.

1 THE IEEE INTELLIGENT TRANSPORTATION SYSTEMS SOCIETY —————————————— President: ...... Charles J. Herget, Alameda, CA 94502, USA President-Elect 2005: ...... Fei-Yue Wang, CAS, China and U. of Arizona, Tucson, AZ 85721, USA Vice President Conferences: ...... Paul Kostek, Seattle, WA 98103, USA Vice President Publications: ...... Emily Sopensky, The Iris Company, Arlington, VA 22207, USA Vice President Finance: ..... William Scherer, University of Virginia, Charlottesville, VA 22904. USA Vice President Technical Activities: ...... Daniel Zeng, University of Arizona, Tucson, AZ 85721, USA Vice President Administrative Activities: ...... Daniel J. Daily, University of Washington, Seattle, WA 98195, USA Transactions Editor: ...... Alberto Broggi, Universit`adi Parma, Parma, I-43100, Italy Newsletter Editor: ... Bart van Arem, University of Twente, Enschede, NL-7500AE, The Netherlands

COMMITTEES

Awards Committee: Chip White (Chair): ...... [email protected] Conferences and Meetings Committee: Paul Kostek (Chair): ...... [email protected] Constitution and Bylaws Committee: Fei-Yue Wang (Chair): ...... [email protected] Fellow Evaluation Committee: Robert Fenton (Chair): ...... [email protected] Finance Committee: Bill Scherer (Chair): ...... [email protected] History Committee: E. Ryerson Case (Chair): ...... [email protected] Long Range Planning Committee: Pitu B. Mirchandani (Chair): ...... [email protected] Nominations and Appointments Committee: Fei-Yue Wang (Chair): . . . . . [email protected] Publications Committee: Emily Sopensky (Chair): ...... [email protected] Standards Committee: Jason Geng (Chair): ...... [email protected] Student Activities Committee: Sudarshan S. Chawathe (Chair): ...... [email protected] Technical Activities Committee: Daniel Zeng (Chair): ...... [email protected]

RECTIFICATION In the first newsletter of 2005, a technical paper ’Rhodes to Intelligent Transport Systems’ was published. Erroneously not all authors were mentioned. The paper ’ Rhodes to Intelligent Transport Systems’ was authored by Prof. Pitu Mirchandani and Prof. Fei-Yue Wang. We apologize to the authors and the readers for the inconvenience. Bart van Arem Editor in Chief

2 Society News

From the Editor by Bart van Arem

Dear Readers,

I am pleased to offer you the third Newsletter of the Intelligent Transport Systems Society of the IEEE. In this newsletter you will find a lot of news on conferences and the development of our ITS Society. We have reports on the recent conferences on Intelligent Vehicles and Intelligence and Security Informatics. Our research review section focuses on research in Japan.

The email announcement is sent to approximately 10.000 addresses around the world. The March and June issue were downloaded 1,600 and 850 times this year respectively. We also observe that readers also tend to download previous issues when a new newsletter is issued.

We have already started new content such as book reviews and reviews of research programs. But in order to develop the content and appearance of the newsletter even further we want to now the opinion of our readers. Please take a few minutes time and complete our on-line questionnaire at: http://www.aida.utwente.nl/en/ITSS Newsletter Survey/ In the December issue of the newsletter we will publish the results. Meanwhile, if you have contributions that you would like to share with our readers, do not hesitate to contact us. I hope you will enjoy this newsletter.

Bart van Arem

Message from the IEEE ITS Society President by Charles J. Herget

I want to start off by expressing our gratitude on behalf of the Society to Stefano Stramigioli for his many years of service to both the Council and the Society. Last year, he served as Editor of this Newsletter, and this year he served as Vice President for Technical Activities and he is currently the Program Chair for this year’s ITS Conference. Earlier this year, Stefano asked to step down from his position as VP Technical Activities because of other commitments. I am pleased to announce that Prof. Daniel Zeng has agreed to take over as VP Technical Activities, and the Board of Governors approved his appointment at its meeting in June in Las Vegas. You can read

3 articles by both Prof. Stramigioli and Prof. Zeng elsewhere in this Newsletter.

The ITS Society is governed by a Board of Governors (BoG) that consists of fifteen elected members and nine officers of the Society. Starting this year, members of the Society will elect five members to the BoG for a three-year term. All of the current elected members of the BoG were elected last year the ITS Council’s Administrative Committee. This year will be the first time the members of the Society elect their represen- tatives.

The Nominations and Appointments Committee has come forward with an excellent slate of candidates. The nominees (named in random order) are Benjamin Coifman, Takaaki Hasegawa, Hsinchun Chen, Steve E. Watkins, Sudarshan S. Chawathe, Brian K. Johnson, Katsushi Ikeuchi, Petros Ioannou, Ka C. Cheok, and Jason Geng. Members of the Society should have received their ballot from IEEE by now or will receive one soon. Please be sure to vote. Ballots are due back at IEEE no later than October 14, 2005.

The Board of Governors elects the officers of the Society. Starting in 2006, there will be nine officers. The present officers were elected by the ITS Council’s Administrative Committee. All of the officers are elected to a two-year term except for the president. The president is elected to serve a one-year term as president-elect before serving a two-year term as president and then serving another one-year term as past- president. The terms of the other officers are staggered so that each year four officers are elected to begin their term on January 1 of the following year.

The next election of officers will be held at the Board of Governors meeting in Vienna on September 17, following this year’s ITS Conference.

The officers to be elected are the Vice Presidents for Conference Activities, Member Activities, and Publi- cation Activities and the Editor-in-Chief of the Transactions on ITS.

I am always interested in hearing if there are any suggestions for improving the operation of the Soci- ety or if there are any other activities that you think the Society should be undertaking. Please send me an email at [email protected] if you have any comments.

Charles Herget President, IEEE Intelligent Transportation Systems Society

Calendar of Society Events by Daniel J. Dailey

Up Coming events for ITSS:

ITS Society Meetings:

IEEE ITS Society Board of Governors Vienna, Austria September 17,2005

IEEE ITSC 2005 September 13-16, 2005 Vienna, Austria Chair: Reinhard Pfliegl - [email protected] Program Chair: Stefano Stramigioli

4 IEEE VES’05 October 15-17, 2005 Xi’an, Shanxi, China Chair: NanNing Zheng - [email protected] Program Chair: FeiYue Wang

IEEE IV’06 June 13-16, 2006 Tokyo, Japan Chair: Katsushi Ikeuchi - [email protected] Program Chair: Hideki Hashimoto, Mohan Trivedi, Christian Laugier

IEEE ITSC 2006 September 17-20, 2006 Toronto, Canada Chair: Baher Abdulhai - [email protected] Program Chair: Issam Kayssi

IEEE IV ’07 Istanbul, Turkey Chair: L. Guvenc - [email protected]

IEEE ITSC 2007 September 29- October 4, 2007 Seattle, Washington, USA Chair: Daniel J. Dailey - [email protected] Program Chair: Brian Johnson

IEEE ITSC 2008 Korea Chair: Hungman Kim

IEEE ITSC 2009 October 4-7, 2009 St. Louis, Missouri, USA Chair: Steve Watkins - [email protected]

ITSS Officers and Committees can be found at: http://www.ewh.ieee.org/tc/its/officers.html

ITSS Calendar and Related Events can be found at: http://www.ewh.ieee.org/tc/its/calendar.html

Web Pages for past, present and future Conferences: http://www.ewh.ieee.org/tc/its/conf.html

Update your IEEE membership to include ITSS at: http://www.ieee.org/itss and click Update

Message from the VP for Conferences by Paul Kostek

Since I last wrote we have held a very successful IV05 in Las Vegas and ISI 05 in Atlanta. Reports on IV05 and ISI 05 are included in separate articles in this Newsletter.

With our growing list of conferences, a Conference Calendar has been created, is included in this newsletter and posted on the ITSS Website. It also will be included in the Transactions. This calendar provides the dates, locations and contacts for our Conferences.

At the last BoG meeting, the Board selected Istanbul, Turkey as the site for IV07. The Chair will be Professor Levent Guvenc: [email protected]. Bids are now being accepted to host IV08. If you are interested in bidding to host IV08 please contact me at [email protected].

5 The BoG also selected St. Louis, Missouri as the site for ITSC09. Dr. Steve Watkins: [email protected] will be the Conference Chair. Our ITS Conferences are now scheduled through 2009. The 2010 ITSC will be held in North America and if you are interested in bidding to host the Conference please contact me.

The IEEE International Conference on Intelligence and Security Informatics was held May 19-20 in Atlanta, Georgia and had over 600 attendees. This was the first time ITSS served as the sponsor of ISI. Planning has begun for next year’s Conference to be held in California. Daniel Zeng: [email protected] is the contact for ISI.

ITSC05 will be held in Vienna, Austria September 13-16. The Preliminary Program, along with regis- tration information, has been posted on the Conference Website: http://www.itsc2005.at/. The General Chair of the Conference is Dr. Reinhard Pfliegl: [email protected].

The fourth Conference ITSS is sponsoring this year is the IEEE International Conference on Vehicular Electronics and Safety. It will be held October 14-18, 2005 in Xi’an, Shaanxi Province,China. The Gen- eral Chair for the Conference is NanNing Zheng: [email protected]. You can learn more at: https://150.135.155.192/ves05/index.php.

IV06 will be held June 13-15 in Tokyo, Japan. The General Chair of IV06 is Katsushi Ikechi: ([email protected]. The Call for Participation and additional information is available at: http://www.cvl.iis.u-tokyo.ac.jp/iv2006/.

ITSC06 will be held September 17-20 in Toronto, Canada and planning is underway. The contact for ITSC06 is Baher Abudalli: [email protected] and the ITSC06 Website is: http://www.itsc2006.org. The Conference Call for Participation will be available soon.

Information on the 2006 ISI and VES Conferences will be provided as soon as it is available. Watch the ITSS Website for the latest.

ITSC07 will be held September 30-October 3 in Seattle, Washington. The contact is Dan Daily: [email protected].

If you have any questions about any of the ITSS sponsored conferences, feel free to contact me. If you are interested in hosting one of our Conferences please contact me and I will provide you with information on how to prepare a bid.

6 Message from the VP for Publications by Emily Sopensky

Research: What interests you?

Most of you know that ITSS is a new society within the IEEE congregation of engineering focused or- ganizations. Many of you are familiar with the predecessor to the ITSS, the ITS Council, which five years ago introduced the IEEE Transactions on Intelligent Transportation Systems. More than a few of you have searched our archives to find the latest research on ITS.

The IEEE ITSS Transactions on Intelligent Transportation Systems serves a dedicated readership. For members of our society, electronic access to the transactions is free. Smart researchers also have known where to turn for up-to-date coverage of ITS research. To give you an example of what our membership was interested in during the 2004 4th quarter, here is a list of the top 10 articles downloaded.

7 Transactions Editor-in-Chief Professor Alberto Broggi is constantly shaping and building the Transactions to reflect the best research available in ITS. His tireless, hard-working efforts have resulted in our Board of Governors approving two increases in the number of pages that will appear in the Transaction per year. That means more first-rate, peer-reviewed papers will be available to you as an IEEE ITSS subscriber.

Online Accessibility Improves with IEEEXplore

IEEE continues to add online accessibilty to its many technical periodicals, papers and articles. IEEEx- plore, IEEE’s vehicle for accessing papers electronically, is constantly being improved. Read more on the next page. We’re happy that ITSS is able to be part of the constant improvement of IEEExplore that provides you with more and better access to papers online.

Please let us know if there is an area of ITS research that you would like to read more about. Email me at [email protected] or Prof. Broggi at [email protected].

8 IEEE Articles Available Through New Yahoo Service

Researchers can now use a new search engine called Yahoo Search Subscriptions to find papers from the IEEE and other subscription-based publishers. Previously, people who needed IEEE papers could use only IEEE Xplore, the institute’s document-delivery platform. That’s because IEEE articles are part of the ”deep Web,” which includes millions of access-restricted Web sites that typical Internet search engines cannot access. To locate this type of material, you’d have to visit and search several sites across the Internet to perform a comprehensive search.

The new Search Subscriptions search engine offers an alternate way to find and access the con- tent of personal subscriptions. Users simply go to the Yahoo Search Subscription Web site, check off ”IEEE publications,” type in the title, subject, or author of what they want, and they’ll be taken straight to a list of relevant articles. Those who have subscriptions to the publications can go further and access the articles they find by logging in with their user I.D. and password. Those without such subscriptions can choose to purchase the articles.

Yahoo Search Subscription users can also expand their queries to include results from any of the parti- cipating publishers. This feature shortens the time it would usually take to conduct a separate search of each publisher’s database, and it provides the ability to compare documents found in different databases.

The IEEE is one of several subscription-based publishers whose content is now available through this free Yahoo feature, which is still in a beta, or test, version. Other participating publishers include The Wall Street Journal, Financial Times, and Consumer Reports.

Visit Yahoo Search Subscriptions at http://search.yahoo.com/subscriptions.

Other notes

Prof. Bart van Arem ([email protected]), the editor of this newsletter, is in the process of broadening the content. See his comments and learn the results of the survey of the readership he conducted. Contact him directly if you are interested in helping with the newsletter content.

If you are an IEEE member, don’t forget to vote. Ballots are due by Noon, Nov. 1, 2005.

Last words from a VP Technical Activities by Stefano Stramigioli

As the reader of this newsletter you may be have noticed, that I have been succeeded by Daniel Zeng as VP Technical Activities. After some long thoughts I realized that my other university obligations, other international duties and my family did not allow me to invest the time ITSS-TAB deserves and I decided to resign. I have been active

9 in the Council and then Society for about six years with different positions and I had the pleasure to see it grow up to a society. I want to thank all the colleagues and friends which have made my time as a volunteer in ITSC and then ITSS a pleasant time and I wish them all and the society to further grow and successfully involve many academics and industries worldwide. During my short period as VP TAB, I generated some ideas for structuring the review process of conferences together with a strict relation with Technical Committees and I really hope this goal will be achieved: I am confident that this will improve quality and efficiency of many activities. Thats all friends! I hope to see you all in Vienna for ITSC2005 and to have the opportunity to thank you all personally. Stefano

Message from the VP Technical Activities by Daniel Zeng

Dear Colleagues, It is a great honor for me to be elected as the VP for Technical Activities of the IEEE ITSS at the last BOG meeting in June. I would like to take this opportunity to thank my predecessor, Prof. Stefano Stramigioli, who has been a tremendous help to me explaining responsibilities of the post, sharing his vision for organizing and coordinating future ITSS technical activities, and providing operational guidance to facilitate the transition. The next 12 months will be critical as our new society is putting in place a formal structure to guide various technical activities. Based on a proposal initiated by Prof. Stramigioli, we are in the process of implementing a new technical activity board (TAB) structure under the leadership of the VP for Technical Activities in close collaboration with the VP for Conference Activities and the VP for Publication Activities. This structure was designed to tackle the issues associated with the increasingly expanding topic coverage of ITS-related research, the growing number of technical conferences under the ITSS sponsorship, and the resulting increased topic variety in submissions to these conferences and the Transactions on ITS. Under this structure, a number of technical committees are being formed based on subject topics. Each technical committee has a chair and formal members with appointment of at least two years. Technical committee chairs are expected to promote certain areas of ITS research by organizing special sessions at the ITSS- sponsored conferences or editing special sections for the Transactions and the Newsletter. Each technical committee chair is also expected to identify a small number of motivated experts to join a centrally-managed international conference program committee, which provides pools of reviewers for ITS-sponsored conferences for quality and timely reviews. Your active participation in these technical committees will be crucial for the success of this new TAB structure. We are hoping that as these technical committees are being established and becoming active players, many ITSS technical functions such as conference promotion, reviewer recruiting, and special topic session organization, will have a broader and more reliable support base. A particular emphasis of my ongoing effort is on establishing technical committees that are interdisciplinary in nature to help raise the awareness and recognition of ITS-related challenges in a broad set of engineering, information technology, and management fields. In future issues of the Newsletter, I will be reporting the status of this reorganization effort on an ongoing basis. We look forward to an exciting and fruitful period of active development and expansion of ITS research. Please drop me an email at [email protected] if you have any suggestions about future ITSS technical activities.

10 2005 IEEE Annual Election by Emily Sopensky

2005 IEEE Annual Election

IEEE is a professional association of over 365,000 electrical and electronics professionals in 150 countries. As a member, one of your benefits and rights is to vote in this annual election.

To learn more about this years candidates, visit the IEEE election site, http://www.ieee.org/organizations/corporate/candidates.htm In addition, each candidate for president- elect has a Web site. See:

Leah Jamieson at http://www.ece.purdue.edu/ lhj/IEEE

Gerald Peterson at http://ghpeterson.home.att.net

James Tien at http://www.rpi.edu/ tienj/IEEE/statement.htm

Ballots for the 2005 IEEE Annual Election of officers are being mailed beginning in August. If you are a full IEEE member, you may vote by mail or electronically (see https://www.directvote.net/ieee/). BALLOTS MUST BE RECEIVED BY 12 O’CLOCK NOON, CENTRAL TIME USA (18:00 GMT) ON 1 NOVEMBER 2005.

Members of IEEE, please vote! Leah Jamieson

I thank the Intelligent Transportation Systems Society for this opportunity to talk about my priorities for the IEEE. I also congratulate the Society and you, the ITSS members, for taking this extra step to be informed voters. In my position statement http://www.ece.purdue.edu/ lhj/IEEE, I outline priorities in four areas:

• Nimbleness in emerging technology areas: The rate of change of tech- nology is accelerating. We must continue to improve our ability to identify new technologies in the IEEE domain, create new technology communities, and establish ourselves as the place to go for novice-to-expert information about new areas.

• Agility in the changing information culture: Publishing is at the heart of both IEEE’s business and its service to the profession. We must be at the forefront in using technology to enhance access and use of our publications. Our content and tools must be relevant to students, researchers, engineers and professionals at all stages of their careers, and, increasingly, to the public. We must also plot a course that reflects the changing economics of publishing.

• Support for engineers throughout their careers: Current estimates put the half-life of engineering knowledge - the time interval in which half of what an engineer knows becomes obsolete - at five years. Just as IEEE is the preferred source for highest quality technical information, it should also become the preferred source for highest quality educational material for lifelong learning and professional development.

11 • Global relevance, local needs: IEEE’s structure gives us the opportunity to knit together the global nature of engineering with an understanding of specific local and regional needs. IEEE must use this structure to serve both the global profession and the changing needs of members throughout the world.

The role of the President of IEEE is to combine the strengths of this outstanding organization with a vision for how it can meet the challenges of the future. I will bring to the position of President a record of strong leadership skills and a long history of service to the IEEE. I will be guided by the key principles - strategic focus, teamwork and communication, sound financial models, value of membership, appreciation for volunteers, value to the profession - in helping IEEE meet the challenges and realize the opportunities that the changing world is presenting us. (Note from the Editor: this section was edited for space by the EiC).

Gerald Peterson

I am honored to be a candidate for the office of IEEE President-elect 2006, and to have this opportunity to share a few brief remarks on my candidacy - please see my web site: http://ghpeterson.home.att.net. Over 37 years, I have held positions in hardware and software design and engineering management and hold one US Patent in the field of telecommu- nications. In the past 17 years I have specialized in industry global strategic standardization. I currently hold the position of Senior Manager Emeritus at Lucent Technologies Bell Labs. I hold Electrical Engineering degrees from the University of Washington and Rutgers University (both in the USA). I am a member of the Tau Beta Pi Engineering Honor Society. In 2001 I was recognized as a ”Who’s Who”in its publication, THE BENT of Tau Beta Pi. Also in 2001 I received the American National Standards Institute’s Finegan Standards Medal for leadership in the development and application of voluntary standards. In addition to my leadership experience in the IEEE, I have served in elected national and international leadership positions that have delivered global technical standards and promoted increased global cooperation. We live in a time of accelerating change and globalization. The IEEE must respond to and help drive these changes if it is to continue to be a preeminent technical society. Key among these changes is how the IEEE delivers value to industry worldwide and, thus, value to the members of the IEEE. Our work and focus is on technology, which we must carry out while maintaining connection to the real word of markets and social benefit. The application of a broad range of technologies applicable to the development and improvement of intelligent transportation systems is clearly a key component of IEEE’s value, now and in the future. I know both the importance and the scope of the responsibility of being IEEE President and Chief Executive Officer, and if elected I am committed to giving my full time and attention to the office of IEEE President in 2007. I appreciate your consideration and welcome your questions, comments, and suggestions. Thank you, Jerry Peterson [email protected]

James Tien

Let me begin by thanking those of you who collected signatures for my petition candidacy; I am now a 2006 IEEE President-Elect candidate because of your hard work! My vision is for the IEEE to be the ”Global Resource of Choice” for scientific, educational and professional products and services. Consequently, IEEE must offer more global and portable member benefits (to support a typical career that includes multiple employers); IEEE must meet the continuing education needs of our members (who must update their knowledge base while being on the job); and IEEE must think and act

12 globally for the profession and think and act locally for the members (who have different cultural and professional needs). One concern that will affect ITS and all our technical societies is the issue of ”open access”; that is, publications - especially those derived from government- funded research - should be readily available and accessible. Although as a researcher I can applaud this stance, I am afraid that IEEE’s financial viabil- ity will be irrevocably undermined unless we take immediate steps to change IEEE’s current financial structure, one that is based on deriving more than 50 percent of our revenues on the sale of our publications. Even if open ac- cess does not necessarily imply ”free access”, it is obvious that we must curtail our dependency on publication revenues. We must develop new intellectual products and services that can be a source for new revenues. Indeed, in 2003 and as the newly elected EAB/VP, I worked with EAB staff and volunteers to launch IEEE/Thomson’s forthcoming Expert Now (formerly known as XELL) web-based learning library; it will contain the best of our conference tutorials and short courses. In addition to meeting our continuing education needs, Expert Now will serve as a new revenue source and possibly grow to replace the lucrative publication business. I have the qualifications to continue to help IEEE become the global resource of choice, especially given my experience as VP of Publications and as VP of Education. Moreover, as detailed in www.jimtien.com, I have extensive leadership experience and demonstrated excellence, including being elected to the U. S. National Academy of Engineering. Finally, I humbly ask not only for your vote, but also for your involvement: Together, We Can Advance IEEE’s Global Value.

13 Bookreview by Algirdas Pakstas London Metropolitan University

Book Review: Perspectives on Intelligent Transportation Systems

Reviewed by Sudarshan S. Chawathe, University of Maine

The book Perspectives on Intelligent Transportation Systems (ITS), by Joseph M. Sussman, is a collec- tion of the author’s writings aimed at professional engineers and students interested in a high-level overview of ITS that includes historical context. The book consists of 22 chapters grouped into five sections. The first section consists of a single chapter that provides a historical context for ITS, followed by a brief overview. The historical discussion notes early initiatives such as the European project Prometheus and the Japanese projects AMTICS and RACS, the formation of the Intelligent Vehicle Highway Society (IVHS) of America and the development of IVHS America’s Strategic Plan in 1992. The overview notes six functional areas of ITS: advanced traffic manage- ment systems; advanced traveler information systems; advanced vehicle con- trol systems; commercial vehicle operations; advanced public transportation systems; and advanced rural transportation systems. It stresses the impor- tance of integration and broad, systemic approaches to transportation. In- stitutional issues such as public-private partnerships, organizational change, and the role of academia, are also discussed. The section titled Transporta- tion and Change provides an informative discussion of the changes resulting from ITS, comparing them with those resulting from the development of the U.S. Interstate highway system. The second section comprises three chapters that focus on organizational issues, regionalism, and the transportation-information infrastructure. The first chapter, one of the larger ones in the book, addresses transportation op- erations, emphasizing organizational and institutional perspectives. Section 5 of this chapter, which discusses the needs of a successful operations mis- sion, is noteworthy. The needs are summarized as 3F/3I/3R, indicating that operations need to be funded, flexible, focused, integrated, intermodal, information- and customer-based, regional, real-time, and routine. In addition to a comprehensive list of references, this chapter includes several sizeable appendices: literature review, companion papers, definitions, and the regional perspective. The second and third chapters in this section briefly addresses deployment concerns, with the latter focusing on the idea of a competitive region. The third section consists of two chapters that discuss the implications of ITS for the transportation pro- fession and for transportation education. The first chapter describes new challenges in educating transporta- tion professionals in the context of ITS. It emphasizes the six I’s: internationalism, intermodal transportation, information and communication technologies, infrastructure, institutional frameworks, and innovation. It proposes a framework for transportation education composed of the foundation of transportation fundamen- tals, emphasizing the concept of transportation as a complex system, and the three areas of technology, systems, and institutions that build on this foundation. The chapter describes a T-shaped professional, suggesting breadth in the above areas and depth within a transportation specialty. The second chapter in this section addresses new roles for the transportation faculty, emphasizing the evolution to engineering systems. It presents three dimensions of transportation systems: the time scale, the geographic scale, and the organizational scale. It describes an I-shaped transportation faculty, suggesting domain knowledge in transportation that is unified, codified, and applied to new domains. The fourth section is a collection of 14 columns originally published as Thoughts on ITS in the ITS Quarterly in the period 1996–2001. These columns address a diverse set of topics, such as ITS and congestion,

14 in Chapter 9, and mega-cities in developing countries, in Chapter 14. The fifth, and final, section comprises three chapters that summarize the current state of ITS and discuss challenges for the future. The first chapter is based on a 2002 U.S. Department of Transportation report What have we learned about Intelligent Transportation Systems. It discusses achievements and challenges in diverse areas of ITS, such as Arterial Management, Traveler Information Systems, Public Transportation, along with crosscutting institutional and technical issues. It suggests assessing ITS using three dimensions: technology, systems, and institutions. The second chapter is organized around the theme of transitions in the transportation world. These transitions, 20 in all, provide a succinct and provocative summary of the evolution of the field. Examples include the transition from capital planning to management and operations; the transition from urban to regional scales in planning; and the transition from a supply-side perspective to a supply-demand equilibrium framework to non-equilibrium systems. The final chapter provides a retrospective analysis of the 1991 U.S. ITS Strategic Plan. Perspectives on Intelligent Transportation Systems Sussman, Joseph M. 2005, X, 229 p., Hardcover ISBN: 0-387-23257-5 Publisher’s recommended price: 125 USD.

Report on IEEE Trans. on Intelligent Transportation Systems by Alberto Broggi

Transactions EiC report, updated Aug 1, 2005

Thanks to the motion approved by TAB, we have the chance to increment the number of printed pages in 2005; this will reduce the backlog of papers to a minimum. We also increased the yearly page budget from 432 to 512 for 2006 thus providing timely publication of accepted papers. The Editorial Board has done a very good job in reducing the time required for the first review; now, this new important step will also reduce the time required from acceptance to actual publication. Furthermore, the Impact Factor of our Journal in 2004 increased to 1.104 (in 2003 it was 0.776). In 2004 the Trans on ITS ranked:

• 2nd out of 21 journals in the category TRANSPORTATION SCIENCE & TECHNOLOGY • 7th out of 79 journals in the category CIVIL ENGINEERING • 60th out of 209 journals in the category ELECTRICAL AND ELECTRONIC ENGINEERING In 2003 it ranked 7th, 10th, and 89th respectively.

We have the following special issues/special sections under way: • special issue connected to IV05 (guest editor: Fei-Yue Wang) • special issue connected to ICVES05 (guest editors: Nenning Zheng, Hironao Kawashima) • special issue connected to the ITS World Congress (guest editor: Chip White) • special issue on Cooperative Intelligent Vehicles (guest editors: Ljubo Vlacic, Toshio Fukuda) We are still working on the inclusion of multimedia content. Scholar One is very late in delivering the new customized version of their ManuscriptCentral (version 3.3). As soon as it is available, we will test it and will go live after a couple of weeks of successful testing. 15 The attached figure shows: in blue the number of papers submitted in each month from April 2003 (when we switched to electronic submission), and in red the number of papers sill without a decision; this means that either the first submission did not come to an end, or that a new revision is currently under evaluation. The figure shows that the trend is positive and, a part from isolated cases, all submitted papers receive a notification in a reasonably short time. The next Editorial Board meeting will be organized in conjunction with IEEE ITSC 2005, in Vienna, Austria, on September 16.

16 IEEE Trans. on Intelligent Transportation Systems - Index by Simona Bert´e

To go directly to the online Transactions Table of Contents, click on ”Index” above. IEEE ITSS members have full access to the papers. Non-members can browse the abstracts, which are provided below.

Vol.6, No.3, September 2005

• Real-time Hazardous Traffic Condition Warning System: Framework and Evaluation , by Cheol Oh, Jun-Seok Oh and Stephen G. Richie

Abstract: This study presents a warning information system based on an innovate methodology to estimate accident likelihood in real time. Bayesian modeling approach implemented by the probabilistic neural network (PNN) is conducted to identify hazardous traffic conditions leading to potential accident occurrence. The proposed system displays warning signs to call drivers’ at- tention for safer and careful driving once hazardous traffic conditions are observed by evaluating accident likelihood. It is believed that the proposed system to produce effective warning informa- tion for real-time safety enhancement could be a valuable tool to highway users and operators.

• State Space Reduction for Non-stationary Stochastic Shortest Path Problems with Real- Time Traffic Information , by Seongmoon Kim, Mark E. Lewis and Chelsea C. White, III

Abstract: Routing vehicles based on real-time traffic conditions has been shown to significantly reduce travel time, and hence cost, in high-volume traffic situations. However, taking real-time traffic data and transforming them into optimal route decisions is a computational challenge. This is in large part due to the amount of data available that could be valuable in the route selection. We model the dynamic route determination problem as a Markov decision process (MDP) and present procedures for identifying traffic data having no decision-making value. Such identification can be used to reduce the state space of the MDP thereby improving its computational tractability. This reduction can be achieved by a two-step process. The first is an a priori reduction that may be performed using a stationary, deterministic network with upper and lower bounds on the cost functions before the trip begins. The second part of the process reduces the state space further on the non-stationary stochastic road network as the trip optimally progresses. We demonstrate the potential computational advantages of the introduced methods based on actual data collected on a road network in southeast Michigan.

• An Approach to Tune Fuzzy Controllers Based on Reinforcement Learning for Au- tonomous Vehicle Control , by A.B. Rad, Xiahui Dai and C.K. Li

Abstract: In this paper, we suggest a new approach for tuning parameters of fuzzy controllers based on reinforcement learning. The architecture of the proposed approach is comprised of a Q estimator network (QEN) and a Takagi-Sugeno type fuzzy inference system (TSK-FIS). Unlike other fuzzy Q-learning approaches that select an optimal action based on finite discrete actions, the proposed controller obtains the control output directly from TSK-FIS. With the proposed architecture, the learning algorithms for all the parameters of the Q estimator network and the FIS are developed based on the temporal difference methods as well as the gradient descent algorithm. The performance of the proposed design technique is illustrated by simulation studies of a vehicle longitudinal control system.

• A Novel Architecture for Multilane-Free-Flow Electronic-Toll-Collection Systems in Millimeter- Wave Range , by Wern-Yarng Shieh, Wei-Hsun Lee, Shen-Lung Tung and Chung-Ding Ho

Abstract: An architecture for simultaneously performing multi-target tracking and multi-data communication suitable for millimeter-wave multilane-free-flow electronic-toll-collection systems is presented. This architecture combines the idea of frequency multiplexing in communication systems and the technique of target tracking in the pulse-doppler radar. For target tracking we

17 make use of pulse ranging by the aid of pulse compression and fine target-direction determination by amplitude comparison to obtain high resolution in radial direction (down range) as well as both lateral directions (cross ranges). This architecture can be utilized for both active- and passive- on-board-unit systems. Another important advantage of this architecture is that it can easily identify the passing vehicles not equipped with on-board unit. This will activate some subsequent enforcement activities against the violation vehicles.

• In Traffic Jam IVC - RVC System for ITS using Bluetooth , by Akihiko Sugiura and Candra Dermawan

Abstract: In the Intelligent Transport Systems field, research continues In-Vehicle Communi- cations, Inter-Vehicle Communications, Road-to-Vehicle Communications etc. All information communications technology, especially radio-communications technology, was applied. This pa- per, to minimize the costs of equipments, simplify a design, equipment, structure of all the systems communication for traffic jam area, we proposed to utilize a wireless Bluetooth technology sys- tem. The whole systems we proposed is connected to the Internet backbone provided some access point area, the Internet can be accessed from inside the vehicle and information, such as news and weather information can be downloaded. It is also possible to know traffic information for each access point area by accessing a data center server. Further more we developed Bluetooth-based IP phone service application. So the whole system, not only for transferring static or dynamic picture but also for voice communication can be performed.

• Comparative Evaluation of Microscopic Car-Following Behaviour , by Hussein Dia and Sakda Panwai

Abstract: Microscopic traffic simulation tools are increasingly being applied to evaluate the im- pacts of a wide variety of ITS applications and other dynamic problems which are difficult to solve using traditional analytical models. The accuracy of a traffic simulation system depends highly on the quality of the traffic flow model at its core, with the two main critical components being the car-following and lane-changing models. This paper presents findings from a comparative evaluation of car-following behaviour in a number of traffic simulators (AIMSUN, PARAMICS and VISSIM). The car-following algorithms used in these simulators have been developed from a variety of theoretical backgrounds and are reported to have been calibrated on a number of different data sets. Very few independent studies have attempted to evaluate the performance of the underlying algorithms based on the same data set. The results reported in this study are based on a car-following experiment which used instrumented vehicles to record the speed and relative distance between follower and leader vehicles on a one-lane road. The experiment was replicated in each tool and the simulated car following behaviour was compared to the field data using a number of error tests. The results showed lower error values for the Gipps-based models implemented in AIMSUN and similar error values for the psychophysical spacing models used in VISSIM and PARAMICS. A qualitative ”drift and goal seeking behaviour” test, which essentially shows how the distance headway between leader and follower vehicles should oscillate around a stable distance, also confirmed the findings.

• Aircraft conflict prediction in presence of a spatially correlated wind field , by Jianghai Hu Maria Prandini, Shankar Sastry and Jianghai Hu

Abstract: In this paper the problem of automated aircraft conflict prediction is studied for two-aircraft midair encounters. A model is introduced to predict the aircraft positions along some look-ahead time horizon, during which each aircraft is trying to follow a prescribed flight plan despite the presence of additive wind perturbations to its velocity. A spatial correlation structure is assumed for the wind perturbations such that the closer the two aircraft the stronger the correlation between the perturbations to their velocities. Using this model, a method is introduced to evaluate the criticality of the encounter situation by estimating the probability of conflict, namely, the probability that the two aircraft come closer than a minimum allowed distance at some time instant during the look-ahead time horizon. The proposed method is based

18 on the introduction of a Markov chain approximation of the stochastic processes modeling the aircraft motions. Several generalizations of the proposed approach are also discussed. • Coordinated Road Junction Traffic Control By Dynamic Programming , by Mark T.K. Ho, Tsan H. Heung and Isaac Y.F. Fung Abstract: This paper presents a novel approach of road traffic control for interconnected junc- tions. With a local fuzzy logic controller installed at each junction, a dynamic programming technique is proposed to derive the green-time for each phase in a traffic-light cycle. Coordination parameters from the adjacent junctions are also taken into consideration so that organized con- trol is extended beyond a single junction. Instead of a pursuit of absolute optimization of traffic delay, this study examines a practical approach to enable simple implementation of coordination among junctions while attempting to reduce delays if possible. The simulation results show that delay per vehicle can be substantially reduced, particularly when the traffic demand reaches the junction capacity. The implementation of this controller does not require complicated or demand- ing hardware and such simplicity makes it a useful tool for off-line studies or real-time control purposes. • Lane Keeping Based on Location Technology , by Jin Wang Stefan Schroedl, Klaus Mezger, Roland Ortloff, Armin Joos and Thomas Passegger Abstract: Vehicle positioning with an accuracy of 10 cm or less will enable lane-keeping assistance in addition to other safety benefits when enhanced lane-level digital map is in place. With constantly evolving technology and sensors, a high precision positioning system that fits into the automotive market can be expected within the next decade. Such a system will incorporate GPS and inertial system (INS) for enhanced positioning performance and availability. In this paper, the technology fields that will have a significant impact on the deployment of centimeter vehicle positioning system will be discussed. Vision based lane recognition systems are relatively mature and have already been introduced to market for lane departure warning etc. However both systems have some limitations. GPS/INS based systems may suffer from frequent satellite signal masking or blockage, while vision based systems do not work well in adverse weather condition or with poor lane signature. Effectively combining these two technologies can make a robust lane departure warning system. A precision map was made for the test area near Stuttgart using RTNA map-making approach. A Mercedes S-class equipped with both a vision system and a high precision GPS/INS was used for the test. The positioning map-matching results and the vision offset are compared and the complementary effectiveness is illustrated.

19 Conference Reports

IVS 05: New Developments and Research Trends for Intelligent Vehicles

Li Li, University of Arizona Jingyan Song, Tsinghua University Fei-Yue Wang, Chinese Academy of Sciences Wolfgang Niehsen, Robert Bosch Corperate Research Nan-Ning Zheng, Xi’an Jiaotong University

sistance and monitoring. Eight years ago, the US Department of Trans- portation launched the Intelligent Vehicle Initiative, focusing on preventing highway crashes by helping drivers avoid hazardous mistakes. This was a sig- Intelligent-vehicle sensing nificant new direction for USDOT safety programs, which had previously focused on crash mitigation- Two kinds of intelligent-vehicle sensing discussed that is, alleviating the severity of crashrelated in- at IVS 05 were out-vehicle environment and vehicle juries to people and property. Europe has also state. (In-vehicle environment is a third type, but paid more attention to road safety in recent years; since it primarily relates to a driver’s state, we dis- the European Road Safety Action Program aims to cuss related research in a later section.) reduce road fatalities by 50 percent by 2010 see Out-vehicle environment sensing involves collect- http://www.europa.eu.int/comm/transport/ ing information about the driving environment. Hot roadsafety/charter en.htm. Further evidence that topics at ISV 05 included driving safety and driver assistance have become worldwide themes appeared in the state-of-art re- • extracting lane boundaries, especially when not search projects presented in June at the 2005 IEEE clearly marked or in bad weather conditions; International Intelligent Vehicles Symposium. • detecting nearby vehicles and estimating their The reason behind this shift in focus is sim- position, speed, and acceleration; ple. According to the USDOT IVI program, in the US alone, more than 42,000 Americans die • recognizing the relevant traffic signs and traffic each year as a result of 6.8 million accidents lights; and http://www.itsdocs.fhwa.dot.gov/index.htm. Even • detecting the unexpected traffic participants farther reaching is the impact of highway injuries- (such as pedestrians) and obstacles. 3 million Americans were injured in 2001. Survivors Vehicle-state sensing focuses on measuring a ve- often sustain multiple injuries and require long hospi- hicle’s movement and monitoring its actuators. For talizations. The cost to society is more than US $230 example, researchers have studied how to detect billion a year-representing a greater share of the na- tion’s health care costs than any other cause of illness • a vehicle’s position, velocity, and acceleration; or injury [1]. The situation is even worse in develop- • an engine’s pressure and temperature; and ing countries. In 2003, more than 104,372 Chinese • a tire’s pressure, temperature, and friction co- died as a result of traffic accidents-on average, 286 efficients. people die each day [2,3]. Here we discuss several selected topics from IVS Novel proposals presented at IVS 05 included out- 05 to provide a broad overview of intelligent-vehicle vehicle sensing for bad weather and the integration research perspectives and innovative projects. Specif- of lane detection, vehicle localization, and vehicle- ically, we focus on advances in vehicle sensing, vehicle departure monitoring. motion control and communications, and driver ass-

20 Out-vehicle environment sensing under adverse Using the same detection theory, Massimo Bertozzi, weather Alberto Broggi, and A. Lasagni have employed an infrared camera for pedestrian detection [5]. The CMOS/CCD cameras, FMCW (Frequency Modu- camera can obtain 2D thermal images (see figure 1), lated Continuous Wave) radar, and LiDARs (light- from which users can analyze more morphological and detection and ranging devices) are the three most thermal characteristics [5]. Using infrared cameras frequently used surround sensors for out-vehicle en- produced a much higher detection rate than that of vironment sensing. However, conventional vision- conventional vision-based methods. based pedestrian detection is a difficult task, because William Herrington, Berthold Horn, and Ichiro pedestrians usually wear clothes in different styles Maski further studied using image fusion techniques and colors and might also carry items such as hats or to combine the relevant information from both visible bags of varied shapes. Moreover, illumination con- and infrared images [6]. The need for high frame rates ditions and moving cars and bicycles also introduce in an automotive application motivates their investi- distortions into the detection process. gation into computationally simple fusion. As figure 2 To conquer such problems, researchers have ap- shows, they applied a computationally simple image- plied thermopile and infrared sensors to intelligent fusion technique, based on the Discrete Haar Wavelet vehicle systems. For example, Dirk Linzmeir and his Transform, to combine three images from cameras op- colleagues have applied a thermopile sensor to mea- erating in different wavelength bands. sure an object’s presence in the sensor’s field of view, Another advantage of thermopile and infrared because objects of interest normally have higher tem- sensors is their ability to detect pedestrians pas- peratures than the environment [4]. sively without illuminating the environment-the sen- sors don’t electronically pollute the surroundings and are environment friendly. The only shortcoming is the cost, so realizing reliable yet lowcost on-vehicle thermopile or infrared sensors should be an interest- ing challenge over the next 10 years. Conventional vision-based pedestriandetection methods have also improved over the past two decades. Researchers recently presented several algo- Figure 1: Preprocessing phase for a stereo infrared rithms that can handle poor illumination conditions, camera system: (a) the original input image and (b) such as rain or darkness. For example, Hiroyuki the focus of attention [5]. Kurihata, T. Takahashi, and I. Ide proposed an in- teresting weather-recognition method that uses a subspace method to judge rainy weather by detect- ing raindrops on the windshield [7]. They define the concept of ”eigendrops” to represent the principal components extracted from raindrop images in the learning stage. Then, they use template matching to detect raindrops. In addition to identifying rainy or fair weather, this method could also help control windshield wipers.

Integrating lane detection, vehicle localization, and vehicle departure monitoring Figure 2: Applying a computationally simple image fusion technique based on the Discrete Haar Wavelet Lane detection and vehicle position measuring Transform to combine three images from cameras op- are two basic intelligent-vehicle functions, and re- erating in different wavelength bands: (a) the color searchers introduced a variety of related techniques visible image, (b) the long-wave infrared image (7-14 at IVS 05. William Travis, Adam Simmons, and µ, m), (c) the monochrome visible and near infrared David Bevly discussed using LiDAR to measure the (up to 1,100 nm) image, and (d) the two-level wavelet lane and vehicle’s heading angle in an indoor scenario fusion result [6]. [8]. However, LiDAR exhibits less measurement accu-

21 racy than Inertial Navigation Systems, so the authors combined the two sensor systems for better results.

Localization is also an important functionality for navigating intelligent vehicles. However, the data ob- tained from GPS and cameras is sometimes uncertain and or even momentarily unavailable (in urban ar- eas, for example). Frederic Chausse, J. Laneunt, and R. Chapuis studied the problem of GPS and vision- sensor-based localization, which combined GPS abso- lute localization data with data computed by a vision system to provide accurate vehicle position and ori- entation measurements [9]. They transform the posi- tion and orientation data into a global reference using a map of the environment and then estimate localiza- tion parameters using a particle filter. This lets them manage multimodal estimations, because the vehicle can be in the left or right lane. The best precision can supposedly reach 48 cm along the road axis and 8 cm along the axis normal to the road. Figure 3: A parking assistance system using a single-camera infrastructure [10]. The top image shows the camera’s view, and the bottom image shows the driver’s view, including the small screen Vehicle motion control and communications installed beside the steering wheel.

Developments in wireless and mobile communi- Intersection collisions represent a significant portion cation technologies are advancing methods for ex- of highway accidents, so roadside communications changing driving information between vehicles and to assist drivers has also recently gained much at- roadside infrastructures to improve driving safety and tention. A potential solution is to supply drivers efficiency. The concept of multiple-vehicle coopera- with timely alerts of imminent collisions. Chingyao tive driving has also recently emerged as a promising Chan and B´en´edicte Bougler set up an experimen- solution to traffic congestion. tal radar, configured to observe a vehicle’s leftturn motions with a trajectory (depicted by the yellow turning curve in figure 4)[11]. The subject vehicle in the figure is initially traveling north and then turning Motion control west. As the figure shows, a left-turn pocket exists for the subject vehicles. The triangle in the figure Drivers often can’t see what’s beside or behind their represents the coverage area of a radar device used vehicle, especially when backing up. When parking, for monitoring the movements of other vehicles in the for example, the driver might have difficulty deter- opposite direction. mining how close the car is to the curb. Furthermore, most vehicles don’t have rear- and side-view sensors because they’re too expensive and technologically challenging. To solve this problem, Yasuhiro Suzuki, T. Fujii, and M. Tanimoto have built an interesting multicamera system in a parking garage [10]. The system can easily identify a vehicle’s position and send the information to the driver in real time (see figure 3).

Other novel parking guidance and monitor projects aim to collect, record, and share all the infor- mation in the garage, including the vehicle’s license plate, available parking bay number, and toll collec- tion.

22 fic, improving traffic control performance. Since then, many others studies have addressed the feasibil- ity and benefits of cooperative driving-for example, California’s PATH project (Partners for Advanced Transit and Highways, www.path.berkeley.edu), the European Union’s Chauffeur project, and Japan’s Demo 2000 Cooperative Driving System (http://www.ahsra.or.jp/index e.html). The latest reports extend cooperative driving technology to road intersections, which is more com- plex than lane changing and merging problems. For example, Li Li and Fei-Yu Wang have analyzed how intervehicle peer-to-peer communications help vehi- cles near an intersection collaborate with each other [12]. They view each vehicle as an individual agent and determine the proper driving schedule through negotiation and planning. Then they modify virtual- vehicle mapping and the trajectoryplanning method to handle the collision-free requirements and vehi- Figure 4: A field observation site for the radar cle (dynamic and geometric) constraints. They also at an intersection [11]. have discussed communication-grouping algorithms, but further discussions are still needed, especially for If the system determines that it’s dangerous to turn multilane-driving scenarios. left based on the radar data (because a vehicle is coming to the intersection from the north at a high speed, for example), it will notify the driver to stop turning left. This cooperative vehicle- infrastructure is a flexible yet sensible solution that could be de- ployable in the near future.

Multivehicle cooperative driving and intersection con- trol

Individual-vehicle-control research focuses mainly on guaranteeing driving safety. Increased traffic conges- tion is making multivehicle-control research an im- portant topic. In 1991, the emphasis was on vehicle platoon control. Then, in the mid ’90s, researchers started examining lanechanging- and lane-merging- control problems. A solution to the former problem is path/trajectory planning technology, which stud- ies how to generate a collision-free driving path or trajectory under constrained vehicle dynamics. On the basis of these studies, researchers now consider Figure 5: Cooperative route contention at inter- cooperative driving with intervehicle communication sections [13]. to be a more promising answer to the problem of traffic jams and collisions.

The concept of cooperative driving was first pre- Yiting Liu, Umit¨ Ozg¨uner,¨ and E. Ekici have pro- sented by JSK (Japan’s Association of Electronic posed a three-level Intersection Warning System with Technology for Automobile Traffic and Driving) in a distance-based warning message generator (see fi- the early 1990s. Using appropriate intervehicle com- gure 5) [13]. Each vehicle approaching the intersec- munication to link vehicles, cooperative driving lets tion transmits its movement information and driv- vehicles safely change lanes and merge into traf- ing plan to the repeater installed at the intersection’s

23 Figure 6: The driver assistance system architecture[14]. center. The repeater then forwards or transfers such and 71,000 injuries (http://www.aaafoundation.org/ information to other vehicles and to the IWS. Simul- resources/index.cfm?button=drowsyfaq). Among taneously, the IWS generates the warning message different approaches in this field, monitoring the based on the received information and broadcasts it driver’s head position has received considerable in- to all the vehicles via the repeater. terest. This could help us detect and infer the driver’s fatigue level (especially when combined with a driver- eye-gaze tracking system) and implement a ”smart” Driver and passenger assistance airbag. Luke Fletcher, L. Petersson, and A. Zelinsky Conventional research has focused on how to make have proposed a novel idea for inferring driver fa- the drive more comfortable- for example, designing tigue, studying the relationship between road scene advanced suspension and chair systems to avoid in- monotony and driver vigilance [14]. Their idea comes jury and implementing smart air-conditioning con- from a psychology perspective that defines monotony trollers that adjust the vehicle’s inside temperature. as an exogenous contributing factor of fatigue. They More recent research addresses how to propose an integrated fatigue detection system that • monitor and analyze the driver’s state, uses driver-head-pose and eyegaze tracking as well as road monotony analysis (see figure 6). They claim • design an advanced vehicle and user interface that this system has better performance than those to more effectively exchange information, and that focus on driver-head-pose and eye-gaze tracking only. • monitor drivers’ behaviors to study, evaluate, and even mimic driving habits. The NHTSA also pointed out that al- though airbags saved over 6,000 lives by Advanced driver assistance systems should be the end of 2000, they also killed over 200 able to ensure that the driver’s reactions are appro- occupants through inappropriate deployment priate and safe. (http://www.iihs.org/safety facts/qanda/ airbags.htm). In response, the NHTSA issued a set of regulations mandating smart airbags that can adapt intelligently to the occupant. The head position al- Monitoring driving behavior gorithm must be robust to lighting conditions and uncontrolled driver postures. Infrared cameras can The driver’s diminishing vigilance level has become a help eliminate the disturbance of poor lighting con- serious traffic safety problem. The National Highway ditions. Algorithms can help reject occlusion and Traffic Safety Administration (NHTSA) estimates the presence of other competing head-like objects in that, in the US, drowsy drivers cause 100,000 acci- the scene. For instance, Stephen Krotosky, Shinko dents each year, resulting in more than 1,500 fatalities Cheng, and Mohan Trivedi have proposed a special

24 Figure 7: Successful detection for difficult examples: the captured images and the disparity images. The detected head location is shown in blue [15]. algorithm to constrain the relative size and disparity xamining all these IVS 05 papers suggests that of an occupant’s head in order to model and validate E intervehicle, vehicle-roadside, and vehicle-driver in- the potential heads in the camera image.15 Results formation sharing is currently the most attractive of ground truth experiments show that the detected trend in intelligentvehicle research. Consequently, head location can accurately estimate the occupant’s an important problem we’ll need to solve is setting 3D location. The demo system was robust to harsh up communication protocols so that products from lighting, partial occlusions, and competing objects different manufacturers can communicate with each such as hands (see figure 7). other. No single company or institution can pro- vide a complete intelligent vehicle, so interoperabil- ity among varied sensors and actuators emerges as a great new challenge. Advanced user interface design Acknowledgments Intelligent assistance systems can present drivers with more information-for example, using smart tire- This work is supported in part by Grants monitor sensors. As a result, information display #60125310 and #60334020 from NNSFC. placement and viewing methods are also hot top- ics. Bernard Champoux has proposed an interface to maximize information representation by collaps- References ing many of the separate dashboard controls, dis- plays, and systems into a single multifunction dis- play (MFD) [16]. A more challenging idea he has proposed is to switch the representation of informa- 1. F.-Y. Wang, P. B. Mirchandani, and Z. Wang, tion to match different driving situations (city ver- ”The VISTA Project and its Applications,” sus highway driving, for example). However, static IEEE Intelligent Systems, vol. 17, no. 6, 2002, displays have their own advantages; information is pp. 72-75. always in the same place and format. Transferring to 2. F.-Y. Wang et al., ”Creating a Digital-Vehicle MFD will break this rule and introduce learning and Proving Ground,” IEEE Intelligent Systems, usage trouble. vol. 18, no. 2, 2003, pp. 12-15. Solving this problem will require significant re- search into driver ergonomics. Also, results from 3. N.-N. Zheng et al., ”Toward Intelligent Driver- aviation assistance might be useful. Assistance and Safety Warning Systems,” IEEE Intelligent Systems, vol. 19, no. 2, 2004, pp. 8- 11.

25 4. D.T. Linzmeier et al., ”Probabilistic Signal In- Proc. IEEE Intelligent Vehicles Symp., IEEE terpretation Methods for a Thermopile Pedes- Press, 2005, pp. 106-110. trian Detection System,” Proc. IEEE Intelli- gent Vehicles Symp., IEEE Press, 2005, pp. 12- 11. C. Chan and B. Bougler, ”Evaluation of Coop- 17. erative Roadside and Vehicle-Based Data Col- lection for Assessing Intersection Conflicts,” 5. M. Bertozzi, A. Broggi, and A. Lasagni, ”In- Proc. IEEE Intelligent Vehicles Symp., IEEE frared Stereo Vision-Based Pedestrian Detec- Press, 2005, pp. 165-170. tion,” Proc. IEEE Intelligent Vehicles Symp., IEEE Press, 2005, pp. 24-29. 12. L. Li and F.-Y. Wang, ”Cooperative Driving and Lane Changing at Blind Crossings,” Proc. 6. W.F. Herrington, B.K.P. Horn, and I. Masaki, IEEE Intelligent Vehicles Symp., IEEE Press, ”Application of the Discrete Haar Wavelet 2005, pp. 435-439. Transform to Image Fusion for Nighttime Driv- ing,” Proc. IEEE Intelligent Vehicles Symp., 13. Y. Liu, U. Ozguner, and E. Ekici, ”Performance IEEE Press, 2005, pp. 273-277. Evaluation of Intersection Warning System Us- ing a Vehicle Traffic and Wireless Simulator,” 7. H. Kurihata, T. Takahashi, and I. Ide, ”Rainy Proc. IEEE Intelligent Vehicles Symp., IEEE Weather Recognition from In-Vehicle Camera Press, 2005, pp. 171-176. Images for Driver Assistance,” Proc. IEEE In- telligent Vehicles Symp. IEEE Press, 2005, pp. 14. L. Fletcher, L. Petersson, and A. Zelinsky, 205-210. ”Road Scene Monotony Detection in a Fatigue Management Driver Assistance System,” Proc. 8. W. Travis, A.T. Simmons, and D. M. Bevly, IEEE Intelligent Vehicles Symp., IEEE Press, ”Corridor Navigation with a LiDAR/INS 2005, pp. 484-489. Kalman Filter Solution,” Proc. IEEE Intel- ligent Vehicles Symp., IEEE Press, 2005, pp. 15. S. J. Krotosky, S.Y. Cheng, and M.M Trivedi, 343-347. ”Real-Time Stereo-based Head Detection Using Size, Shape and Disparity Constraints,” Proc. 9. F. Chausse, J. Laneurit, and R. Chapuis, ”Ve- IEEE Intelligent Vehicles Symp., IEEE Press, hicle Localization on a Digital Map Using Parti- 200, pp. 550-556. cles Filtering,” Proc. IEEE Intelligent Vehicles Symp., IEEE Press, 2005, pp. 243-248. 16. B.B. Champoux, ”A Mode of Interaction for Driver Vehicle Interface (DVI),” Proc. IEEE 10. Y. Suzuki, T. Fujii, and M. Tanimoto, ”Parking Intelligent Vehicles Symp., IEEE Press, 2005, Assistance Using Multi-camera Infrastructure,” pp. 795-800.

26 Report on IEEE International Conference on Intelligence and Security Informatics (ISI-2005)

Hsinchun Chen, Daniel Zeng, Fei-Yue Wang

session focusing on the perspectives and future re- ntelligence and Security Informatics (ISI) can be I search directions of the government funding agencies; broadly defined as the study of the development and and several invited panel sessions including one on use of advanced information technologies and systems technical ISI research and community building with for national and international security-related appli- Charles Herget and Fei-Yue Wang, the current and cations, through an integrated technological, orga- next ITSS Presidents, as panelists. ISI-2005 also in- nizational, and policy-based approach. In the past cluded a track on Terrorism Informatics, which is a few years, ISI research has experienced tremendous new stream of terrorism research leveraging the latest growth and attracted substantial interest from aca- advances in social science methodologies, and infor- demic researchers in related fields as well as practi- mation technologies and tools. tioners from both government agencies and industry. In addition to the main sponsorship from the U.S. We held the IEEE International Conference on In- National Science Foundation, the Intelligence Tech- telligence and Security Informatics (ISI-2005) in May nology Innovation Center, and the U.S. Department 2005 in Atlanta, Georgia. This conference was an of Homeland Security, the conference was also co- ITSS-sponsored event with technical co-sponsorship sponsored by several units within the hosting univer- from the IEEE SMC Technical Committee on Home- sities including: the School of Communication, In- land Security. It provided a stimulating intellectual formation and Library Studies at Rutgers; the Cen- forum for discussion among previously disparate com- ter for Discrete Mathematics and Theoretical Com- munities: academic researchers in information tech- puter Science at Rutgers; the Eller College of Man- nologies, computer science, transportation systems agement and the Management Information Systems and safety, public policy, and social studies; local, Department at University of Arizona (UA); the NSF state, and federal law enforcement and intelligence COPLINK Center of Excellence at UA; the Mark and experts; and information technology industry consul- Susan Hoffman E-Commerce Laboratory at UA; the tants and practitioners. There were 92 regular paper Artificial Intelligence Laboratory at UA; the Program and 12 poster submissions. About 40% of these sub- for Advanced Research in Complex Systems at UA; missions were from outside of U.S. Acceptance rate the College of Computing at GATECH; and the Geor- for regular (long) papers was about 40%. The two- gia Tech Information Security Center. day program included one plenary panel discussion

27 Technical Contributions

Bridge Collapse Detection and Motorist Warning System

James Justin Mercier, P.E., Senior Member, IEEE [email protected] Robert A. Marshall, Senior Member, IEEE [email protected]

I. Introduction (26 m). The bridge is owned and maintained by the Texas Department of Transportation (TxDOT) Pharr District Office. After a bridge span collapses, an unusual haz- In September 2001, two spans of the QIC were ard is created. Approaching motorists often cannot knocked out by an errant barge, leaving a 160 ft (49 see the missing span until there is insufficient stop- m) gap in the roadway (Fig. 1). Motorists driving ping distance, even under ideal visibility conditions, from the mainland side could see the missing spans due to the bridge’s vertical curvature. Until some just before the peak of the bridge and were able to form of human intervention finally stops the flow of stop in time. However, those driving from the island traffic, whether spectators, other motorists, or au- could not see the missing spans until after they had thorities, motorists may proceed into the danger. crested the peak of the bridge and were about 200 ft A bridge collapse detection and motorist warning (61 m) from the edge of the gap. The stopping dis- system that can immediately and automatically warn tance for an average car at the normal travel speed of approaching motorists of the eminent danger could 60 mph (96 Kph) is approximately 303 ft (91 m). In save many lives in the event of a bridge collapse. the fifteen minutes before emergency personnel could Unlike traditional smart structures which provide be notified and could close the South Padre Island highly detailed structural information, this system’s approach to the bridge, ten cars drove off the bridge goal is to provide reliable, immediate, and actionable into the water resulting in eight fatalities. advice. The requirements of such a system, being a safety system, include a high degree of reliability in both reporting emergencies and not generating false alarms. False alarms would quickly undermine pub- lic confidence in the system. Durability and main- tenance simplicity are additional requirements of the system. Such a system has been operational on a bridge in Texas, USA, since March 5, 2004.

II. Motivation Figure 1: Queen Isabella Causeway following the collapse of a third span, one day after a barge acci- Located at the south tip of Texas, the Queen Is- dent in September 2001. abella Causeway Memorial Bridge, formerly known as the Queen Isabella Causeway (QIC), is a 2.4 mile Unfortunately, this is not an isolated incident. (3.8 Km) long bridge crossing the Gulf Coast Intra- Other bridge collapses in other places have also re- coastal Waterway to connect the city of South Padre sulted in the loss of multiple vehicles because the Island to the mainland city of Port Isabel. The struc- hazard was not apparent to the drivers. Shortly ture’s construction consists of 147 concrete spans, 80 after the QIC was repaired and reopened, TxDOT ft (24.4 m) each, with a 780-ft (238 m) steel section initiated the design of a system that would detect a over the ship channel itself. The peak height is 85 ft bridge collapse and alert motorists to stop.

28 Figure 2: Basic system architecture showing major components: master and slave controllers, collapse detection sensor (breakers + fibers), motorist warning beacons, and emergency personnel notification.

III. System Architecture signals, no alarm is activated; the flashing beacons are normally-off and the traffic gates are held-open The basic concept of the system is straightforward by the master and slave controllers. Should either (Fig. 2). Critical parts include a controller subsys- signal be lost, the master controller will command tem, a collapse detection sensor, a motorist warning the autodialer to notify the TxDOT Pharr District subsystem, and an emergency personnel notification Office of a maintenance alarm. subsystem. Only when both signals are lost will the master The controller subsystem on the QIC consists of activate a full emergency alarm. Any damage to the a master controller on one end of the bridge and a bridge severe enough to cause one or more spans to slave controller on the other end. The collapse de- collapse will also sever the fiber optic cable. The loss tection sensor is a fiber optic cable run in a metal of both presence signals will cause the controllers to conduit that is attached underneath the bridge deck. activate the warning beacons and lower the gates, The motorist warning subsystem is a series of flash- and the master controller will command the autodi- ing red traffic signal heads as warning beacons spaced aler to alert emergency personnel to a full emergency along the bridge roadway with associated regulatory alarm. The controllers can also be configured to send warning signs explaining their function plus traffic preemption signals to existing traffic signals or to dy- gates at each bridge approach. The emergency per- namic message signs on the roadway to direct traffic sonnel notification subsystem is a telephone autodi- away from the affected bridge. aler, programmed to call the appropriate personnel Backup power for the controllers, beacons, and in the event of an alarm. gates is provided through a pair of uninterruptible When in operation, the master controller on one power supplies (UPS), one located with each con- side of the bridge sends presence signals through two troller. The UPS’s may need to power the controllers fibers to the slave controller at the opposite end of for hours under normal conditions, but for only half the bridge; the slave controller returns the two signals an hour or less under emergency conditions. Once through another two fibers to the master controller. the traffic has stopped, the hazard will be known, so As long as the master controller receives both extended operation is not necessary. Half an hour

29 should be sufficient time for emergency first respon- ered: audio, vibration, sonar, radar, free-space laser, ders to arrive on the scene. and metallic signal cable. The vertical and horizon- tal geometry of the bridge, as well as the risk of airborne interference, corrosion, and even vandalism A. Controller limited the choices to either metallic or fiber optic ca- The controller type selected for the QIC system ble enclosed in a conduit. But, signal losses in metal- was an Allen-Bradley PLC-5. A Programmable Logic lic cable required signal boosters spaced about 0.75 Controller (PLC) receives commands or inputs from miles (1.2 Km) apart, each requiring 120 volt elec- sensors such as thermostats or switches, and responds trical power to operate. Metallic cable also may be by directing actions such as turning on motors or ad- susceptible to lightning and other electrical interfer- justing flow rates. In this case, the PLC senses the ences. loss of signal and initiates the appropriate alarm. Fiber optic cable was selected because it required The master controller is housed in a stainless steel only one booster. The 2.4 mile (3.8 Km) length of cabinet located at the Port Isabel end of the bridge. the bridge required approximately 13,800 ft (4.2 Km) The slave controller, dubbed a Remote Rack due to of cable between the master and slave controllers. its lack of a coprocessor, is essentially a relay housed Then, Allen-Bradley newly released a long range 10 in a cabinet on the South Padre Island end of the Km fiber optic laser which eliminated need for any bridge. The Remote Rack controls the motorist warn- boosters. ing gate and beacons for the traffic traveling from But, fiber optic cable presented its own unique South Padre Island to Port Isabel. Although exposed problems. Both gel-filled and dry fiber optic cables to subtropical sun, neither cabinet is air conditioned; are designed with added stretch, slip, and reinforce- each has a thermostatically-controlled draft fan which ment to prevent breaking under the tension of normal draws in filtered fresh air. installation or usage. These properties may be ideal The Allen-Bradley PLC-5 was selected due to its for preventing cable breaks in telecommunications ap- history of reliability in harsh industrial and commer- plications, but are specifically contrary to the needs cial settings and due to district personnel familiar- of an elongation-based collapse detection sensor. The ity, in spite of its higher cost and more advanced TxDOT standard of a gel-filled or loose-buffered ca- programming experience required. PLC equipment ble is not recommended for this type of application reliability was a critical consideration in the original because the cable is designed to stretch at least 5% design which specified one PLC and Remote Rack. without damage to the fibers. The gel could also con- Before construction was complete, TxDOT decided tribute to the fibers slipping in the buffer tube instead to add a redundant PLC and Remote Rack in the of breaking to initiate the full emergency alarm. master and slave controllers. Each PLC was pro- A typical dry or tight-buffered cable stretches grammed to monitor the functionality of the other; much less before breaking. A loose lay fiber bundle is and both have access to the autodialer. tightly wrapped within a layer of Aramid (a tough synthetic yarn), and contained within a polyethy- lene sheath. The helical lay of the fibers provides B. Collapse Detection Sensor a little extra fiber length, increasing elongation be- The fiber optic cable selected for the collapse fore breakage. The Aramid yarn decreases friction detection sensor was a helical lay, six fiber, dry within the cable, permitting the fibers to move eas- buffered cable, supplied by Optical Cable Corpora- ily within the sheath. In addition, the reinforcing tion of Roanoke, Virginia, USA. The cable was in- fiberglass strength member within the cable assumes stalled in a rigid metal conduit fastened to the bridge any tension that would otherwise be imparted to the underneath the deck. Only one sensor conduit was in- fragile optical fibers. Unaddressed, the stretch, slip- stalled since concrete beam bridges do not normally page, and strength issues could prevent the fibers twist without collapsing the way that steel structures from breaking during a collapse. might. The cable was anchored and weakened at se- Several mechanical methods to ensure fiber break- lected locations to ensure that the fibers would break age were considered but were rejected as either not instead of the cable stretching or slipping during a reliable or risky to premature damage. The accepted bridge span collapse. The design of the collapse de- solution was to weaken the cable at selected places tection sensor was the most challenging aspect of the by cutting the sheath, strength member, and filler project. tubes or fibers, but leaving the optical fibers intact. Before settling on fiber optic cable, several other TxDOT and Practical Technology, LLC, of Caren- options for the collapse detection sensor were consid- cro, Louisiana, USA, developed separate devices to

30 accomplish the controlled breakage of the fibers. Al- would be transferred directly to the fibers, which are though both devices have been tested on a limited locked in place by friction. During installation any scale by their designers, no full scale test has been slack within the breaker should be taken up by ad- performed on either design. justing the distance between the spindles, minimizing The TxDOT device consisted of a condulet-style the amount of movement before the spindles lock. conduit box with clamps to guide the cable and a cen- ter post. About 6 inches (15 cm) of the optical fibers were exposed and routed on both sides of the post (Fig. 3). The fibers were divided so that the sending and receiving fibers of each signal would straddle the post. The design intent was that the collapse of a span would pull the cable until the sheath contacted the post. The toughness of the sheath would transfer all tension to the more delicate fibers causing them to break. Because the clamps in the TxDOT design only Figure 5: Top view of spindle anchor winding ar- guided the cable instead of clamping it securely, some rangement. Tension on the dressed fiber in either concerns arose that the fibers might slip in the sheath direction will ”lock” the spindles and transfer ten- for the length of the cable instead of breaking, es- sion to the fibers. pecially with the use of a straight lay cable. Should the fiber slippage occur, a span collapse might not Cable clamps in the spindle anchor effectively se- initiate a full emergency alarm unless some other fac- cure the cable so that strains from normal activities tor happened to break the fibers. A helical lay cable won’t damage the dressed fibers. The spindle an- twists the fibers tighter when pulled, breaking the chors also secure the cable to prevent the possibility fibers sooner, but suffered from installation issues. of runout - the cable elongating and slipping through the TxDOT breakers. Based on the bridge elevation, cable stretch, and fall geometry, the spindle anchors were installed every 1440 ft (439 m), or every 18 spans. The TxDOT breakers were installed every 240 ft (73 m) or every 3 spans. Spindle anchors were also mounted at each end of the bridge and in each of the three splice boxes on the bridge. During the initial field controller tests, heavy sig- nal losses were noted in the fiber one week after the Figure 3: TxDOT style fiber breaker. cable’s installation. An inspection found the dressed fibers wrapped tightly around the posts in the Tx- The spindle anchor designed by Practical Technol- DOT devices. Because the clamps in the TxDOT ogy (Fig. 4) was developed specifically to address devices did not anchor the fibers, the helical lay in the slippage issue of standard fiber optic cable for a the cable began to relax, transmitting the twist to collapse detection sensor application. the dressed fibers at the posts. The damaged cable was replaced with a special order flat lay cable from Optical Cable Corporation. Ultimately, the spindle anchors were the only uniquely manufactured item of the system. All other parts and components were either standard materials or special stock items from regular manufacturers.

Figure 4: Practical Technology style fiber breaker, or spindle anchor. III. Motorist Warning The motorist warning subsystem consists of traf- The spindle anchor prevents fiber slippage by wind- fic gates and warning beacons. In the event of a full ing the individual dry-buffered fibers around a set of emergency alarm, the flashing red beacons warn those anti-slip coated spindles in a counter-wound fashion on the bridge to stop immediately and not proceed, (Fig. 5). Tension on the cable from a span collapse while the traffic gates at the bridge approaches stop

31 anyone else from entering the bridge. Beacons be- to vandalism determined the best placement to be 3 yond the point of failure are disconnected, allowing ft (1 m) beyond the outside bridge rail. Placing the that traffic to exit the bridge normally. beacons on the center concrete traffic barrier was con- The most feasible way to stop traffic in an emer- sidered but rejected because the location appeared gency is with a flashing red signal. The warning inviting to vandalism. The beacon mounts were de- beacons each consist of two 12 in (300 mm) red light signed for 100-mph (160 Kph) wind speeds because emitting diode (LED) traffic signal heads which flash the QIC is located in a hurricane zone. During the alternately when activated (Fig. 6). near miss by the 2005 hurricane Emily, wind gusts in Port Isabel were reported at 79 mph (127 Kph), but only one beacon head was damaged.

Beacon spacing considerations included sufficient vis- ibility time, adequate stopping distance, and minimal scenic intrusiveness to arrive at a spacing of 560 ft (171 m). The Texas Department of Public Safety Driver’s Handbook lists the average stopping dis- tance for a passenger vehicle traveling at 60 mph (96 Kph) on dry pavement under normal conditions as 303 ft (91 m). Adding a margin of three seconds to cover reaction time and non-ideal road conditions results in a spacing of 567 ft (173 m). Since the bridge spans are 80 ft (24.4 m) each, spacing was set at 560 ft (171 m), the end of every seventh span. This spacing is intended to be sufficient for effective warning without being excessive during the normal non-emergency periods. The beacons were installed along the entire length of the QIC to provide protec- tion from collapse at any point along the bridge, not just in proximity to the ship channel.

The placement of warning beacons along a non- movable bridge is not a standard practice and war- rants explanation. Regulatory warning signs were mounted in each travel direction to instruct the mo- torists to stop immediately should the beacons ac- tivate: ”STOP when flashing DANGER/PELIGRO con luz intermitente ALTO.” Bilingualism was ap- Figure 6: Installed series of flashing red traffic propriate because of the proximity to Mexico and the signal beacons warn motorists of a bridge collapse. large number of Mexican tourists to the island. The first sign and beacon were mounted on the traffic gate at each bridge approach (Fig. 7). Additional signs The LED’s low power draw and monochromatic- were spaced across the bridge to remind motorists of ity make it ideal for this application. The LED’s the purpose of the beacons. are visible at least 1 mile (1.61 Km) in full daylight under normal coastal conditions of haze and bright sunshine. The beacons are controlled by one flasher module in their respective controller cabinets which causes them to flash synchronously on each side. The syn- chronous flashing is intended to relay urgency to the affected motorists. No guidelines exist for the placement or spacing of the beacons. Considerations of driver view ability, ease of electrical power supply, and inaccessibility

32 D. Electrical Issues Three primary electrical issues were identified: Protecting the warning beacon circuits from electri- cal shorts during a collapse, reliable electrical power, and protection of the PLC from lighting and transient voltages. The collapse of a span would probably shred the electrical conductors feeding the beacons causing the conductors to short together which would trip the circuit breaker, leaving all the beacons dark. To re- duce the likelihood of such a situation, brass welding cable connectors were used as quick disconnects for the beacon power conductors. Mounted in line with the conductors in the junction boxes at each beacon, (Fig. 8) the collapsing span would pull the conduc- tors causing the connectors to disconnect prior to the conductors shredding and shorting out. All the bea- cons up to the missing span would continue to oper- ate and warn the motorists. Under tests performed by TxDOT, the brass connectors pulled apart easily even when twisted into the lock position. South Padre Island and Port Isabel often suffer utility power outages, voltage sags, and voltage tran- sients, the effects of which were negated through an ITS quality uninterruptible power supply (UPS) in- stalled in each of the controller cabinets. ITS and traffic signal UPS’s are designed for high temperature extremes in remote, uncontrolled environments. To Figure 7: Installed traffic gate at bridge approaches mitigate the heating problems, the UPS was placed prevents additional traffic from entering the bridge on the highest shelf at the top of the cabinet while the during a full emergency alarm. batteries were placed on the lowest shelf at the bot- tom. The air circulation is drawn in near the bottom and exhausted through vents at the top. Lightning presented a risk to the controllers and equipment because of the geographic location and physical exposure. To provide a suitable ground path, the cabinet anchor bolts in concrete foundations were bonded to the reinforcing steel for a Ufer ground. All conduits in the system were rigid metal installed without coating to provide additional grounding and were bonded to both the cabinet and the reinforcing steel, including the feeder conduit from the electrical service. At each electrical service, a Phoenix Con- tact Combo-Trab lightning/transient surge suppres- sor was installed because the electrical utility lines in that area are particularly susceptible to direct strikes by lightning.

Figure 8: Connectors pulled to the right by a col- IV. PERIODIC TESTS lapse will leave the current beacon operating but protect the beacon circuit from a total blackout. A full emergency alarm test is performed quar- terly to check the operation of the motorist warning subsystem and the autodialer. A fiber break is sim-

33 ulated by removing jumpers in the master controller anchor designs. The tests could also provide infor- cabinet instead of disrupting the delicate fibers. Prior mation to determine maximum allowable spacing for to each test, the police in both cities are directed to each design. stop traffic at the scheduled time, and all parties on The QIC collapse detection and motorist warning the emergency call list in the autodialer are notified system is intended to be maintainable by state or of the test. city traffic signal personnel. Additions of cameras, The automated emergency notification of the en- sound detectors, weather stations or other devices tire call list is confirmed, and the traffic gates are are not recommended because each one adds a level checked for lowering times. A drive-around allows of complexity and associated failure modes to an in- each beacon to be checked; any repair needs are docu- tentionally simple system. mented. As soon as the drive-around is complete, the jumpers are replaced and traffic is released to travel, usually within ten minutes. Minor issues could cause delays of up to thirty minutes.

V. RECOMMENDATIONS FOR FUTURE DE- SIGNS AND USES

The QIC collapse detection and motorist warn- ing system is suitable for bridges in seismic hazard zones as well as those at risk over navigable water- ways by providing immediate warning of a bridge col- lapse until law enforcement or highway authorities can respond. Isolated rural structures can be col- lapsed without any awareness by authorities but still present a hazard to motorists. The total contract cost of about $900,000 (US) Figure 9: Activated system stops traffic from ap- for the QIC included basic materials costs for a 2.4 proaching the danger detected. mile (3.8 Km) long bridge and some expensive change orders to correct material errors. The largest single material cost item was the 38,000 ft (11.5 Km) of rigid metal conduit for the three separate runs enclos- ACKNOWLEDGMENT ing the fiber optic cable and the two warning beacon power circuits. A shorter bridge will not warrant as J. J. Mercier, P.E., thanks Mark Bloschock, Tx- many warning beacons or perhaps no warning bea- DOT Bridge Division Special Projects Engineer, cons at all where the bridge is short enough and traf- for exploring the idea of a collapse detection and fic gates are installed at the approaches. motorist warning system and supporting his de- Industrial computers may be used in place of high- sign efforts. He also specifically thanks all the powered PLC’s; the redundancy negates the necessity IEEE members and associates who shared their ideas for a singularly highly reliable PLC. With industrial and knowledge to help in the design of this sys- computers, each UPS could be downgraded to the tem. Juan Marfil and Eligio Alvarez oversee the reliable and functional traffic signal UPS. Industrial QIC Collapse Detection and Motorist Warning Sys- computers would also simplify maintenance because tem. Both are Electrical Engineering graduates of they are similar to the traffic signal controllers with the University of Texas Pan American in Edinburg, which the various departments of transportation in Texas, USA. They may be reached through the Tx- the United States and elsewhere are familiar. DOT Pharr District Office at 956-702-6100. Their A full scale test of the collapse sensor subsystem is email addresses are: [email protected] and still desired and encouraged. The ideal test would be [email protected]. a bridge scheduled for demolition where a pier could R. A. Marshall thanks Fred R. Marshall, be knocked out to trigger a span collapse. The use of Practical Technology, LLC, http://www.practical- high-speed cameras to record the sequences including technology.com for his development of the spindle an- when each signal is lost could prove conclusively the chor specifically for this project. effectiveness of both the TxDOT breaker and spindle

34 REFERENCES Mr. Mercier is a Licensed Professional Engineer in Illi- nois, Indiana, Michigan, Ohio, Texas, and Wisconsin, and 1. James. Justin. Mercier, P.E., ”The Queen a licensed master electrician in both the State of Texas Isabella Causeway Bridge Collapse Detection and the City of Austin. He is a past Chairman of the and Motorist Warning System,” Proceedings of Austin, Texas Chapter of the IEEE Power Engineering the Texas Section ASCE Spring Meeting, April. Society. 2005. Robert A. Marshall (SM IEEE) earned the B.S. de- gree in electrical engineering in 1994 from Louisiana State James Justin Mercier, P.E. (SM IEEE, M ASCE) is University, Baton Rouge, LA, USA. both a construction electrician and a Civil Engineer. He graduated from the International Brotherhood of Electri- He is currently an Independent Consultant in George- cal Workers (IBEW) Apprenticeship program in 1979 as town TX. His previous experience includes co-founding a a Journeyman Wireman (construction electrician), where rural local exchange telephone company. He also spent his electrical work experience included power plants, steel many years developing DSL, Ethernet, and CDMA cellu- mills, refineries, schools, traffic signals, and office build- lar technologies with Cisco Systems and Motorola, and a ings prior to his returning to college. He earned a B.S. short time at The Superconducting Super Collider. degree in civil engineering from the University of Cincin- nati, Ohio, USA, in 1986, and worked as an Environmen- Mr. Marshall has 19 patents awarded or pending. His tal Engineer in the private sector prior to joining TxDOT. patent for the spindle anchor described here is licensed to Since joining TxDOT in 1994, he has been a Traffic Practical Technology, LLC. He has also developed a com- Engineer in Roadway Illumination and in Traffic Signals, plete pre-programmed package subsystem implementing and a Hydraulics Engineer with the Design Division. the controller functionality described above.

35 Non-Society ITS News

A Glimpse on the Web by Alessandra Fascioli

This department is dedicated to catching a glimpse on the WWW trying to discover interesting ITS related Web resources. Reviewed sites range from research programs and projects, to software packages, databases, associations, non-profit companies, and more. Every suggestion or contribution is welcome and should be addressed to [email protected].

• PIARC, the World Road Association, is a non-political and non-profit association whose mis- sion is exchanging knowledge on roads and road transport policy and practices within an integrated sustainable transport context. PIARC wants to be an international forum for analysis and discus- sion of the full spectrum of transport issues, related to roads and road transport. PIARC creates and coordinates Technical Committees, organizes a World Road Congress and various technical seminars and publishes a large number of documents including a quarterly magazine. Link to PI- ARC site: http://www.piarc.org/en/ • The aim of the Society of Automotive Engineers is sharing information and exchange ideas for advancing the engineering of mobility systems. SAE is a resource for standards development, events, and technical information and expertise used in designing, building, maintaining, and operating self-propelled vehicles for use on land or sea, in air or space. SAE International, through the work of committee members and participants, maintains a number of technical standards and related documents. Through its ITS initiative, SAE has identified key intelligent vehicle (IV) functional areas as prime candidates for development to improve basic vehicle functions that, in turn, provide the basis for improved safety/security and information/entertainment functions. Link to SAE site: http://www.sae.org • The University of Maryland’s Center for Advanced Tranportation Technology (CATT) Lab works to provide safe and efficient transportation systems through improved operations and management by means of research and development, technology implementation, training and ed- ucation. Research focuses on Archived Data User Services, Data Visualization, Image Processing, Traveler Information Systems. Link to CATT site: http://www.cattlab.umd.edu/ • The Bureau of Transportation Statistics (BTS) is a statistical agency aimed at data collec- tion, analysis, and reporting and to ensure the most cost-effective use of transportation-monitoring resources. BTS brings a greater degree of coordination, comparability, and quality standards to transportation data, and to fill important gaps. Its mission is developing transportation data and information of high quality, and to advance their effective use in both public and private trans- portation decision making. Link to BTS site: http://www.bts.gov

36 Results of AHS demonstration experiment by Yuichi Odawara, Hiroshi MAKINO, Hiroyuki Mizutani

Results of AHS demonstration experiment

Yuichi Odawara, Assistant Director, Road Traffic Control Division, Road Bureau, Ministry of Land, Infrastructure and Transport, Japan, E-mail:[email protected] Hiroshi MAKINO, Senior Researcher, Intelligent Transport System Division, National Institute for Land and Infrastructure Management, Japan E-mail: [email protected] Hiroyuki Mizutani, General Manager, Advanced Cruise-Assist Highway System Research Association, E-mail:[email protected]

Introduction Particularly on the Tokyo Metropolitan Expressway, 21% of accidents are concentrated on accident-prone curves (6% of total length). Ministry of Land, Infrastructure and Transport For many of the accidents, 75% of the cause lies has been developing Advanced cruise-assist Highway in the actions of the driver right before the accident, Systems (AHS) that uses roadside sensors to detect such as a delay in recognition or an error in judgment stationary and low-speed vehicles (including rearmost (Figure 2). congestion) positioned ahead on curves with poor visibility. This system has undergone research and development for a service that uses on-board equip- ment in vehicles entering curves to provide immedi- ate advance information to drivers (Forward Obstacle Collision Prevention Service). The Sangubashi Sec- tion of the Route No. 4 Shinjuku Line of the Tokyo Metropolitan Expressway is one of the seven points at which the National Institute for Land and Infras- tructure Management (NILIM) is implementing road tests of AHS. Demonstration experiment of Forward Obstacle Collision Prevention Service was carried out Figure 1: Accident rates in urban expressways by for a three-month period starting in March 2005. Of- curve radius fering information to general drivers using the *3- Media VICS-enabled1 car navigation system, which is already commercially available and widespread in the Japanese market, the service reduces rear-end crash accidents and sidewall collision accidents and decreases the number of dangerous situations involv- ing near miss. This paper is to report on the results.

Risky Curve Sections of Urban Expressways

On urban expressways of Japan, the accident rate at curves with a radius of 200 m or less is 2.6 times higher than average (Figure 1). There are 470 such points throughout Japan, and it is estimated that the annual loss due to accidents is 10 billion yen. Figure 2: Breakdown of accident factors

1*-media VICS: There are two types of VICS receiver, one receives only FM multiplex broadcasting, and the other can receive information from beacons (radio/optical) all together (this kind of device is called ’3-media-receiver)

37 It is thus important to inform the driver of the situation of the road ahead beforehand, in order to reduce accidents. Although the safety of the vehicles themselves to recognize forward obstacles is contin- uing to increase due to the use of on-board sensors, etc., detection by a vehicle alone is difficult on sharp curves (R ≤ 250 m). Support from the road side, or so-called vehicle-highway cooperation, is therefore indispensable (Figure 3).

Figure 5: Overview of service

Figures 4 and 5 show the locations where the demon- stration experiment was carried out and a general outline of the offered service. The service provided showed a simple figure depicting the rear-end of traf- Figure 3: The difficulty for a vehicle to recognize fic congestion with the words ”caution, congestion an obstacle ahead in a sharp curve section ahead” on the screen of car navigation systems with beep sound for stationary vehicles, low-speed vehi- AHS has been researched and developed the Forward cles or congestion ahead, which were detected by Obstacle Collision Prevention Service to provide in- AHS image processing sensors of road side. Figure formation detected from road-side sensors concerning 6 shows the appearance of the information actually stationary or low-speed vehicles (including the rear- provided to car navigation systems. end of traffic congestion) ahead of curves, which have poor visibility. The information is provided just ahead of time to the driver using a 3-Media VICS- enabled car navigation system on the vehicle that will be entering the curve.

Outline of Demonstration Experiment Targeting General Vehicles

As it was confirmed through the proving test in FY2003 that the technology level for the ser- vice was feasible, a demonstration experiment was Figure 6: Forward obstacle information shown on implemented for the period from March 1, 2005 3-Media VICS-enabled car navigation system through May 31, 2005. The trial used the cur- rently widespread 3-Media VICS-enabled car naviga- tion system to provide safe driving support informa- tion to general vehicles. Confirmation of Effectiveness through Demonstra- tion Experiment

In order to measure the effectiveness of the ser- vice, it is possible to compare the number of acci- dents that occurred before and after introducing the service, however a long-term accumulation of acci- dent data is necessary for this. In response to this, a method has been proposed in AHS R&D for eval- uating the effect of traffic safety measures over the Figure 4: Locations for pilot project short-term by using a behavior analysis of each ve-

38 Figure 7: Situation of accident occurrence on Route No.4 Shinjuku line, including Sangubashi hicle captured by AHS image processing sensors. In considered near miss, such as sudden decelera- the analysis of the proving test conducted in 2003, it tion and approaching the curve at a high speed became to know that behind a single rear-end crash etc., decreased 9-14%. In the future, it can be accident, there are 80 cases of sudden deceleration thought that the effect will be further improved what are thought to be near miss. by, for example, greater penetration of car nav- igation systems, and the addition of providing 1. Basic analysis results information through roadside message sign. The figures for March to April, when the demonstration experiment was implemented, were compared with the number of accidents that occurred in the past fiscal year on the Route No.4 Shinjuku Line. (Figure 7) For the all lines of Route No. 4 as well as sim- ilar sharp curves, there was a increasing trend for 2005 compared to 2004, but the Sangubashi curve stands out in 2005 with a decrease. Be- Chart 1: Comparison before/after introduc- cause the drainage pavement on the Sangubashi tion of service based on sensor data analysis curve was being fixed in 2005, it is not possible to make a simple comparison, but it is possi- 3. Evaluation of satisfaction based on collection of ble to consider that safety measures, including drivers’ opinions the service offered as a demonstration experi- ment, had an effect. The inbound lane of the For this time’s demonstration experiment, test Sangubashi is a heavy-traffic section with daily monitors were recruited ahead of time from traffic of about 47,000 vehicles and is a place among general drivers, and their satisfaction where the occurrence of congestion is chronic level was evaluated by collecting a question- in the mornings and evenings. As a result of naire on their opinions when driving through surveying the vehicles that drive through this the curve section. There were 259 test moni- section, it was learned that about 10% of all the tors. Also, opinions were collected from other motor vehicles were equipped with a 3-Media general drivers through the Internet, etc., so to- VICS-enabled car navigation system. gether with the test monitors, there were 296 re- sponses on the experience of receiving the infor- 2. Results of traffic flow observation mation. The results of the collection of opinions The frequency of vehicles suddenly decelerating show that it is a service that will generally be in the curve section when there was an obsta- accepted by drivers, and drivers take expected cle ahead and speeds when the curve was being actions such as being cautious or gradually de- approached were observed using data detected celerating after receiving the information. by AHS image processing sensors (Chart 1). Also, dangerous actions such as being surprised As a result of the analysis, it was learned that by the information and making sudden maneu- because of the service offering advance informa- vers were not occurring. On the other hand, tion on obstacles ahead, behavior that can be there were also opinions that in the experience,

39 Figure 8: Overall evaluation of test monitors

the information provided was not useful. On that can be thought to be near miss through AHS analyzing the reason for that, it was learned image processing sensors. It thus became clear that that because it was already congested up un- ITS can be an effective method in terms of the anal- til the mouth of the curve, the information was ysis and evaluation of traffic safety measures. Future not necessary. In order to maintain the effect of tasks include service improvements reflecting drivers’ providing safety information, it is thought that opinions, and plans to undertake understanding the providing the necessary information at the nec- stability of the effect throughout the year. Aiming essary time is important, and this is a point for for the continued reduction of traffic accidents in the improvement in the future. Also, a question- future, research and development toward the realiza- naire was collected from the test monitors for an tion of an AHS service will be promoted. overall evaluation of the service experience over the course of the demonstration experiment pe- Future plans riod (three months) (Figure 8). Looking at the results of this, the service was highly evaluated, Smartway was placed as a national strategy aim- with 85% of test monitors evaluating it as effec- ing to realize a full-fledged ITS society through the tive, and 90% desiring the continuation of this concrete objectives of Zero Fatal Traffic Accidents, service for the Sangubashi curve. Safe Driving Support for the Rapidly Growing Num- ber of Elderly Drivers (Figure 9). Conclusion 1. Concept of vehicle-highway cooperation is up to now and spread to the world As a result of conducting a demonstration exper- 2. The effort to realize a full-fledged ITS society iment of a forward obstacle collision prevention ser- in 2007 is underway vice through 3-Media VICS-enabled car navigation 3. Early implementation is the goal for informa- systems, even though the rate of compatible car nav- tion provision igation systems being equipped was 10%, it was con- firmed that the service changed the behavior of ve- hicles toward the safe side due to the provision of information, and that the system could be accepted by general drivers. The results also showed that it was possible to clarify the system’s potential as a new traffic safety measure. However, in order to maintain the effect of the service, it became clear that improve- ments are needed so that the necessary information is provided at the necessary time. Additionally, as a method of evaluating traffic safety measures, it was Figure 9: AHS Development shown that it is effective to analyze vehicle behavior

40 Call for Papers

2005 IEEE International Conference on Vehicular Electronics and Safety

(IEEE ICVES’05) Sponsored by the IEEE Intelligent Transportation Systems Society October 14 - 18, 2005, Xi’an, Shaanxi, China

Advisory Committee Co-Chairs The International Conference on Vehicular Electronics and Safety (ICVES’05) is an Charles Herget annual forum sponsored by the IEEE Intelligent Transportation Systems(ITS) Society. It IEEE ITS Society gathers researchers from industries and universities to discuss research and applications for vehicle electronics, and vehicle safety systems. Papers dealing with all aspects of Konghui Guo Jilin Univ. , China vehicle electronics and vehicle safety-related intelligent systems are solicited for ICVES’05. Guoqing Sun China National Technical PROGRAM TOPICS Committee of ITS Standardization • Active and Passive Safety Systems • Navigation and Localization Systems Gang Wan • Telematics • In-Vehicular Network Tongji Univ. , China Vehicular Power Networks Vehicular Measurement Technology • • General Chair • X-By Wire Technology • Vehicular Signal Processing Nanning Zheng • System-On-a-Chip • Micro-electromechanical Systems Xi’an Jiaotong Univ., China • Vehicular Sensor • Image Sensor [email protected] • Vehicle Bus • Vehicle Control • Sensor Network • Driver Assistance Driving Systems Program Chair • Embedded Operation System • Adaptive Cruise Control Systems Feiyue Wang • Electro Magnetic Compatibility • Pattern Recognition for Vehicles Univ. of Arizona Inter-Vehicular Network Human Machine Interaction [email protected] • • • Human Factors • Others Program Co-Chairs Alberto Broggi Univ. of Parma, Italy [email protected]

Hironao Kawashima Keio Univ., Japan [email protected]

Richard Bishop IVSource.net [email protected]

Tutorial Chair Li Li Univ. of Arizona SPECIAL SESSION ORGANIZATION is encouraged. Organizers should contact [email protected] Prof. Xin Guan at [email protected].

Finance Chair PAPER SUBMISSION Yuehu Liu Xi’an Jiaotong Univ., China Complete manuscripts in PDF format must be electronically submitted to conference [email protected] website: https://150.135.155.192/ves05/index.php. Submitted manuscripts should be Publication Chair within six(6) pages in IEEE two-column format including figures, tables and references. Tsuneo Takahashi A Latex style file and a Microsoft Word templates are available from the IEEE website NF company, Japan (http://www.ieee.org/pubs/transactions/stylesheets.xml). However, submission MUST be PDF FORMAT. Publicity Chair Shuming Tang Please refer to the following websites for the most up-to-date information: Academy of Sciences, China [email protected] https://150.135.155.192/ves05/index.php Special Session Chair htttp://iv.xjtu.edu.cn Xin Guan http://www.ieeeves.org Jilin Univ., China http://www.ewh.ieee.org/tc/its/conf.html. [email protected]

Exhibits Chair Important Dates Haijun Gao Academy of Sciences, China [email protected] Paper submission deadline------June 15, 2005 Notification of acceptance------July 15, 2005 Registration Chair Camera-ready copy due------August 15, 2005 Hong Cheng Xi’an Jiaotong Univ., China [email protected] Call for Demonstrations Live vehicle demonstrations are planned for Oct. 19~Nov. 4, 2005 from Xi’an to Dun Local Arrangements Chair Huang, Perspective demonstrators should contact Prof. Lin at [email protected] Binggang Cao Xi’an Jiaotong Univ., China before Jul. 15, 2005. [email protected]

Student Activity Chair You can get the information of the conference agenda from Hong Cheng Grey Lai ([email protected]). University of Arizona, USA [email protected]

41 42

The IEEE Transactions on Intelligent Transportation Systems

Special issue on Cooperative Intelligent Vehicles Focus This special issue, being focused on a broad spectrum of cooperation among intelligent vehicles - regardless of whether they are driverless or driven by human beings - seeks submissions from scientists and researchers affiliated with both academic and industrial research institutes, as well as government institutions, on topics related (but not limited) to:

Decision & control algorithms for cooperative driving by driverless intelligent vehicles Decision & control algorithms for avoidance of a collision with dynamic road obstacles such as pedestrians, vehicles and other road users (i.e. cooperative driverless driving in time varying road environments) Vehicle-to-Vehicle communication and real-time data exchange techniques Vehicle-to-Infrastructure communication and real-time data exchange techniques Cooperative Intelligent Vehicle architectures Architecture of a System of Cooperative Intelligent Vehicles Cooperative Intelligent Vehicle On-Board Sensors

Submissions addressing practical applications and on-road (in-field) experimental testing are strongly encouraged. Authors intending to submit survey or tutorial papers may wish to decide to contact Professor Ljubo Vlacic before formal submission.

Guest Editors: Professor Ljubo Vlacic, Griffith University, Australia Fax: +61 7 3875 5198, Email: [email protected]

Professor Toshio Fukuda, Nagoya University, Japan Fax no. +81-52-789-3115. Email: [email protected]

Important Dates: 15 June 2005: Call for Papers 15 August 2005: Expression of Interest (please submit your letter of intent to [email protected]) 20 September 2005: Opening Paper Submission 15 October 2005: Closing Paper Submission 15 February 2006: Paper Acceptance Announcement 15 April 2006: Submission of reviewed versions and Final Review of the Accepted Papers 15 May 2006: Camera-ready Paper Submission September 2006: Expected Publication

Electronic submission process Letters of intent are not compulsory. This means that any paper submitted without prior submission of a letter of intent will be reviewed. However, the authors are strongly encouraged to inform the guest editor Professor Ljubo Vlacic of their intention to submit a paper. Authors are advised to follow the electronic paper submission requirements, as per the Information for Authors Guidelines published at http://its-ieee.manuscriptcentral.com/index.html?mode=instruction

The entire paper electronic submission process will be managed via the journal’s web site at http://its- ieee.manuscriptcentral.com. In case of any difficulty with the submission process, authors are encouraged to contact the Assistant to the Editor-in-Chief, Dr. Simona Berte, [email protected]

43 Upcoming Conferences, Workshops, or Symposia by Massimo Bertozzi

This section lists upcoming ITS-related conferences, workshops, or exhibits. Contributions are welcome; please send announcements to [email protected].

Biennial on DSP for in-Vehicle and Mobile Systems October 14–18 http://dspincars.sdsu.edu IASTED Intelligent Systems and Control Sesimbra, Portugal Cambridge, USA September 3 October 31–November 2

IASTED Artificial Intelligence & Soft Computing IASTED Robotics and Applications Benidorm, Spain Cambridge, USA September 12–14 October 31–November 2

ISPA2005: Computer Vision in Intelligent Transport ITS Worls Congress Systems http://www.itsworldcongress.org http://www.isispa.org/sscvits.html San Francisco, USA Zagreb, Croatia November 6–10 September 15–17 Intelligent Vehicles and Autonomous Navigation 2005 National Rural ITS Conf. http://www.isvc05.net/IVAN.pdf http://depts.washington.edu/trac/nrits2005/ Lake Tahoe, NV, USA Spokane, WA, USA December 5–7 September 11–13 IEEE 63st Semiannual Vehicular Technology IEEE 62st Semiannual Vehicular Technology Conf. (spring) Conf. (fall) http://vtc2006spring.com http://vtc2005fall.org Melbourne, Australia Dallas, USA May 8–10, 2006, September 25–28 ♦ submission by Semptember 16

Smart Demo - Intelligent Vehicle Trial Applications of Advanced Technology in Trasporta- http://www.sapro.com.au/smartdemo/smartdemo2005.htm tion Conf. Adelaide, Australia Chicago, USA September 29–30 August 13–16, 2006, ♦ submission by August 15 AET European Transport Conf. Strasbourg, France IEEE 64st Semiannual Vehicular Technology October 3–5 Conf. (fall) http://www.ieeevtc.org/vtc2006fall IEEE International Conference on Vehicular Elec- Montral, Canada tronics and Safety September 25-28, 2006, Xi’an, Shaanxi,China ♦ submission by February 13, 2006

44