SECOND ON-SITE IARP WORKSHOP on

HUMANITARIAN DEMINING

PRISHTINA, KOSOVO

June 19-20, 2003

Introduction

INTRODUCTION

The International Advanced Robotics Programme (IARP) has, till now, organized four workshops devoted to the possible use of Robotics Systems for speeding the demining of infested countries: the map shown on the first page summarizes the importance of the problem and the on this map coloured countries still suffer on this plague. The first IARP Workshop took place in Toulouse (France): a scientific meeting which gave some scientists the opportunity to present their theoretical results in the domain of the Outdoor applications for mobile robots; the most papers focused on the navigation and the control of robots while some papers also described existing advanced mechanical detection-devices (example: the UAV Camcopter) and their possibilities.

Following this first workshop, a working group HUDEM has been entrusted to Dr Tom Martin (Germany) who organized a first ON-SITE workshop in Zimbabwe (more precisely at the border Zimbabwe-Mozambica in the vicinity of Mutare, where the company MINETECH, funded by the UNO, hosted the members of the WG ) The first members of the WG Hudem received there a realistic perception of the demining problems, allowing some of them to re-orientate their research activities by taking into account with the operational requirements of such an application.

Dr Tom Martin concluded this workshop by inviting the members of the WG to introduce scientific proposals under FW-5 European Programme.

The third Workshop, hosted by Professor P.Kopacek in the Technical University of Vienna, and co- chaired by Prof Y.Baudoin, new chairman of the WG Hudem, again gave some scientists the opportunity to present the last results of their research in the domain of the detection of anti-personnel mines, of the mechanical mine-clearance and of the roboticized demining tasks.

Prof Yvan Baudoin concluded this workshop by inviting the members of the WG to pursue their efforts and to take into account with the ITEP directives (International Testing and Evaluation Programme), while Prof Peter Kopacek underlined the necessity to extend the detection tasks ( entrusted to mobile robots or other automated systems) to the mine-removal tasks (the effective mine-clearance)

The fourth workshop, again cosponsored by Austria and Belgium, is the second On-Site IARP workshop on Hudem and is hosted by the Technical University of Prishtina: a ‘come-back’ to the reality of the Demining problems in very unstructured infested areas where the use of mobile robots is far from obvious…

On behalf of the the cosponsoring members of the IARP, let us thank the National Organizing Committee and the participants to this workshop: their experience will allow us to prepare a detailed state-of-the-art report that will be sent to all the members of the Working Group Hudem and to the Secretary of the ITEP as well as to the Secretary of the GICHD (Geneva International Center for Humanitarian Demining) : we hope that this report will allow an efficient cooperation between those institutions and the End-Users in charge of the ON-SITE Demining

EUDEM-2, the funded European Network focusing on Hudem, will organize a (by the EC funded) scientific international symposium (Brussels, 15-19 September 2003) : don’t hesitate to join this event: it will be our next Rendez-Vous this year. We also hope to organize a fifth IARP Workshop in 2004, in close cooperation with the ITEP. Be ready to contribute…

Yvan Baudoin (BE) Peter Kopacek (Austria)

2 SECOND On-Site IARP WORKSHOP on Humanitarian Demining

Background of of IARP and IARP WG-HUDEM The International Advanced Robotics Programme (IARP) is an international project initiated at the Versailles Economic Summit of 1982. All countries/members of the IARP have agreed on the general objective: ”to foster international cooperation aiming to develop robot systems able to dispense with human exposure to difficult activities in harsh, demanding or dangerous conditions or environments. All members of the WG Hudem have agreed to foster international cooperation aiming to develop performant techniques and robotics systems speeding up the demining of infested countries.”

Why this Workshop in Kosovo?

Short History of Kosovo Kosovo covers an area of 10.887 km2, 53% are mountains and with about 2 million inhabitants. It is surrounded by Albania, Macedonia, Serbia and Montenegro. The capital of Kosovo is Prishtina and other major cities are Prizren, Peja, Mitrovica, Gjakova, Gjilan and Ferizaj. The highest mountains are situated in the southwest (Mali i Sharrit with peaks up to 2750m), in the west (Bjeshket e Nemura with peak Gjeravica of 2.656m) and in the north (Kopaonik). From Kosovo flow rivers toward three seas: Adriatic Sea, Aegean Sea and Black Sea. It is very rich with minerals which is a good foundation for industrial development. In the former times it was an important commercially crossroad between Adria and the Balkans. The earliest known inhabitants of Kosovo were called Illyrians by both Greeks and Romans. In 1389, in the famous Battle of Fushe Kosove / Kosovo Polje, the Serbs and their allies were defeated by the Ottoman Turks and shortly Kosovo became part of the Ottoman Empire. Kosovo was reoccupied by Serbs during the First Balkan war 1912 and remained a part of Yugoslavia until 1999 when NATO-led forces entered and set a United Nations administration. Places of interest: The Kllokot Spa, The Rugova Gorge, Brezovica, The Marble Cave, The Mirusha Canion. Heritage: Decan Monastery, Peja Patriarchy, Bajrakli Mosque in Peja, Hadum Mosque in Gjakova, Gracanica Monastery near Prishtina, Novo Brdo Castle, Prizren Castle etc. Facts about Prishtina Prishtina is Kosovo’s capital with more than 500,000 inhabitants. It’s a administrative, economic, cultural and education center. The history of Prishtina goes back to the Neolithikum and is situated near the ancient town called Ulpiana. Because of the minerals it was in the mid age (14th and 15th century) a very important market place. Now, it’s a modern city with three national TV stations, University with more than 20,000 students, National Library, National Theatre, National Museum, Academy of Science and Arts, Catholic Church, Orthodox Church and many old Mosques. Near Prishtina is based Airport as a link with outside world. In outskirts of city Thermo Power Plants are installed which supply Kosovo with electric power. Recreative and sportive center is situated about two kilometers from the center in Germia Valley. Memorial for the Kosovo Battle and Sultan Murat’s Tomb are just few kilometers outside Prishtina.

Demining Problem in Kosovo

At least 120 people have been killed and 370 wounded by landmines and cluster bombs left over from the Kosovo war since the conflict ended in June 1999. More than 4,500 mined fields were marked by international agencies. By the end of 2001 the United Nations Mine Action Coordination Centre (UNMACC) has disposed of 20,000 landmines and international peacekeepers have removed 7,000 of around 45,000 located. Since then Kosovo Protection Corps (KPC-TMK) is in charge for demining and explosive device clearance. According to international standards mine clearance can be guaranteed up to 99.6% and Kosovo needs up to seven years to reach that rate. Therefore, the Workshop in Pristina, Kosovo will examine how it could be possible to speed up the demining process using sophisticated devices as are mobile (wheeled, tracked or legged) robots and other modern equipments. http://www.ihrt.tuwien.ac.at/HUDEM03/HUDEM.htm - background

3 Sponsorship

This Workshop is sponsored by Austria and Belgium and co-sponsored by France, Russia, United States. The IARP representatives of the sponsoring and co-sponsoring countries are : Austria P. Kopacek Vienna University of Technology Belgium Y. Baudoin Royal Military Academy, Brussels Canada E. Dupuis, Canadian Space Agency, Quebec J.C. Piedboeuf France G. Giralt LAAS-CNRS, Toulouse Russia V. Gradetsky Russian Academy of Sciences Spain M. Armada Instituto de Automatica Industrial (CSIC) USA E. Marsh National Science Foundation (NSF)

International Programme Committee:

A. Almeida (PRT) M. Armada (ESP) Y. Baudoin (BEL) : Chair T. Borangiu (ROM) S. Buza (KOS) N. Caplan (USA) : IARP President J.G. Fontaine (FRA) : Representative P. Gonzalez (ESP) V. Gradetsky (RUS) M. Jimenez (ESP) P. Kopacek (AUT) : Chair B. Kuchen (ARG) C.W. Lee (KOR) E. Marsh (USA) T. Martin (GER) A. Pajaziti (KOS) C.E. Pereira (BRA) J. Sa Da Costa (PRT)

Local Organizing Committee

• I. Gojani (University of Pristina, Mechanical Engineering Faculty, Kosovo) • A. Pajaziti (University of Pristina, Mechanical Engineering Faculty, Kosovo) • Sh. Buza (University of Pristina, Mechanical Engineering Faculty, Kosovo)

Mechanical Engineering Faculty University of Prishtina Bregu i Diellit pn. Prishtina Kosovo, UNMIK Tel/fax. +381 38 554 997

4 PROGRAMME

JUNE 19, 2003

09.30-10.00 H Opening - Welcome Dean of the M.E.Faculty (Ko) Prof P.Kopacek (AU) Prof Y.Baudoin (BE) Dr A.Pajaziti (Ko) 10.00-10.30 H Demining Problem in Dr S.Buza (M.E.F Kosovo Prishtina) 10.30-11.00 H Demining in Kosovo Major D.Rexha (Kosovo Protection Corps) 11.00-11.30 H Coffee Break 11.30- 12.00 H Anti-Personnel mines JC. Robert-Peillard, neutralisation Y.Riebernon (Hitachi Furukawa, FR) 12.00-12.30 H Fuzzy-genetic algorithm Dr A.Pajaziti, and obstacle path I.Gojani,S.Buza,A.Shala generation for walking (MEF, Prishtina, KO) robot 12.30-13.45 H Lunch Break 13.45- 14.15 H Robotics for humanitarian Yvan Baudoin, Eric Colon, demining: a need? Ping Hong, Jean-Claude Habumuremyi, Ioan Doroftei,, Hichem Shali, Dragomir Milojevic, Jérôme Weemaels 14.15-14.30 H Existence of an Indian A.Rao (Goldspark, market for Robotics Bangalore) Demining 14.45 – 15.15 H Sensory and Robotics for Prof M.Acheroy Humanitarian Demining Prof Y.Baudoin (RMA, BE) 15.15- 15.45 H Trajectory tracking by Dr A.Shala, using fuzzy logic R.Likaj,A.Geca,A.Pajaziti, controller on mobile robot F.Krasniqi (MEF, Prishtina, KO) 15.45-16.00H Coffee Break 16.00-16.30 H Low cost, modular robot E.Cepolina (Un Genova, for landmine detection in IT) thick vegetation 16.30-17.00 H Humanoid Robots for Prof P.Kopacek (TU Wien, demining AU) 17.00-17.30 H Utilisation of Burmester Dr theory in design of planar L.Gojani,A.Pajaziti,B.Pira, mechanisms/manipulator P.Marango (Polytechnical Un Tiran, ALB) 17.30 H PANEL DISCUSSION IARP-KPC

5

MR. J.C. ROBERT-PEILLARD

COMPANY OVERVIEW

HITACHI-FURUKAWA LOADERS EUROPE S.A.S. (H.F.L.E.) is established by leading companies in the industry, namely HITACHI CONSTRUCTION MACHINERY COMPANY and FURUKAWA Co. Ltd.

H.F.L.E. is located in the suburbs of Lyon (Genas) and benefits from an outstanding location, in the neighbourhood of the main motorway network, the speed rail links (TGV) and Lyon's international airport.

The principal activities of the company are as follows : 1. Manufacture and design of wheel loaders and their related attachments. 2. Manufacture and design of mine clearance equipments and systems.

Wheel loaders manufactured by the company are distributed exclusively through the HITACHI's sales network.

CATEGORY : ANTI-PERSONNEL MINES NEUTRALIZATION.

EQUIPMENT NAME :

A 15 ton Excavator (prime mover) equipped with an Electronic Kit to address and neutralize through "Tool Box Concept" anti personnel mines located in hardly accessible areas..

Description :

The HITACHI-FURUAKWA mine clearance equipment and systems offers different solutions to attack and neutralize anti-personnel mines located on hardly accessible ground or surfaces such as slopes, ditches, along walls, fences, trenches, dykes, etc... The purpose of these tools is to concentrate on small areas and to support or complement other demining processes such as manual mine clearance and high scale mechanical techniques.

A simple and robust electronic box connected to angular sensors installed as a kit on the excavator, allows coordination of the movements of the system kinematics in space, and consequently at the end of the boom stick. This enables adjustment and automatic operations of the tools attached in the selected area.

Normal operation procedures can be achieved through an optional remote control system. A camera monitors and enables the operator to visualize and control the mine clearance sequence.

7

EQUIPMENT NAME :

Mine Sweeper AMDS ("Off the shelf tool").

Description :

The anti-personnel mine demining system (AMDS) is an area reduction tamping device for identification of the presence of anti-personnel mines in suspect areas and for the neutralization of these mines by detonating any near to surface or surface pressure mines.

Vibrating tamping capacity, combined with a simple electronic instrumentation, allows demining of areas that are difficult to access by manual demining or with other doming mechanical equipment.

Six independent and articulated footpad kits exert pressure on the ground through tamping, 1 ton per foot, and consequently detonate the anti-personnel mines. Additionally, these feet serve as a fuse to absorb the blow to the detonation. Upon explosion, the plastic pipe feet can be easily and quickly replaced.

The total AMDS ground tamping process can also be achieved through a remote control system. A camera monitor set enables the operator to visualize and control the operation. The weight of the AMDS is 700 kg.

EQUIPMENT NAME :

DiggerExcavator / Sifter UXOES ("Off the shelf tool").

Description :

The HITACHI-FURUKAWA excavator sifter (UXOES) is designed for sifting soil in difficult access areas where anti-personnel mines are a threat (i.e. slopes, ditches, channels, dykes, etc.)

The UXOES tool has a combination of "mill/rippers" or earth removing teeth and a "sifter drum basket" which sorts out the earth removed up to a depth of 25 cm.

The tool can dig and sort out the earth through the front and back swing of the "drum basket". At the end of the sequence the tool allows the automatic repositioning of the mill/ripper. In this manner, no surface remains unexplored. This particular anti-personnel mine tool is ideally suited for sandy ground and weights 350 kg.

8

EQUIPMENT NAME :

Vegetation Slasher / Bush Cutter ("Off the shelf tool").

Description :

This tool is a solid bush cutter and is designed with modular hood protection to accommodate different working applications.

For cutting vegetation, two types of cutters can be easily assembled to cope with different vegetation clearance requirements.

The bush cutter can also be transformed to a mini flail head (long straight blade) and has limited applications for clearance of anti-personnel mines down to a ground penetration depth of 5 cm. It is particularly effective on sloped terrain such as channels, trenches and dykes.

POINT OF CONTACT :

MR. J.C. ROBERT-PEILLARD Mobile : + 00 33 6 09 85 83 04 Phone : + 00 33 4 72 23 28 81 Fax : + 00 33 4 78 90 09 56 e.mail : [email protected]

COMPANY INFORMATION :

35, rue Roger Salengro – B.P. 211 F – 69742 – GENAS CEDEX (FRANCE)

9

A FUZZY-GENETIC ALGORITHM AND OBSTACLE PATH GENERATION FOR WALKING ROBOT WITH MANIPULATOR

Pajaziti, A.; Gojani, I.; Buza, Sh. & Shala, A. Mechanical Engineering Faculty, University of Prishtina, Kosova

Abstract

We present eight-legged walking robot Scorpion with manipulator that is controlled with a Fuzzy – Genetic algorithms basic approach. The goal is to develop walking robots for outdoor environments for the use of landmine detection. The eight-legged robot has to plan its path as well as gait simultaneously, while moving on flat terrain in presence of obstacles and ditches. It is a complicated task and no single traditional approach is successful in handling this problem. Due to the computational expenses there is a need to develop the faster and efficient algorithm for generation of optimal paths and gaits. The Fuzzy-Genetic algorithms approach developed in this paper enabled the tetrapod to do its job with minimum time of travel and with an optimum number of legs on the ground. The effectiveness of the algorithm is tested through computer simulations.

Keywords: Walking Robots; Fuzzy Control; Genetic Algorithm; Simulation; Path and gait generation

1. Introduction

Mobile platform systems, consisting of a walking robot platform equipped with one or more manipulators, are of great importance to host applications, mainly due to their ability to reach targets that are initially outside of the manipulator reach. Applications for such systems abound in mining, construction, forestry planetary exploration and the military. A wide category of such systems employs wheeled mobile robots, walking robots, which is the type of system under study in this paper. Walking robots continue the world-wide efforts in realizing such mechanisms with the goal to understand walking on the one side and to push forward a technology on the other side which might be useful for quite a lot of applications where wheel-driven machines are not reasonable. Research on legged robots typically concentrates on determination of vehicle’s trajectory, foothold selection and design of a sequence of leg movements. Motion planning for walking robots with manipulator is concerned with obtaining open loop controls, which steer a platform from an initial state to a final one, without violating the nonholonomic constraints. A comprehensive survey of developments in control of nonholonomic systems can be found in Kolmanovsky and McClamroch (1995). Moving mobile manipulator systems present many unique problems that are due to coupling of holonomic manipulators with nonholonomic bases. Seraji (1998) presents a simple on-line approach for motion control of mobile manipulators using augmented Jacobian matrices. The approach is kinematics and requires additional constraints to be met for the manipulator configuration. Perrier et al. (1998) represent the nonholonomy of the vehicle as a constrained displacement and try to make the global feasible displacement of the system correspond to the desired one. Foulon et al. (1999) consider the problem of task execution by coordinating the displacements of a nonholonomic platform with a robotic arm using an intuitive planner, where a transformation was presented. The same authors introduce other variations of local planners, which are then combined to constitute a generalized space planner (Foulon, Fourquet, and Renaud 1998). Papadopoulos and Poulakakis (2000) presented a planning and control methodology for mobile manipulator systems allowing them to follow desired end-effector and platform trajectories simultaneously without violating the nonholonomic constraints. The problem of navigating a mobile manipulator among obstacles has been studied by Yamamoto and Yun (1995) by simultaneously considering the obstacle avoidance problem and the coordination problem. To reduce the computational complexity of motion controls of mobile manipulator systems, some heuristics have also been developed by several researchers. Fuzzy Logic Controllers (FLCs) have been used by

10 several researchers in the recent past (Deb 1998, Pratihar, 2002, and Mohan, 2002) to solve the walking robot with manipulator navigation among stationary obstacles. The optimization problem involves finding an optimal fuzzy rule base that the walking robot should use for navigation, when left in a number of scenarios of stationary obstacles. Once the optimal rule base by using the Genetic Algorithm (GA) is obtained off-line, the walking robot can then use it on-line to navigate in other scenarios of stationary obstacles. In the present study, we concentrate on navigation problem, where the objective is to find an obstacle-free path between a starting point and a destination point, requiring the minimum possible time of travel. This paper consists of five sections: Section 2 describes the walking robot with manipulator model. The possible solution of the problem using FLC is proposed in Section 3. Section 3.1 discusses the Fuzzy and GA approaches. The results of computer simulations are presented and discussed in Section 4. Some concluding remarks and scope for future work are made in Section 5.

2. Mobile Platform and Manipulator System Model

The mobile platform and manipulator system model consists of two parts: walking robot with 8 legs, and manipulator with two segments (see Fig. 1). Because nonholonomy is associated with the mobile platform, while the manipulator is holonomic, the system is studied as two connected subsystems, the holonomic manipulator and its nonholonomic platform. This allows one to find an admissible path for the mobile platform that can move it from an initial position and orientation to a final desired one. Next, using known techniques for manipulators, joint trajectories are calculated for the manipulator so that its end-effector is driven to its destination. An advantage of this approach is that it is very simple to extend the method to mobile systems with multiple manipulators on board.

Fig.1. Eight-legged walking robot Scorpion with manipulator model during mine detection on a flat terrain

2.1. Nonholonomic Mobile Platform Subsystem

The eight legged robot was adopted as a mobile platform that is 65 cm long. The minimum height of the robot is 20 cm, the maximum height (legs outstretched) is 35 cm. It consists of a central body and eight legs with three segments each (see Fig. 2). The legs consist of a thoracic joint for protraction and retraction, a basal joint for elevation and depression and a distal joint for extension and flexion of the leg. With 6 degrees of freedom (DOF) of the central body the mobile platform has 32 DOF. Each leg has three DOF, turning around the α, β and γ axes, while the manipulator has two DOF (θ1 and θ2), which is installed on the front side of the mobile platform.

11

8 12

Central body α

C

γ Segment 2 v

M β θ1

θ2 Segment 3

H Segment 1

Manipulator Detector

S

Fig. 2. Model of the walking robot with a single leg and manipulator

Position of point M (the point where manipulator is attached to the central body) is given by equation:

x = s + []s + s cos(β ) + s cos(γ − β ) sin(α) M 1 2 3 (1)

Velocity of robot platform is given by:

vM = []− s2 β& sin(β ) − s3 (γ& − β&)sin(γ − β ) sin(α ) + (2) + [s1 + s2 cos(β ) + s3 cos(γ − β )]α& cos(α )

swing z s2 s

M s1

s x 3 stance γ s2 β s3 t M’ s 1 y S F

H x

α x S Fig. 3. Kinematics parameters of the central body and leg L1

12 During the swing/stance phases (see Fig. 3) the body height H and foot distance S should remain constant:

H = −s sin(β ) + s sin(γ − β ) = const. 2 3 (3) S = []s + s cos(β ) + s cos(γ − β ) cos(α) = const. 1 2 3

2.2. Holonomic Manipulator Subsystem

The Cartesian coordinates of the center of mass C1, center of mass C2 and the of detector D to the mobile platform joint M are given by (see Fig. 4). Position of C1 - center of mass for link 1: l x = v ⋅t ⋅ cos(ϕ) − 1 cos(ϕ −θ ) C1 o 1 2 (4) l y = v ⋅t ⋅sin(ϕ) − 1 sin(ϕ −θ ) C1 o 2 1

Position of C2 - center of mass for link 2:

l 2 x C = v o ⋅ t ⋅ cos( ϕ ) − l1 cos( ϕ − θ 1 ) + cos( ϕ − θ 1 + θ 2 ) 2 2 (5) l y = v ⋅t ⋅sin(ϕ) − l sin(ϕ −θ ) + 2 sin(ϕ −θ +θ ) C2 o 1 1 1 2 2 Position of detector D on xy plane is defined by:

d M = vo ⋅t , x x1

D

l2

C2 θ2

C1

θ1 l1 M’ ϕ dM y M ϕ y1 C mobile platform

Fig. 4. Kinematics parameters of the central body and manipulator

13 or by its components:

x = v ⋅t ⋅cos(ϕ) − l cos(ϕ −θ ) + l cos(ϕ −θ +θ ) D o 1 1 2 1 2 (6) y = v ⋅t ⋅sin(ϕ) − l sin(ϕ −θ ) + l sin(ϕ −θ +θ ) D o 1 1 2 1 2

Velocity of detector D is defined by:

x = v cos(ϕ) − v ⋅t ⋅ϕ sin(ϕ) + l (ϕ −θ& ) ⋅ & D o o & 1 & 1 ⋅sin(ϕ −θ1 ) − l2 (ϕ& −θ&1 +θ&2 )sin(ϕ −θ1 +θ 2 ) y = v sin(ϕ) + v ⋅t ⋅ϕ cos(ϕ) − l (ϕ −θ& ) ⋅ & D o o & 1 & 1 (7) ⋅cos(ϕ −θ1 ) + l2 (ϕ& −θ&1 +θ&2 )cos(ϕ −θ1 +θ 2 ) or in matrices form:

⎡x& D ⎤ p& D = ⎢ ⎥ , (8) ⎣y& D ⎦ with

⎡ 1 ⎤ ⎡vo cos(ϕ) −vo ⋅t ⋅ϕ& ⋅sin(ϕ)⎤ ⎡ l1 sin(ϕ −θ1 ) − l2 sin(ϕ −θ1 +θ 2 )⎤ J = ; J = ; q = ⎢ ⎥ ; w−r ⎢ ⎥ m ⎢ ⎥ & ⎢ ϕ& −θ&1 ⎥ ⎣vo sin(ϕ) + vo ⋅t ⋅ϕ& ⋅cos(ϕ)⎦ ⎣− l1 cos(ϕ −θ1 ) l2 cos(ϕ −θ1 +θ 2 ) ⎦ ⎣⎢ϕ& −θ&1 +θ&2 ⎦⎥

where is J = [J w−r J m ] central body and manipulator Jacobian, that consists of two sub-jacobians, Jw-r and

Jm, where Jw-r multiplies the mobile platform velocities corresponding to the uncontrolled degrees of freedom, and Jm multiplies the manipulator joint velocities corresponding to the controlled degrees of freedom. q& is the vector of central body and manipulator joint velocities. The matrix form of the equation (8) is given by:

p& D = J ⋅q& (9)

2.3. Mobile Platform and Manipulator System Dynamics

Kinetic energy of the manipulator is given by equation:

1 2 2 1 2 2 Ek = mD (x&D + y&D )+ m1(x&C + y&C )+ 2 2 1 1 1 2 2 1 2 1 2 (10) + m2 (x&C + y&C )+ JC (ϕ& −θ&1) + JC (ϕ& −θ&1 +θ&2 ) 2 2 2 2 1 2 2 where the moment of inertia for center of link 1and 2 is:

1 1 J = m l 2 , J = m l 2 , C1 12 1 1 C2 12 2 2

Second order Lagrange equations of motion for the mobile platform and manipulator is:

d ⎛ ∂Ek ⎞ ∂Ek ⎜ ⎟ − = τ (11) dt ⎝ ∂q& ⎠ ∂q

14 where τ - represents forces and torques exerted by walking robot and manipulator joint torques, Ek is the kinetic energy of system, while q and q& is the vector of walking robot and manipulator joint positions and velocities.

2.4. Gait patterns

Fig. 5. left shows the tetrapod gait, that describes the phase characteristics of the leg movement of the walking robot. The tetrapod gait, where four legs are on the ground at the same time, can be observed at low speeds of walking robots. During walking, a eight legged robot uses its 2 tetrapods not unlike a bipped stepping from one foot to the other – the gait is simply shifted alternately from one tetrapod to another. Since 4 legs are on the ground at all times, this gait is both “statically and dynamically stable. A cycle consists of a leg promotion period (swing) and a leg remotion period (stance„). Phase, swing, and stance are the parameters of the robot’s motion pattern (Fig.5, right) together with the step size of a leg.

1 2 L1 R1 R4 L3 R2 L2 R2 L1

L3 R3 L4 R3 L2 L4 R4 R1

Fig. 5. Tetrapod walking gait (left) and an according motion pattern beginning from the start position (right)

3. Solution of the problem by using Fuzzy Logic and GA Controller

The problem can be stated as follows: A eight-legged robot will have to plan its path as well as gait simultaneously, while moving on flat terrain with occasional hurdles such as ditches and in presence of stationary obstacles. The eight legged robot needs to do this job by avoiding to collide with any obstacles and not falling into ditches and all within the minimum time of travel and with an optimum effort to gain ratio. Moreover, its body height H and foot distance S should always be same to ensure static stability. To perform above tasks, in practice, the walking robot will have to do the following sub-tasks optimally (with minimum travel time): 1. Move along straight-line paths (only translation). 2. Take sharp circular turns (only rotation). In order to simplify the problem the following assumptions are made: 1. The contact of the feet with flat can be modeled as point contacts. 2. Each obstacle is represented by its bounding circle. 3. The terrain is discretized into cells and the center of each cell is considered as a candidate foothold. 4. The mass of the legs is reduced into the body and the center of gravity is assumed to be at the centroid of the body. 5. The detector should cover the surface with constant foot distance S during the forward motion of the walking robot.

3.1. Used controllers

15 In our proposed fuzzy and Fuzzy Genetic algorithm, two potential tools, namely fuzzy logic controller (FLC) and genetic algorithm (GA) have been merged to utilize the advantages from both. It is to be noted that the performance of an FLC depends on its rule base and membership function distributions. It is seen that optimizing rule base of an FLC is a rough tuning process, whereas optimizing the scaling factors of the membership function distribution is a fine tuning process. The FLC and GA controllers which are used for solving the problem of combined path and gait generations simultaneously of a walking robot with manipulator. The FLC finds the obstacle-free direction based on the predicted position of the obstacles in the next time step. The inputs, namely position and velocity of the central body and manipulator are fed to the FLC and there is the control vector to compensate the uncertainties of the model. A GA is used to control the body height H and foot distance S. The walking robot with manipulator will have to reach its destination starting from an initial position in minimum traveling time and its generated gaits will be optimal (with minimum number of ground-legs). This criteria is considered on the selection of the GA parameters, namely population size, crossover probability, mutation probability and the number of generations. Thus, the output from the GA block was also the control vector that compensated the gait errors.

4. Simulation results For simulation, according to the biometric approach the patterns are transferred to the model of robot, we have use sine function f(t) to avoid too abrupt accelerations, Fig. 6. The function f(t) for leg L4 is repeated by the other legs as g(t) after a partition of cycle.

⎧ π 30 ⋅sin(2πt − ) ⎪ 2 ⎪ ⎪if 0 ≤ mod(t, 2.5) ⋅ 0.5π ⎪ (12) f (t) = ⎨ ⎪ t − 0.5 π 30 ⋅sin(π + ) ⎪ 2 2 ⎪ ⎪ ⎩if 0.5 ≤ mod(t, 2.5) ⋅ 0.5π + 2 The phase ϕ between two legs remains constant independent of the velocity.

⎧ π 30⋅sin(2πt − +ϕ) ⎪ 2 ⎪ ⎪if 0 ≤ mod(t, 2.5) ⋅0.5π ⎪ (13) g(t) = ⎨ ⎪ t − 0.5 π 30⋅sin(π + +ϕ) ⎪ 2 2 ⎪ ⎪ ⎩if 0.5 ≤ mod(t, 2.5)⋅0.5π + 2

During the walk, all legs of robot, when they are in swing-position, distance S = 17 [cm] (distance between central body axe and foot - end side of segment 3) is constant. Also during the walk, central body axe height is constant value H = 15 [cm]. Length of segments and manipulator links are: s1 = 5[cm] s2 = 8 2[cm] , s3 = 23[cm], l1 =12[cm] and l2 = 24[cm]. Desired velocity of robot - central body is vo = 0.5 [m/s]. The rotational motion of the joints is transferred to the forward motion of the walking robot.

16 deg 30

15

f (t) 0

15

30 5 5.5 6 6.5 7 7.5 8 8.5 9 9.5 10 Ltrace4 1 Time [sec] R3trace 2 L2trace 3 R1trace 4

Fig. 6. A tetrapod group of thoracic hip joints for swing/stance = 3/5 and phase ϕ = 0.1

Control scheme (Fig. 7) of walking robot is realized in Matlab/Simulink.

Fig. 7. Control scheme of walking robot using Fuzzy – Genetic Algorithm

In control scheme shown in Fig. 7, through “Inverse dynamics” block, as output we have actuator torque on leg joints and manipulator joints (τ). From block “Fuzzy Logic Controller” also as output is torque (τflc). Through the Genetic Algorithm we have design as optimization algorithm for body height H and foot distance S of the walking robot, which are needed to be constant values. Using instructions given by Kalyamoy Deb where GA implementation is using binary and real coded variables, we have developed a

Genetic Algorithm Block. Inputs are estimated body high H and foot distance S. Outputs are torques (τga). Block “Walking robot and manipulator” represents the “Direct dynamics” solution of dynamic model for walking robot and manipulator given by differential equations:

D(q) ⋅ q&& + h(q,q&) = τ + τ flc +τ ga (14) where D(q) is symmetric inertia matrix, h(q, q&) is the vector of Coriolis, centrifugal and gravitational torque’s and other unmodeled disturbances, and q&&is the vector of joint acceleration. All simulated variables can be viewed namely, the paths, velocities and errors.

17 sim b Fig

y

u .

u Path [cm]

s 8 show l 26 60 a i Path [cm]

10 n

70 17 t 27 .5 57 . i 10 g th 5 4 ons (se 70

velocity [m/s]

15

.5 . 50 0 5 0. 0. 0.4

0.

50 0 52 54 0. 49 e s t

8 0 5

Fu

0 h

e e zzy-Gen Fi p a g. th of 1 4 0

Fig ) cen etic alg . Fig .

8 t r 37 . . Fig a 9 Pathof 37 5 l bod 8 .

5 o

Path . rith 10 y an . Velo m

of th s 12 . Th w e w d detecto x a ci d x d lk a e v ty i l rec k ing i r e ing e o 16 t c ion lo f t ro i walk r o

city of , 25 n r [cm] 25 bo o wh [ 0 c 18 b 0 m

t - o Ti

ile i ] t n m 20 -

g cen cen

e walk Fig [ ro s ] t tr bo r .

a al 9 show in l t - bodyan bod g rob 24 cen Legend y an L t s t ege o c P ral bo T – c P Pa ent 12 t r a 12 a h ac israth d d t 28 th nd: th h e 5 e 5 ral b e o o nt f o

: o p f e ra o f f w d a

d f tect w o

l ody y th of f a e b o a t

lk te

o o lking er co L1 d t ing

o c

y L1 t r o

32 cen

r r D robot D o n

b

s o t tant d t r

a –

l bod

3 6 0 0 u

ring

y an x x

th 40

d foo

y e entire z

y z t L1

,

0.15 0.66

0.10 0.33 m] m] c c S [ S [ - 0.05 - 0 E E r r

rro 0 rro 0.33 E E

0.05 0.66 0 4 8 12 16 20 24 28 32 36 40 0 4 8 12 16 20 24 28 32 36 40 Time [s] Time [s] a) b) Fig. 11. Stance – foot distance S error: a) with GA and b) without GA

0.12 0.40

0.06 0.30 m] m] c c [ [ H H - 0 - 0.20 E E r r

rro 0.06 rro 0.10 E E

0.12 0 0 4 8 12 16 20 24 28 32 36 40 0 4 8 12 16 20 24 28 32 36 40 Time [s] Time [s] a) b) Fig. 12. Error on walking robot - central body height H: a) with GA and b) without GA

In case of keeping the constant foot distance there were significant differences between the used algorithms with GA and without GA (see Fig. 11 a and b). The GA algorithm has also influenced in an optimal manner to reduce the central body height error (see Fig. 12).

5. Conclusions

Position of walking robot – central body and leg L1 had no significant deviation from desired values, but after 10 seconds the deviations increased considerably. The velocity error of the walking robot – central body in comparison with desired values was not so significant. The performances of controller GA keeping constant values of the body height H and foot distance shows that the GA decreased four times. This work can be extended further where the fuzzy rule will be designed by a GA that would be suitable for on-line implementation due to its computational complexity.

References

Deb, K., Pratihar, D. K., and Ghosh, A.1998. Learning to avoid Moving Obstacles Optimally for Mobile Robots Using a Genetic-Fuzzy Approach. A.E. Eiben et al. (Eds.); PPSN V, LNCS 1498, pp. 583-592. Foulon, G., Fourquet, J.-Y, and Renaud, M. 1998. Planning point to point paths for nonholonomic mobile manipulators. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), pp. 374-379. Foulon, G., Fourquet, J.-Y, and Renaud, M. 1999. Coordinating mobility and manipulation using nonholonomic mobile manipulators. Control and Engineering Practice, vol. 7, pp. 391-399. Gojani, I., Pajaziti, A., and Shala, A. 2002. Mobile robot navigation using cognition models and genetic algorithm-based approach. The 13-th International DAAAM Symposium, pp. 189-190.

19 Klaassen, B., Linnemann R., Spenneberg D., and Kirchner, F. 2002. Biometric walking robot SCORPION, Control and Modeling, Robotic and Autonomous System, 41, pp. 69-76. Kolmanovsky, I., and McClamroch, H. 1995. Developments in nonholonomic control problems, IEEE Control Systems, pp. 20-35. Mohan, A., and Deb, K. 2002. Genetic-Fuzzy Approach in Robot Motion Planning Revisited, N.R. Pal and M. Sugeno (Eds.): AFSS 2002, LNAI 2275, pp. 414-420. Papadopoulos, E., and Poulakakis J. 2000. Planning and model-based control for mobile manipulators. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), pp. 1810-1815. Perrier, C., Dauchez, P., and Pierrot, F. 1998. A global approach for motion of nonholonomic mobile manipulators. Proc. Of the IEEE Int. Conf. on Robotics and Automation, pp. 2791-2796. Pratihar, D. K., Deb, K., and Ghosh A. 2002. Optimal path and gait generations simultaneously of a six- legged robot using a GA-fuzzy approach, Robotic and Autonomous System, 41, pp. 1-20. Seraji, H. 1998. A unified approach to motion control of mobile manipulators. The International Journal of Robotics Research, 17(2), pp. 107-118. Yamamoto, Y. and Yun, X. 1995. Coordinated obstacle avoidance of a mobile manipulator. Proc. of the IEEE Int. Conf. on Robotics and Automation, pp. 2255-2260.

20 Robotics for Humanitarian Demining : a need ?

Yvan Baudoin, Eric Colona*, Ping Honga, Jean-Claude Habumuremyia, Ioan Dorofteia, , Hichem Shalib, Dragomir Milojevicc, Jérôme Weemaelsd

a Robotics Centre, Royal Military Academy, Avenue de la Renaissance 30, B-1000 Brussels Belgium b Department of Electronics and Information Processing VUB-ETRO-IRIS, Vrije Universiteit Brussel, Pleinlaan 2, B-1050 Brussels c Service des Systèmes Logiques et Numériques, Université Libre de Bruxelles 50, Av. F.Roosevelt, CP 165/57 B-1050 Bruxelles d SCMeRO, Université Libre de Bruxelles 50, Av. F.Roosevelt, CP 165/42 B-1050 Bruxelles phone: +32 2 737 6552 fax: +32 2 737 6547 [email protected]

Abstract. destruction percentage. This work is very slow, tedious, dangerous and costly. Furthermore, In the current situation, there is a need to the detection is not always reliable. correctly define the usefulness and Improvements can be made by developing new requirements of robotics solutions , essentially sensors, by automating the detection sequence in pre- and post-mine detection (minefield and by using different sensors simultaneously delineation and quality assurance), to develop (Baudoin & Colon, 1999). The Royal Military a network of research-centers focusing on this Academy, leading the Belgian project Hudem, kind of solutions , to define and continuously is focusing on the development of new data update generic modules of the used Robotics processing and fusion algorithms Systems . Beside the correct orientation of (Milisavljevic 1999), on improvement of research activities , deduced from such Ground Penetration Radar (GPR) (Scheers definitions, it will be necessary to develop test 1999) and on robotic systems that carry mines methods and procedures in order to assess the detection sensors. performances of the 'System' in highly, cost- effective and most generic way. The Network , Among the different ways robots could help with teams focusing on work-packages related human deminers, the scenarios described to the modules defined in the picture 1, could below are the most realistic. Small help to clarify the role of the Robotics Systems autonomous vehicles equipped with different (or Mechanical assistance) and assist future sensors could run around an area to delimit the T&E activities of ITEP. (International Test surface that is really polluted with mines. This and Evaluation Programme). This paper phase when done manually is the most summarizes the results of the research dangerous one because deminers are working activities on Robotics in Humanitarian faster and are taking more risks than during Demining, conducted by the RMA, as well as systematic detection. To study this first aspect, some important conclusions regarding the a small wheeled electrical vehicle named needs of such research objectives? Tridem has been developed (figure 1). A first indoor prototype was build to demonstrate the Keywords: Robotics, Control, concept feasibility (Alexandre, Weemaels, Doroftei & Colon 1998). A second version with more powerful motors and a modified 1. Introduction. frame has been developed for outdoor trials.

Once the actual mined area is delimited, a The terrible antipersonnel landmines plague systematic scanning process can begin. It has represents a real challenge for the research been proved that the use of different sensors community. Antipersonnel mines kill or could drastically improve the detection mutilate tens of people every day. efficiency and reliability. However, the data Humanitarian deminers still use classical fusion process requires the registration of the manual methods because heavy demining data acquired by the different sensors. This vehicles cannot achieve a satisfying requirement justifies the work done on the two

23 other systems: the first one which is a sliding The last aspect considered in this project is the pneumatic robots with a 2 degrees of freedom determination of the robot's location in the scanner (figure 2) and the second one which is field. This is required for navigation but also a modified Explosive Ordnance Disposal for automatic production of detection maps. (EOD) vehicle that carries a three-dimensional For this purpose, a visual servoing system Cartesian scanning system (figure 3) (Colon, based on a pan-and-tilt colour camera has been Baudoin & Alexandre 1998). The fact that the developed. (Hong, Sahli & Baudoin 1998). three systems have very different robotic This system tracks a colour beacon mounted architectures introduced some additional on the robot and sends in real-time the three- challenges in the development of the control dimensional position of the sensor to the main and interface architecture. Also the different control computer. The principle is shown in sensors used in the project and more figure 4. particularly their interface influenced the Besides mobility trials, the systems have been system implementation. used on dummy domestic minefields to record data for the searchers working on sensor development and data fusion algorithms.

Figure 1. The Tridem with the Metal Detector Figure 2. The pneumatic walking robot AMRU 4

24

Figure 3 The Hunter with a blue beacon Figure 4 The tracking system principle

Figure 5a. AMRU-5 : the six-legged robot Figure 5b. Possible configuration of AMRU-4

Finally, although this may be considered as a systems. Next, the control and communication long-term study, we also decided to develop a architecture is described. The emphasis is put six-legged robot (figure 5) and to study multi- on the control of the scanning and on the tools systems for the AMRU 4 (figure 5b). synchronisation between the distributed processes. Finally some results are presented This text continues with an overview of the and future work is discussed. different components of the robotic acquisition

2. System Overview

Basically, a Robotics system may be described by the next scheme:

25 VEHICLE CONTROL NETWORK

Control Transmit / transceiver Receive Vehicle status

Vehicle host

Motion control Sensor data Sensor data Transmit / Y

processing transceiver Receive R O T I POS

Sensor Sensor data E deployment fusion A R DAT Robot Mission ION

positioning Data acquisition Sensor S management S sensor process I M

Cutter control HMI Data acquisition Sensor 2 process

Data acquisition Sensor 3 process NK I

CONTROL TION L

A STATION BACKPLANE

Data acquisition Sensor n UNIC process MM

DATA O NETWORK C

SENSOR SUITE

During our project we control system of the vehicle has been focused on the shaded interfaced with a microcontroller such that the modules. Other work vehicle can now be controlled with a groups of our Hudem-team computer. The scanner is actually a Cartesian focused on the other ones. robot with three degrees of freedom. A DC The scanning robotic systems are composed of servomotor coupled with a planetary gear-head the following elements: actuates each axis. An optical digital encoder - The vehicles with their possible scanning is used for position and speed feedback and device, contact switches have also been placed at the - The mine detection sensors, end of each axis. The useful area is 850 by 500 - The tracking and location system. mm and the vertical axis has a travel distance of 500 mm. The system can be used on both As described in the preceding section the sides of the vehicle and is powered with robotic systems can be divided into two batteries or with a power supply. categories: the ones having a scanning device The second robot (AMRU4) is a pneumatically that can be equipped with different sensors and actuated walking machine (figure 2). Two the wheeled robot that simply carries a single sliding frames allow a linear motion and a sensor. In this case, the scanning of an area can rotating cylinder is used for changing the be obtained by moving the robot body itself. motion direction. The upper part of the robot is The mechanical systems need to be controlled a two degrees of freedom scanner that can and their motion must be synchronised with carry different kinds of mine detection sensors. the data acquisition. The vehicle motion is also As mentioned in the introduction, to develop synchronised with the tracking and location efficient detection methods, precise and process. reliable data are essential. This fact justifies The vehicles and the scanners are described in the development of those two systems, which this section; the control and communication are able to acquire multi-sensor data . architecture is presented in section 3. The third vehicle called Tridem (figure 1) has a triangular shape and stands on three wheels. The first vehicle available in the project is The wheels are connected to the frame as a called Hunter (figure 3); it is a small caterpillar three arms star. The frame supports the vehicle that was used by the Belgian Army for electronics and the batteries. The robot is anti-terrorism operations. The original manual equipped with three driving and three steering

26 motors. The power is transmitted to the driving electrical connections are also very easy; the motors by copper brushes. This solution gives wires (control signal and power) are connected the wheels a complete rotation freedom around through standard DB9 connectors which are the steering axis. System modularity was one embedded into the frame. It is an of the basic requirements for this robot. The omnidirectional vehicle that has been designed wheels can be removed and replaced very for highly unstructured grounds. This platform easily because all the wheels are identical and could be used to detect mines in areas which are fastened with fast screw connections. The are dangerous for human deminers.

3. Control Architecture

3.1 The components of the control architecture

The whole system has a multi-processing architecture and comprises the following components (figure 6): - The HMI computer, - An embedded computer for data acquisition and communication with the HMI computer, - The motion controller (microcontroller), - The visual tracking and location computer.

Figure 6 The general system architecture

The following function are distributed among the different processing systems:

- Motion control

27 - Processes Synchronisation - Data Acquisition

In this application, two kinds of interfaces are friendliness is reached through the use of needed: communication interfaces (serial, efficient programming techniques (see section Ethernet) and acquisition interfaces (serial, 3.7). GPIB, frame grabber). The 32-bit microcontroller is very versatile and The subsequent subsections give and overview powerful; it performs different functions of the different components of the control depending on the system: architecture. The last subsection describes the integration and synchronisation of the different - It controls electrical motors (scanner, systems and processes. Tridem), - It monitors signals coming from contact switches, 3.2. The Master PC - It generates commands (replacing a PLC) for controlling the valves on the The control architecture comprises two loops: pneumatic robot, a high level control loop that runs on a PC and - It generates analogue signals for the tracks synchronises high level processes and a low of the caterpillar vehicle, level control loop running on a microcontroller - It hosts the main communication loop with which performs common control functions the Master PC. (position and speed control). The Master PC is responsible for the synchronisation of the robot The specific characteristics of the control motion with the positioning system and for the methods are given in the remainder of this synchronisation of the scanning with the data subsection. acquisition. It sends configuration and trigger What concerns the Hunter, two control levels commands to the microcontroller trough a can be distinguished: a software only loop that serial link and performs the data acquisition. manages communication through the serial interface and the motion control functions The role of the Master PC can be played by the which are interrupt-based. The serial port is HMI PC, when it is directly connected to the polled and a switch-case instruction fires the vehicles and to the sensors, or by the function corresponding to the received code. embedded PC which communicates via a Main high level control functions for the TCP/IP link with the HMI PC. This remote Hunter are: set scanning limits, set position, set acquisition can for the moment only be speeds, move to area limits, start scanning ... performed with the metal detector (RS232 During the scanning, synchronisation signals interface). The direct connection is required are sent to the Master PC to trigger the data when acquiring data from the GPR because it acquisition. Two scanning modes are is an instrumentation system that cannot be available: a continuous mode and a discrete easily displaced. It uses a normal PC HPIB mode. In the discrete mode, the scanner stops interface card connected to a scope for at every step waiting for the end of the data acquiring data. When a portable sensor is acquisition. The acquisition step can be available, it will be possible to easily integrate changed in the set-up dialog box of the HMI. it into the mobile system. In the moveTo function, the desired goal co- ordinates are copied into shared variables that are read by the motion control functions. These functions are called by interrupts at fixed 3.3. The motion and scanning intervals (a 4 ms control period is used). The controller PI controller controls the motor speed by using the Time Processing Units (TPU) of the The three robotic systems use the same basic microcontroller which are configured to electronics; the low-level control is realised generate PWM signals. A cascade negative with a 32-bit microcontroller that feedback control combining two loops has communicates through a serial link with the been implemented: the first loop corresponds Master PC. They can be remotely controlled to a speed negative feedback control (the inner through the same HMI, a program called loop) while the other one is for position CoRoDe (Control of Robots for Demining). negative feedback control (the outer loop). The HMI manages all control aspects: vehicle The microcontroller also generates analogue and scanning control, communication and data signals to drive the tracks of the Hunter. Two acquisition. A high interactivity and user- TPU's channels produce two independent

28 PWM signals for right and left tracks. The controller. The TPU's generate six PWM PWM signals are filtered to generate signals and capture the digital encoders' signals continuous bipolar analog signasl compatible from the steering motors. The speed of the with the old interface. As the driving motors driving wheels are measured with tacho- do not have encoders it actually acts as an generators whose signals are converted by the open loop controller. internal AD converters. A classical PI algorithm is used for controlling the wheels. In the case of the walking robot, the microcontroller is essentially used as a Different configurations are possible (figures software PLC. The control is implemented like a finite state machine and all outputs are binary 7a, 7b and 7c respectively): signals. The TPU's are used to read the position signal from incremental linear sensors - Translate: all wheels are parallel, placed along the pneumatic cylinders. The - Spin: the wheels are tangent to a circle, main loop manages the communication with - Free: the normal motion mode in which the master PC. the three wheels are separately commanded to follow arbitrary The control of the Tridem is similar to the one trajectories. of the Hunter but requires all TPU's of the

This later mode is preferred for the detection because it keeps the metal detector pointed along the motion direction.

Figure 7a Translate mode Figure 7b Spin mode Figure 7c Free mode

3.4. Sensors and acquisition scanning process and the way the control is realised. interfaces

Three different sensors have been successfully 3.5. Communication used in the project: a metal detector (MD), a Ground Penetration Radar (GPR) and an A serial communication allows the infrared camera. A fourth one, a pyrometer, transmission of commands between the Master has been tested but abandoned due to PC and the microcontroller (the transmission unsatisfactory results. The data acquisition speed is 9600 baud). Radio Ethernet links process requires different interfaces: the metal (protocol 802.11) are used to communicate detector has a serial interface, the GPR data are between the HMI PC and the embedded PC. read through the GPIB interface of a high speed oscilloscope, the images coming from 3.6. Location and tracking the infrared camera are captured with a frame grabber (through a cable or a wireless A location system determines the robot's connection). We will see in the subsection 3.7 positions which are used to automatically how the sensors' characteristics influence the generate a map of the detections (Hong 2001). A pan-and-tilt colour camera tracks a coloured

29 target mounted on the vehicle. Every frame the of the camera is also known, the ball is ball object is extracted from the image (figure completely located in the field. In favourable 8) and an estimation of its position is conditions (uniform background, constant computed. As the robot moves, the camera illumination) a precision of 30 cm at distances follows the target in order to keep it in the up to 15 m is reached, which is enough for this centre of the image. Furthermore, the size of application. In order to increase the precision the target is kept constant by using a motorised of the location process in varying conditions zoom and the value of the zoom gives an the coupling of the camera with a laser estimation of the distance. As the orientation telemeter is now considered.

Figure 8 Colour target identification

3.7. Human Machine Interface ƒ One-dimensional data merged into a 3D volume (GPR). The graphical user interface of the control The map contains a reference frame and program CoRoDe is shown in figure 9. This consecutive positions of the robot and scanned program offers the following functions: areas. The user can switch between either - Control of the vehicle, configuration by clicking on a button in the - Configuration and control of the scanning toolbar. system, - Configuration of the sensors, Data acquired during the scan process are - Data visualisation, saved in two different formats: first as binary - Data archiving, data for later processing (double for GPR, - Mapping. double word for MD) and as 8 bits grey scale The main area is devoted to data visualisation. raw images for direct visualisation. The data It can alternatively display a map with acquisition, scanning, location computation successive locations of the robot or the sensors' and vehicle motion are integrated into a data. sequence that is controlled by the user with Displayed data can result from two different button commands lying in a single toolbar. The acquisition methods: interface is simple and intuitive thanks to the - Global acquisition: use of well-known symbols (VCR-like) and ƒ Infra-red images standard colours (see figure 8). In this ƒ Video images (used to view and application, it was a requirement to let the user record pictures from the scanned area) keep the control of the process; at every - Sequential acquisition: moment the user can pause, resume or stop the ƒ Single values organised as a 2D operations. image (MD)

30 Figure 9 The CoRoDe data visualisation window

It is also essential to provide information the force-feedback joystick drivers, now during internal processing or timeouts. In this Windows2000 or XP could be used as well). case, sensors' data are drawn on the screen as The communication with the microcontroller the scanning progresses. The position of the runs in its own thread in order to keep the scanner relative to the maximum positions, the interactivity of the user interface. Each event status of the scanning sequence and the main coming from the microcontroller (begin or end options are also presented to the user and of line, position trigger for data acquisition) regularly updated. Finally the use of additional through the serial port is intercepted by the communication threads (see next section) communication thread. For each event, preserves the interaction with the user corresponding messages are posted to the view interface. The next section provides window (the program is implemented with the implementation details about the MFC document-view scheme). Each message communication and the synchronisation of results in a call to the corresponding function processes. of the window class (examples of such messages are acquiredata, savedata,...). When 3.8 Control Process, using the discrete scanning mode, this process is synchronous and the microcontroller blocks communication and until it receives the acknowledgement of the synchronisation main process. The consequence is that the scanning is not continuous but stops at every The robotic systems used for sequential data step (typically every 2 cm). This was required acquisition have a more complex control by the metal detector because when triggered, architecture than the wheeled robot. The this sensor has an acquisition latency of about division of the control between the 200 ms which corresponds to 2 cm for a speed microcontroller and the Master PC requires a of 10 cm/s and consequently to a 4 cm shift permanent synchronisation of the processes. between the forward and backward scanning The main program runs on a computer under lines. This synchronisation also removes the the Windows operating system (OS) need to take into account the delays introduced (Windows98 was used because at that time it by the OS message queuing mechanism and was the only windows version compatible with the acquisition duration. The communication

31 thread then enters in a blocking section and Master PC because the communication process waits until the data acquisition has been is essentially asynchronous. The performed and the lock released to send the microcontroller receives configuration or acknowledge message to the microcontroller. motion commands and executes them. An acknowledgement is sent after the reception of Realising the communication between two the command, the execution is started and a threads by messages allows to completely new command is read from the serial port. uncouple them (the other way is to pass objects' pointers to the control thread and to directly call functions from it). At each step, 4. Results and future work the position of the scanner in the user interface is updated, data is acquired, plotted and saved. The two systems equipped with scanners have been successfully used to test the GPR However, this asynchronous control has two prototype and to acquire registered multi- drawbacks: the scanning process is slower and sensors data (figure 10). These data have been it can produce vibrations due to resonance very valuable for the searchers working on phenomena. As mentioned earlier, the first fusion methods. The wheeled robot has also limitation is also due to the sensors and cannot been demonstrated as a remotely controlled be avoided without changing the sensors mine detection system. themselves. Introducing random delays to avoid resonance frequencies can eliminate the Different elements have contributed to the second problem. Other possibilities for dealing success of this project: with the acquisition latency is to calibrate the delays and to shift the data every two lines at - The use of the same microcontroller for the end of the acquisition or to always scan in different robots, the same direction. The first solution was - The use of object-oriented programming rejected by the data processing team because it techniques and language that allow to introduces some extra pre-processing and can easily reuse building blocks in developing alter the data. The second solution has as new applications or modifying existing consequence to double the scanning duration ones, and to cancel the benefit of the continuous - The hierarchical model of the control, scanning; it was consequently not considered. - The user-friendly user interface. What concerns the Tridem, there is no such synchronisation between the robot and the -

Figure 10 Data acquisition using the prototype GPR and a metal detector

32

But despite these benefits, the application presents several limitations: 6. Final conclusions. Need for

- It is a monolithic application where Robotics in Humanitarian control and interface cannot be easily Demining ? separated, - It is OS specific, There exists a big difference between the - Adding a robot with new capabilities or a military and the civilian mine clearance. The sensor with new display requirements military demining operations accept low rates would demand important modifications of of Clearance Efficiency (CE). For these the program. purposes it is often sufficient to punch a path through a minefield. For the humanitarian These drawbacks are not unique for this demining purposes, on the contrary, a high CE application but are common characteristics of is required (a CE of 99.6% is required by UN). classic programs. This can only be achieved through a ‘keen New ways for implementing the control and carding of the terrain and an accurate scanning visualisation software are consequently of the infested areas’ that implies the use of considered: components and distributed sensitive sensors and their slow systematic components like CORBA or Java RMI are displacements, according to well-defined possible candidates. The requirements for the procedures or drill rules, on the minefields. new architecture are listed below. It should: The robots, carrying the mine-detectors, could play here an important role. - Provide an universal User Interface, The automatization of an application such as - Have Plug & Use capability, the detection and removal of AntiPersonnel - Be open to existing and future robots, mines implies the use of autonomous or - Allow the concurrent control of different teleoperated mobile robots following a robots, predefined path, sending the recorded data to - Allow the easy reconfiguration of the their expert-system (in charge of processing application (addition or removal of the collected data), marking the ground when a components: robots, sensors, users), mine is detected with a probability of - Allow several users to collaborate, predefined level and/or, possibly removing the - Allow the users to incorporate their own detected mine. control algorithms. This automatization is unrealistic : the technologies allowing it exist, but the integration of those technologies in a Robotic 5. First Conclusions System moving in unpredictable outdoor environmental conditions is not yet mature. This summary has presented the results achieved by the Robotic Workgroup within the Several workshops, a.o. organised by the IARP Belgian funded project Hudem. Robots using (International Advanced Robotics Programme) different locomotion techniques have been allowed a discussion on the possible R&D successfully used to acquire sensor data and to activities for solving the problem(s): the test and validate utilisation methods in robotic systems was not felt as the most different scenarios. In this summary, the promising solutions, due to their high cost, the control and programming architectures of use and maintenance difficulties, the varying these systems have been described. The use of (daily changing) terrain conditions, etc. object oriented techniques for application However, specific tasks could be entrusted to development and the reuse of the same control mechanical mine disposal systems (or, if hardware have contributed to the success of efficient, robotised sensor-carriers): this project. However, in the next future new programming techniques will allow developers 1) the cutting of the vegetation : it’s a to write more open and reusable applications. mechanical (tele-operated or not) At this stage, it has not been proven that assistance that doesn’t need research robotic detection of mines works better or activities, but adaptation of existing faster than human deminers. But the obtained mobile cutters, results are encouraging and pave the way for 2) the detection tasks in very dense and an integrated solution that will some day help dangerous areas : such tasks could to solve this terrible plague. imply the realisation of effective multi-

33 legged robots, a difficult long-term I also want to thank my assistant, technical challenge, director of our UGV (Unmanned Ground 3) the delimitation of the borders of a CeVehicle) center, Eric Colon, who composed suspected area : task that could be the major part of this paper. entrusted to aerial tools : the EU will probably encourage the R&D focusing on this task: UAVs (Unmanned Aerial References Vehicles) could be useful for this aim. 4) the inspection of an area after manual Alexandre P., Weemaaels J., Doroftei I. & demining or mechanical clearance. Colon E., (1998) Development of a high mobility wheeled robot for humanitarian mine We are convinced that both tasks 2 and 4 may clearance, Proc. Robotic and semi-robotic be entrusted to Robotics Systems for so far it ground vehicle technology, Aerosense - SPIE, will be taken into consideration with the next Orlando, USA. recommendations: Baudoin Y. & Colon E. (1999) Humanitarian 1) No any R&D may be conducted without a Demining and Robotics: a difficult challenge, permanent realistic contact with the End- Proc. Clawar99, Portsmouth, UK. Users 2) No any ‘dummy’ minefield may lead to Colon E., Baudoin Y., Alexandre P. (1998) realistic trials: experiences conducted in Development of mobile robots for mines laboratories on ‘sandboxes’ distort the detection, Proc. Mechatronics98, Skovde, comprehensive perception of the Sweden. problems. 3) R&D AND parallel trials on real Hong P., Sahli H. & Baudoin Y.(1998) Color minefields are needed, even limited to Target Detection and Tracking, Proc. Clawar minor betterment of existing techniques ‘98, Brussels, Belgium (example: adaptation of existing agricultural vehicles for cutting the Hong P. (2001) Visual servoing for robot vegetation): some companies , NGOs , etc. navigation: application in humanitarian are ready to welcome and test proposed demining, Ph. D. Thesis, Free University of solutions. Their involvement from the Brussels, Belgium. beginning of a project to its end must be seen as a warranty of success. Milisavljevic N. (1999) Mine shape detection 4) Improvement of existing sensors by well- and data fusion considerations, Proc. Hudem trained local demining teams and Symposium, RMA, Brussels, Belgium. temporary help of the teams by well- accepted and optimally interfaced (HMI) Scheers B. (1999), Development of a small light-weight, easy controllable and laboratory UWP GPR system for maintainable mechanical systems for well- minedetection, Proc Hudem Symposium, defined tasks (vegetation cutting, sensor RMA, Brussels, Belgium. carrying in high-dense minefields, pre- inspection or post-inspection of a suspected or cleared area, etc.) are welcome for so far they improve the cost- effectiveness and the safety of the demining operations.

Acknowledgements

This project would not have been possible without the contribution of all the searchers involved in the Hudem project and without the funding by the Belgian Ministry of Defence and the Secretary for Cooperation and Development. We also want to thank our partners from the European Network CLAWAR (Climbing and Walking Robotics) and from the WG Hudem of the IARP.

34 Demining in the Indian context

Akshay Rao Goldspark Bangalore

Over the last Century, India has suffered many conflicts, though the wounds of war may have healed but the danger still lies buried in the ground, waiting to kill and maim even years after the conflict is over. These are the true wounds of war and require a concerted effort by the world community to heal them.

India has a long and hostile border where extensive mining and de-mining operations are held due to continued escalation and de-escalation along the border. This as the rest of the world understands is an unjustified problem, but in the race to balance powers across our border this is one of the ways which are justified in stopping hostile personnel from crossing the border into India.

A great part of this border is mountainous and rocky with lots of caves and rock faces where it is possible for hostile personnel to hide. Kargil and Afghanistan are examples of this. Surface penetrating radar can find these, thus helping the location of these hostile personnel India and Pakistan have laid large numbers of such mines along their common border since coming close to war over Kashmir in December 2001, the International Campaign to Ban Landmines said in a report. India is not a State Party to the Mine Ban Treaty and uses and produces antipersonnel mines. At 4-5 million it has the fifth largest stockpile of the weapon in the world. Since December 2001, Indian Government forces have been involved in what appears to be the largest mine-laying operation in years and possibly decades.

Following the attack on the Indian parliament on December 13th, 2001 India began the largest mine laying operation along the 2900 km border. Initially only the 440-kilometre Line of Control was mined but following December's attack on parliament, the international borders have also been mined. The regions of mining in India are

1) Mountainous part of Kashmir Valley,

2) Plain and fertile land of Jammu & Punjab

3) Desert of Rajasthan. 4) Landmines are also used heavily by the "Naxalites" in Andhra Pradesh, Unfortunately in the Indian context, India and Pakistan are respectively the fifth and sixth largest landmine countries in the world and have an estimated stockpile of about 11 million landmines between them. Parts of the border between the two countries are already regarded as some of the most densely-mined areas in the world. India and Pakistan have not signed the 1997 Ottawa Treaty banning anti-personnel mines .The United Nations estimates that landmines still kill about 10,000 people a year around the world, and activists said the devices injure about another 10,000.Up to 40 percent of all mine victims are children under 15, according to the United Nations

The above statistics are not new, but show how far we have gone into making a “mess” for ourselves. This problem can be solved technologically in a way that does not jeopardize human life. At the current moment there are many technologies that can be used for this purpose but the world is still to a certain extent “groping in the dark” .This groping leads to technological improvements. Like the television, internet and so many other technologies we know, it only gets better, and the vision of improving our “today” leads us into a better tomorrow.

21 There are many technological aspects of this problem. Like the development of many other robots, the robotic de-miner will have to undergo much iteration to get to a more efficient model which can rid the world of landmines – in an economical way. Goldspark is developing these robotic models with our eye on the latest technology. There are two aspects of the de-mining problem. Firstly location of the mine and secondly the defusing of the mine. For the first there are many ways which are being used. Here again we answer another question, do we just locate and bypass the mine or do we also defuse it? Goldspark is developing track vehicles to be used in robotic de-mining. These will be tracked vehicles to help them negotiate uneven and uncertain terrain. Thanks to the help and guidance given by Prof James Trevelyan and Prof. Claudio Bruschini whose advice has been provided within the framework of the EC EUDEM2 project, we are in the process of making a technologically superior vehicle for this purpose.

There are many interesting ways of approaching this problem.

The development of the Fireant is an example of how the defusing technology improves itself with a proportional reduction in the cost. Bringing down the cost from 1000$ to 15$.also Goldspark is looking at chemical neutralization which will render the mine inactive through chemicals being injected into it. The chemical neutralization will be further achieved by means of dispensing mechanisms which we will mount upon the mobile robot vehicle in order to render the mine useless. The reason Goldspark looks at chemical neutralization is that the army claims that always blasting the mine causes noise which could alert the enemy in a hostile scenario. Therefore chemical neutralization and other non-blasting techniques can be employed. The advantage of this is that the mine is destroyed without actual blasting thereby preventing alerting the enemy in the battle scenario.

Also to be used in our de-mining vehicle is a Low-Depth Ground Penetrating Radar. Such de- miners can also be used in their idle time to locate other objects buried in the ground. Our development process would essentially consist of the following. After the conceptual design, performing of Finite Element Analysis on various aspects of the vehicle and its motion, capability of taking shocks and vibrations, also done are necessary analyses for shielding for all parts of the system requiring it. In addition to this the design and validation of the motion control and sensing systems and their integration into the main system, to be accompanied with the adequate design of the Robot Arms, Dispensing mechanisms and other parts which facilitate de-mining. Most important is the final integration of the whole system at which stage the entire system, functions to actually sense the mine and defuse it

The Indian Army currently uses conventional de-mining techniques; these are well known as causes of death and incapacitation. Currently the Army is involved in Operation Sarp Vinash, which is to block infiltrators from across the border. This has taken a deployment of an additional 10,000 in two regions of our northern state Efforts are on to introduce technology into this process. We are uniquely placed technologically to handle developments in the field of robotic de- mining. We have been acknowledged the world over for our software prowess. Having companies which possess the capability of making advanced systems requiring an integration of the nature which is required for this product, India already designs and manufactures transmitters and receivers covering from metric to millimetric ranges of the electromagnetic spectrum. Further is the advantage of being strategically placed in the subcontinent to lead this effort in this region. This part of the world is infested with landmines and India can become the leader in this context. The immediate neighbours of India will also be able to benefit from the lead that India will take in this regard. In India, we believe that we can take this lead as this is one of our focus areas. A solution which works in one region can be implemented in another region too. In summary we can say India is one of the countries largely affected by land mines. De-mining in India is being done by the conventional method. Once the lead is taken we can surely reduce the death rates by introducing robots in de-mining process.

Goldspark contains the team of experts in this context who can make de-mining an easy task by implementing the latest technologies in an economical way.

22 Humanitarian demining: sensor technology status and signal processing aspects

Marc Acheroy Dept. of Electrical Engineering, Royal Military Academy Avenue de la Renaissance 30, B-1000 Brussels, Belgium

Abstract— This paper presents the status of sensor tech- very important for assessing the probability of an nology, including operational characteristics without aiming alarm in a given location of the minefield. at being exhaustive. Signal processing aspects and important

lessons on data fusion are also discussed briefly. The detection is considered as a global process in which the outputs of the The detection probability is the probability of having sensors, considered as skilled specialists, are integrated in a an alarm in a given position of a minefield for a fusion operation. given detection material, if there is a mine in that position. This probability gives indirectly a measure I. INTRODUCTION of the non-detection probability of that material as A huge amount of antipersonnel (AP) mines are pol- well. luting the environment in about 60 countries. Thanks to the Ottawa Convention, mine clearing operations have The probability of false alarm is the probability of been organized in a more controlled and effective way. having an alarm, for a given material, in a given Nevertheless, mine clearance remains a very slow process. location if there is no mine in that location. It is estimated that, on average, a deminer is clearing an area of 10 m2 every working day if he is using conventional tools, i.e. metal detectors and prodders. To The two latter definitions are extremely important to give an idea of the extent of the problem, in Cambodia, understand the humanitarian demining problem and for only approximately 260 km2 have been cleared during designing demining systems. the last ten years. Therefore, humanitarian mine clearance It is indeed particularly important that the detection prob- operations must be understood and designed correctly, ability should be as close as possible to one. It is easy keeping in mind that their main goal is to provide efficient to show that evaluating the detection probability also aid to innocent people, who may be severely injured by amounts to evaluating the risk of the occurrence of a this dreadful pollution. mine which has not been detected. This risk is concerned Furthermore, the analysis of actual demining campaigns with human preservation and is therefore of the utmost not only reveals the far too long time needed to clear importance. No such risk is acceptable and it is therefore polluted terrain, but also brings to the fore a far too an absolute requirement that a demining system should large false alarm rate, the threat of plastic mines (which decrease the probability of such a risk to the lowest upper are difficult to detect by classical means i.e. by metal bound possible. detectors), and the large variety of mine clearance Besides, although one indirectly saves human lives by scenarios, depending on the country, the region, the decreasing the false alarm risk thanks to the acceleration climate and the place of the pollution (houses in villages, of the demining operations, the false alarm risk is also roads, agricultural fields, etc). a question of cost. Indeed, a demining method which minimizes the false alarm rate results in an acceleration The important parameters, which characterize the mine of the demining operations which results in spending less detection problem, are the mine occurrence probability, money. the detection probability of a given material and the false Therefore, any demining operation enhancement must alarm probability of a given material [9]: result in the highest possible detection probability (close

The mine occurrence probability in a given position to one) and in the smallest possible false alarm rate and of a minefield expresses the local mine density of that at the lowest price. Generally, it is accepted that the that minefield as well. Obviously, it is impossible most efficient way for increasing the detection probability to control this parameter because it depends on the while minimizing the false alarm rate consists in using reality of the terrain. Nevertheless, this parameter is several complementary sensors in parallel and in fusing the information collected by these sensors. and still need a lot of developments. The second step As a matter of fact, it is imperative to evaluate the de- is sketched out in section 3 and consists in fusing the tection probability when optimizing the performances of a information produced by the different sensors with their system. However, the detection probability, as it is defined dedicated processing tools. before, assumes that a mine is present in the considered position. Since, during organized trials, the position of the II. SENSOR DESCRIPTION mines is well known, the condition of the occurrence of In this section, it will be tried to describe succinctly a mine in the given position where the performances of sensors of different types without claiming exhaustivity. a system must be evaluated is always realized. This latter we will subdivide the description into four categories: remark is of particular importance because it justifies the 1) Prodders, seismic and acoustic sensors organization of trials and the construction of models, to 2) Electromagnetic sensors (Metal detector, GPR, be validated by trials, in order to evaluate the detection Micro-wave radiometer, Electrical Impedance To- probabilities. mography, Electrography, Imaging with handheld Furthermore, assuming in the following as the first sensors) approximation that the sensors are independent1, the 3) Electro-optic sensors (visible, IR, multispectral, hy- detection probability can be maximized by optimizing perspectral, LIDAR) separately the design of each sensor and of the associated 4) Explosive detectors (NQR, X-rays, Neutron activa- signal processing. Next, it can easily be shown that the tion, Biosensors, Trace explosive detection) detection probability increases if the number of different The three first sensor categories are not able to discrim- sensors increases and that maximizing the overall detec- inate between an explosive material and any material tion probability of a set of independent sensors clearly with the same electro-magnetic, thermal and/or optical comes to the same as maximizing the detection capabilities properties, but often offer good localisation capabilities as of each individual sensor. This justifies the use of several well as 2-D and even 3-D capabilities for some of them. complementary sensors and of data fusion techniques to The last category aims at detecting explosive material, increase the detection probability. Among the most cited often offers poor localisation capabilities and often lacks sensors one finds the metal detectors, the radars and the for spatial resolution as well as for 2-D or 3-D capabilities. infrared sensors. A lot more are presented in section 2. Most of the time these sensors require a long integration Finally, the false alarm risk, i.e. the probability of time, which makes them more suitable as confirmation having an alarm if there is no mine, cannot be as easily device. In the latter case, they are used in combination evaluated as the detection probability because of the use with sensors of the three first categories. of data fusion methods which favor the manual or auto- For each of these categories, a table will describe for matic cancellation of false alarms. Furthermore, it is very each sensor its status of maturity (“R&D”, “in devel- difficult to evaluate the risk of false alarm because it is opment, “in use”), its cost (“Low”, “Low to medium”, very difficult to define in a general way what is not a mine. “High”, “Very High”), its clearance speed (“Low”, “Low In this context, it should be particularly inappropriate that to medium”, “High”) and its effectiveness (“Unknown”, a demining system, whatever it may be, makes decision “Low”, “Low to medium”, “High”). instead of the final user whose own physical security is involved. Therefore, a well designed system should help A. Prodders, seismic and acoustic sensors the user in the decision making, not by replacing him, but Manual prodders can also be enhanced by addition to by implementing efficient data fusion methods. For this them of an ultrasonic sensor at the prodding extremity. purpose, methods which are able to deal with uncertainty This so-called “Smart Prodder” allows to have a better by making proposals including the doubt to the user seem guess of encountered objects and to exercise less pressure to be promising. on these objects (in this way, decreasing the risk of This paper will only consider close-in detection in two accidents). steps. In the first step sketched in section 2, will consists Seismic devices are listening to the response of the in describing potential technologies for mines detection ground to a shock applied at the ground surface from a and, when appropriate, their associated processing. Some safe position. of these technologies are already in use (e.g. prodders and In the case of an acoustic sensor, an ultra-sonic wave is metal detectors). Some will be fielded very soon (e.g. the sent into the ground and the backscattering of this wave ground penetrating radar (GPR)). Others will only appear against buried objects and the environment is analysed. in the field in the long term because they are not mature The application of this method is limited to soil with a very high level of moisture or in the water (paddy 1the independence is a particular case of the complementarity: two in- dependent sensors are complementary, but the contrary is not necessarily fields) for which the propagation of ultra-sonic waves true. is possible. This latter characteristic makes ultra-sonic sensors complementary to the ground penetrating radar (GPR), the performances of which are optimal when the moisture level of the soil is low. The operational characteristics of these sensors are described in table II-E.6.

B. Electromagnetic sensors 1) Metal detectors: There exist three families of metal detector: the first one, based on electromagnetic induction (EMI), sends a primary magnetic signal in the ground in an emitting phase during which it creates eddy currents in the buried metallic objects which in turn create a secondary magnetic field. During a listening phase, the emission is Fig. 1. Metal detector — deconvolution of a metallic ball at 5.5cm and stopped and the system listens to the secondary magnetic 8.0 cm field which induces eddy currents in the coils of the detector. These currents are characteristic of the buried metallic objects and of the soil. There exist two types of EMI devices: the first one sends a magnetic pulse, the second one a continuous wave at different frequencies in a stepped frequency mode. In the second family, the detector, called magnetometer, mesures the local perturbations of the earth magnetic field. In the third one, the detector, called gradiometer, mea- Fig. 2. Metal detector — deconvolution of a straight metallic wire sures the magnetic field gradient in a given direction depending of the sensor configuration. The most used family of detectors is the first one, based two-dimensional horizontal slice (parallel to the ground on EMI. surface) in a set of adjacent B-scans. Surprisingly, the metal detector of the first family The GPR includes an emitting system (transmitter) and (which is the most common detector), considered as an a receiving system (receiver). The transmitter emits a pulse imaging device, can also provide very useful information wave or a continuous wave at given frequencies. The re- on the shape of metallic pieces included in mines. ceiver collects the waves backscattered by discontinuities Unfortunately, the point spread function (PSF) of a in permittivity. Discontinuities can be provoked by buried metal detector is a function of the depth (see Fig. 1) and of objects like landmines (useful signal) but also by natural the nature of the buried metallic object (eddy currents are discontinuities of the soil (clutter). This means also that a different in a close and in an open circuit) and the image GPR is able to detect plastic objects buried in the ground. formation process is non linear. However, the in depth There are mainly two important types of GPR depending modeling of the metal detector behavior as a function of the emitted signal: the first one sends a short pulse into the type of buried object by the RMA [13], has shown the ground (Ultra wideband pulse GPR), the second one that it is possible to derive the depth of a buried object sends a continuous wave in a stepped frequency mode. The from the original data and thus to derive the corresponding advantage of the second type is that it provides directly PSF to allow a correct de-convolution (see Fig. 2, 3 the Fourier transform of the received signal and that more and 4). Further, information on the symmetry properties energy can be send into the ground at a given frequency. of the buried metallic objects can easily be extracted. Ground penetrating radars and passive radiometers are This subject is still under investigation. This interesting intended to function as anti-personnel mine detectors. consideration shows that the metal detector, known as a Their performances depend upon parameters such as type cheap mine detection system, remains a promising device. and texture of soil, soil water content, soil density and 2) The ground penetrating radar (GPR): Useful defini- operating frequency. In order to evaluate the performances tions to understand what follows have to be given first. An of microwave technologies in land-mine detection, the A-scan is a one-dimensional signal taken perpendicular to electrical properties of soils must be extensively evaluated the ground surface and is the basic echo signal produced (see [10]). by a GPR. A B-scan is a two-dimensional signal resulting Current GPR are working in a frequency range com- from a collection of adjacent A-scans along a straight prised approximately between 0.4 and 6.0 MHz. line horizontal to the ground surface. A C-scan is a Fig. 5 shows a sample A-scan of a PMN mine in loam. Fig. 3. Metal detector — deconvolution of a metallic open loop

Fig. 4. Metal detector — deconvolution of a metallic closed loop

Fig. 6 presents two B-scans of a PMN mine respectively in loam and in sand and a C-scan of a PMN mine in loam at 4cm depth.

Fig. 6. From top to bottom, PMN B-scans in sand and in loam and a Fig. 5. Example of a PMN A-scan in loam PMN C-scan in loam

One advantage of GPR systems is that it is possible capabilities. In the scope of the HOPE project2, The DLR to become a 3-D representation of objects buried in the has developed a MWR which is capable to work at more ground. In order to recover the correct 3-D shape of buried than 32 different frequencies between 1 and 8 GHz. An objects, RMA [4][5][6] has developed algorithms based on example of images obtained with a manual scan is given the convolution, by modeling the behavior of the GPR in in Fig. (8). the time domain. The developed algorithms are faster than 4) Other electromagnetic sensors: In Electrical the classical migration methods and provide better results impedance tomography, the soil impedance is measured as it can be seen on Fig.7. between selected locations on the ground. By solving a non-trivial and non-linear inverse problem, it is possible 3) Microwaves radiometers: The microwave radiome- to detect anomalies. ter is a passive ground penetrating radar (only receiving In electrography, the corona effect is used to detect antennas are present) which uses the EM waves (a few K) explosives (typically TNT) in a liquid phase. emitted by the sky and reflected on the ground surface and The operational characteristics of electromagnetic sen- subsurface. They also can generate clear two-dimensional sors are described in table II-E.6. images of surface, shallowly buried and buried objects (metallic and plastic). The spatial resolution and the 2the HOPE handheld system, project funded by the European Com- penetration depend on the frequency. As for the GPR, the mission, includes a metal detector (Vallon, GmbH & RMA), a stepped frequency GPR (RST, GmbH) and a multifrequency MWR (DLR), all performances are also depending on the soil conditions: with imaging capabilities through the use of a high precision positioning a high level of moisture can largely limit the detection system (RMA) recording by means of a linear polarizer four different im- ages corresponding to four different polarization directions [15].

Fig. 7. From left to right: raw GPR 3-D image of a PMN mine, restored 3-D image of a PMN mine and restored 3-D image of a 10cm barbed wire with three barbes

Fig. 9. From left to right: original image of a buried mine (5 cm below the ground surface), preprocessed image, detected edges and original image with detected mine-like shape (ellipse)

3) Scanning Laser Doppler Vibrometry (SLDV): In Fig. 8. DLR microwave radiometer images of a metallic plate, obtained with a manual scan, at 8 different frequencies between 1 and 6 MHz this technology, which is not properly said an electro- optic technology, an acoustic power transmitter sends an acoustic wave in the ground. If an object is present in the soil, at the ground surface a backscattered wave induces C. Electro-optic sensors soil vibrations measured by a laser Doppler vibrometer. Classical cameras have a poor detection capability even This technique has been tried by FGAN (Germany) on a for mines laid on the ground and shallow buried ob- test site in ISPRA (European Commission Joint Research jects. LIDAR and therahertz imaging systems have still Centrum). Results are shown on Fig. 10 The operational to demonstrate their usefulness for mine detection. They indeed use shorter wavelengths than ground penetrating radar and hence suffer from significant limitations in soil penetration. Further, wild groing vegetation offers a strong limitation to most of electro-optic devices. Nevertheless, special attention must be paid to hyperspectral, thermal infrared sensors and Scanning Laser Doppler Vibrometry (SLDV). 1) Hyperspectral sensors: Hyperspectral techniques take into account the very selective properties of the ma- terial reflectivity. Laboratory experiments[7], in the course Fig. 10. buried mines acquired with a SLDV (FGAN - Germany) of which very narrow wavelength bands have been used, have demonstrate the capabilities of wavelength tuning to characteristics of electro-optic sensors are described in discriminate between different surface laid materials. table II-E.6. 2) Thermal infrared: Mine detection by means of ther- mal infrared sensors can be achieved in two different D. Explosive detectors: biosensors approaches. The first approach consists in measuring the 1) Dogs and educated rodents: Actually, one of the appearing temperature difference of the soil, induced by most efficient “sensor” for mine detection is the dog. But the differences in emissivity and/or by the differences in it appears that rodents are easier to educate and to feed thermal flux due to the presence of a shallow buried or and that they can work longer than dogs. Furthermore, buried object (see Fig. 9and [12]). A second approach the rodents are much lighter and have a better olfactory consists to take advantage of the polarisation properties capacity and a better immunity. The non profit organiza- of manufactured surfaces, by analysing the information tion APOPO and the University of Antwerpen (Belgium) contained in the images of shallow buried mines produced are currently making operational tests in six different by the three Stokes parameters which characterize the po- countries, which are representative of the mine threat. larisation state, i.e. the degree of polarisation, the azimuth Rats have already booked interesting results in Tanzania and the ellipticity. Those parameters can be evaluated by in 2002. and functional groups in the molecules. Nitrogen is a quadrupole atom that appears in every type of explosive. Because of very distinct NQR frequencies the false alarm rate due to other nitrogen containing materials is extremely low [1]. Fig. 12 shows respectively a time responseof a NQR system and the corresponding spectrum.

Fig. 11. Rodent

Fig. 12. From left to right, time domain NQR response and correspond- ing spectrum 2) Artificial nose: Some reseach activities have been devoted to technologies that try to mimic the olfactory 2) Thermal Neutron Activation (TNA): Explosive ma- system of a dog. But up to now no significative results terials are rich in Nitrogen-14 (14N) which is the pre- have been booked in the field of humanitarian demining. dominant stable isotope of Nitrogen. When a neutron is Another approach consists in using antibodies sensitive captured by a Nitrogen nucleus the following reactions

to TNT. these antibodies are fixed on a quartz crystal. occur:

When the sensor is in contact with TNT free molecules, 14 15 15 γ ¢£¡ n N ¡ N N the fixed antibodies leave the crystal. The result is that the

weight of the quartz crystal is changed. This weight loss The excited Nitrogen nucleus (denoted by ¢ ) de-excites to is measured by measuring the crystal frequency change. its ground state in less than a picosecond by emitting one A relatively long integration time, due to the extremely or more gamma-rays with unique energies up to 10.83 low concentration of explosive vapours in the air, makes MeV. The gamma-rays can be detected by conventional this technology more suited to confirmation than to the detectors such as NaI and/or GeLi. Specific software can detection itself. Therefore, this technology should be ac- be used for signal processing and identification. companied by a detection equipment such as a metal 3) Fast Neutron Activation (FNA): Fast (14 MeV) detectors or a GPR or a combination of them. neutrons and the associated alpha particles (3.5 MeV) are The operational characteristics of biosensors are de- generated by a fast neutron source. The neutrons produce scribed in table II-E.6. prompt gamma rays in inelastic scattering with the nuclei of the materials. Each neutron is always produced in as- E. Explosive detection: nuclear and chemical methods sociation with an alpha particle which are always emitted This family of technologies includes Nuclear at 1800 to the neutron direction; hence, direction of each Quadripole Resonance, Thermal Neutron Activation neutron is known from the direction of each alpha particle (TNA), Fast Neutron Activation (FNA), Trace of associated with it. Alpha particles are detected with an explosive detection using chemical processes, X-ray array of scintillation detectors; the position of the detector backscattering and X-ray fluorescence. Again, the hit by the alpha determines the alpha direction which, relatively long integration time needed to detect the in turn provides the neutron imaging (see Fig 13). The explosive molecules and their high cost make this analysis of the gamma-rays provides information about technology more suited to confirmation than to the the stoichiometric composition of hitted materials in terms detection itself. Therefore, this technology should be of Carbon, Nitrogen and Oxigen. Depth information of a accompanied by a detection equipment such as a metal given neutron is provided by measuring the time of flight detectors or a GPR or a combination of them. of the corresponding alpha particules. 1) Nuclear Quadripole Resonance: Advanced Nuclear 4) X-ray backscattering: X-ray backscattered radiation Quadrupole Resonance (NQR) techniques can be used to is detected during active illumination of the ground with detect explosives in any surroundings. The quadrupole X-rays, and basically determines whether or not an object charge distribution of the atom results in alignments of is made up predominantly of light chemical elements (i.e. nuclear spins. A radio frequency pulse (RF-pulse) gener- low atomic number Z). The technique is intended for ated by a transmitter coil causes the excitation of nuclear bulk explosive detection, although AP mines have been spins to higher quantized energy levels. When the nuclear imaged as well; smaller, man-portable detectors based spins return to their equilibrium position, they follow a on radioactive sources have also been proposed. The particular precession frequency. This specifies the atoms systems which have been developed are said to be able Low level fusion can be performed even using a het- erogeneous set of sensors if the data are co-registered. Our experience has shown that higher level data fusion is possible but accounting for the following facts:

Learning processes are very difficult and risky be- cause of the inter and intra variability of the scenar- ios. The heterogeneous character of a given minefield and of the huge set of possible minefields makes Fig. 13. Image produced by a FNA system, the colours represent the generalisation unpractical if not dangerous. stoichiometric composition in C, N and O (image provided by courtesy High level fusion must rely on qualitative instead of of High Energy (CA - USA)) quantitative a priori knowledge, therefore methods like Bayes decision theory will often fail.

The absolute (objective) confidence in specific sen- to produce a 2D image with a resolution of some cm. sors resulting from extensive trials must be included Potential problems come from shallow penetration, system in the fusion model (principle of objective discount- complexity, sensitivity to soil topography, sensor height ing). variation, and safety aspects due to the use of ionising The relative (subjective) confidence that the deminer radiation [16]. has in specific sensors must also be (interactively) 5) X-ray fluorescence: X-rays are not penetrating included in the fusion model (principle of subjective deeply in the ground. Therefore, they are most of the discounting). time not detecting directly the explosive encapsulated in Since in this domain of application one has to deal with a mine but molecules of explosive which are migrating uncertainty, ambiguity, partial knowledge, ignorance and from the mines to the ground surface. When migrated qualitative knowledge, it is important to chose for an explosive molecules are illuminated by X-rays a series of approach where they can be appropriately modeled, e.g. changes occurs in the electron configuration resulting in belief functions within the framework of the Dempster- the emission of photons characteristic of the material and Shafer theory. A main motivation for working within this that can be captured and analysed. framework is to be able to easily model and include exist- 6) Chemical methods: Chemical methods are also used ing knowledge regarding: chosen mine detection sensors, to detect explosive in the air and in the ground. The diffi- mine laying principles, mines, and objects that can be culty consists in collecting enough explosive molecules in confused with mines [11]. the air or in solution because of the very low concentration In any case, we need to be aware that the ultimate of explosive available. decision must belong to the deminer because his life is The operational characteristics are described in table II- involved. E.6.

III. SIGNAL PROCESSING AND DATA FUSION IV. CONCLUSIONS For each type of sensor, specific signal processing The purpose of this paper is to present the status techniques are used in order to extract useful information. of sensor technology, including their operational charac- The used techniques mainly include signal conditioning or teritics without having the pretension to be exhaustive. preprocessing (e.g. signal detection, signal transformation, Furthermore, this paper presents the detection as a global noise reduction, signal restoration and enhancement (see process wherein the outputs of the sensors, considered as [2], [3] and [8], which are a very important steps before skilled specialists, thanks to their associated processing, further processing) and pattern recognition techniques can be integrated in a fusion process. aiming at increasing the expertise of each sensor sepa- rately. Nevertheless, it has been shown in the previous ACKNOWLEDGMENTS section that no sensor is perfect for all scenarios and all conditions (moisture, depth, cost, etc). The authors wish to thank all the researchers of the The analysis of the principles of operation of different HUDEM3 and MsMs4 projects without whom it was sensors, their complementary information, and the factors impossible to write this paper. that affect their operability, have led to the conclusion that their fusion should result in improved detectability 3The Belgian project on humanitarian demining (Hudem) has been and reduced number of false alarms in various situations funded by the Belgian Ministry of Defence and the Belgian State Secretariat on Development Aid (different types of mines, of soil, vegetation, moisture, 4The Multi-sensor, Multi-signature project is supported by the Euro-

etc ¡ ¢ ). pean Commission Joint Research Centre (ISPRA) Sensor family Sensor Maturity Cost Speed Effectiveness Prodder In Use Low Very Low High Prodders & Smart Prodder In Use Low to Medium Very Low High Acoustic In devel. Seismic & acoustic R&D High Medium High (in wet soil) EMI devices In Use Low to Medium Low to Medium High Magnetometer In Use Low to Medium Low to Medium High Gradiometer In Use Low to Medium Low to Medium High Electro- GPR In Use Medium to High Low to Medium High (in dry soil) magnetic MWR In Devel. Medium to High Low to Medium Medium Electrical Imp. Tom. R&D Low to Medium Low to Medium Unknown Electrography R&D Low to Medium Low to Medium Unknown Visible OK Low to Medium Medium Low Infrared OK High Medium Medium Infrared Polar. R&D High Medium Medium Electro-optic Prototype Multi & hyperspectal R&D High Medium Medium LIDAR R&D Very high Medium Low Terahertz R&D Very high Medium Low SLDV R&D Very high Medium Medium to high Dog OK Medium to high Medium to high Medium to high Biosensors Rodents In devel. Medium Medium to high Medium to high Artificial nose R&D Medium to high Medium Medium NQR R&D Medium to high Medium Medium Prototype TNA R&D High Medium Medium Prototype FNA R&D Very high Medium Very high Nuclear & X-ray backscattering R&D High Medium Low Chemical Prototype X-ray fluorescence R&D High Medium to high Medium Prototype Chemical detectors R&D High Medium Unknown TABLE I SENSOR OPERATIONAL CHARACTERISTICS

V. REFERENCES enth international conference on ground-penetrating radar, Kansas, USA, 1998. [1] A. Engelbeen. Nuclear quadrupole resonance mine [5] B. Scheers, M. Piette, A. Vander Vorst. The detection detection. In CLAWAR’98, pages pp.249–253, Brus- of AP mines using UWB GPR. In IEE Second In- sels, Belgium, November 1998. BSMEE. ternational Conference, The Detection of Abandoned [2] A. Pizurica. Multiresolution techniques for im- Land Mines, Edinburgh, UK, October 1998. IEE. age restoration in mine detection problems. In [6] B. Scheers, M. Piette, M. Acheroy and A. Vander CLAWAR’98, pages pp.225–230, Brussels, Belgium, Vorst. A laboratory UWB GPR system for landmine November 1998. BSMEE. detection. In GPR2000, Sidney, Australia, June 2000. [3] A. Pizurica, W. Philips, I. Lemahieu and M. Acheroy. [7] F. Dupuis. Acquisition d’images hyperspectrales. Speckle Noise Reduction in GPR Images. In Interna- Master’s thesis, RMA, Brussels, Belgium, 1998. tional Symposium on Pattern Recognition “In Memo- [8] L. van Kempen, H. Sahli, J. Brooks, and J. Cornelis. riam Prof Pierre Devijver”, Brussels, Belgium, New Results on Clutter Reduction and Parameter Februari 1999. Royal Military Academy, RMA. Estimation for Landmine Detection Using GPR. [4] M. Piette B. Scheers. Short-pulse response of anti- In GPR 2000, Eighth International Conference on personnel landmines to UWB GPR signals. In Sev- Ground Penetrating Radar, Gold Coast, Australia, Australia, May 2000. [9] M. Acheroy. L’infrarouge thermique, principes et applications au deminage´ humanitaire. Revue HF, (3):p. 13–24, 1998. [10] M. Storme, I. Huynen, A. Vander Vorst. Characteri- zation of wet soils from 2 to 18 GHz, experimental results. In CLAWAR’98, pages pp.237–239, Brussels, Belgium, November 1998. BSMEE. [11] N. Milisavljevic,´ I. Bloch, and M. Acheroy. A first step towards modeling and combining mine detection sensors within Dempster-Shafer framework. In The 2000 International Conference on Artificial Intelli- gence (IC-AI’2000), Las Vegas, USA, 2000. [12] N. Milisavljevic.´ Comparison of Three Methods for Shape Recognition in the Case of Mine Detection. Pattern Recognition Letters, 20(11-13):1079–1083, 1999. [13] P. Druyts and L. Merlat and M. Acheroy. Modeling Considerations for Imaging with a Standard Metal Detector. In Detection and Remediation Technologies for Mines and Minelike Targets V, volume 4038, pages 1431–1451, Orlando, FL, USA, April 2000. SPIE Press. [14] P. Verlinde, M. Acheroy, G. Nesti, and A. Sieber. First Results of the Joint Multi-Sensor Mine- Signatures Measurement Campaign (MsMs Project). In Detection and Remediation Technologies for Mines and Minelike Targets VI, volume 4394, Or- lando, FL, USA, April 2001. SPIE Press. [15] D.L. Jordan G.D. Lewis. Infrared polarisation signa- tures. In Proceeding NATO-IRIS Vol. 41 N5, pages p.71–82, BRUSSELS, 1996. NATO-IRIS. [16] EUDEM2 - Mine Action Technology website - http://www.eudem.vub.ac.be. TRAJECTORY TRACKING BY USING FUZZY LOGIC CONTROLLER ON MOBILE ROBOT

A. Shala, R. Likaj, A. Geca, A. Pajaziti & F. Krasniqi, [email protected], [email protected], [email protected], [email protected], [email protected]. Mechanical Engineering Faculty Pishtina - Kosova

Abstract: In this paper a Conventional PD Controller and Fuzzy Logic Control for trajectory tracking of mobile robot platform is presented. Using computed torque method is given input – torque on mobile robot wheels. The locomotion control structure is based on integration of kinematics and dynamics model of mobile robot base. The proposed Control scheme and Fuzzy Logic Algorithm could be useful for building an autonomous non-destructive testing system based on wheeled mobile robot. Key words: mobile robot, Conventional, Fuzzy Logic, trajectory tracking, platform, wheel

INTRODUCTION yC y

Recently, the development of engineering ωCz technology drives people to devise goods for xC convenience in everyday life, and the research utilizing the robots for the better of 1 r v the life is conducted in many areas. And Cy vC such robots are applied not only at the a C v industrial sites but also at the houses, Cx medical uses, and etc. However, due to the b diversity and complexity of the project 2 formation, different from the former robots, which performs simple repetitive works only, the robots with the ability to make appropriate decisions as to fit the situation ϕ x and those with ability to appropriate O correspond are needed. Thus, the system Figure 1: Kinematics coordinate system proposed in this paper was constructed in assignments. order that the mobile robot can follow desired trajectory. It was made possible by The nonholonomic constraint states that the using the Conventional PD Controller and robot can only move in the direction normal Fuzzy Logic Control. to the axis of the driving wheels i.e., the mobile base satisfies the conditions of pure KINEMATICS MODELING rolling and non-slipping: vCy cosϕ − vCx sinϕ −b⋅ωCz = 0 ...... (1) We have use for modeling mobile robot with From Fig. 1: two degrees of freedom (DOFs): x - vCx = vC cosϕ − b⋅ωCz ⋅sinϕ translation and either y - translation or v = v sinϕ + b⋅ω ⋅cosϕ z-rotation. Cy C Cz ϕ& = ω Cz or in matrix form: q& = J (q)⋅v ...... (2)

44 ⎛vCx ⎞ ⎛cosϕ −b⋅sinϕ ⎞ ⎛τ1 ⎞ ⎜ ⎟ ⎜ ⎟ ⎛ vC ⎞ τ = ⎜ ⎟ - torque on wheels 1 and 2. ⎜vCy ⎟ = ⎜ sinϕ b⋅cosϕ ⎟⋅⎜ ⎟ ⎝τ 2 ⎠ ⎜ ⎟ ⎜ ⎟ ⎝ωCz ⎠ ⎝ ϕ& ⎠ ⎝ 0 1 ⎠ In our case E p = 0 because the trajectory of

⎛ xC ⎞ ⎛vCx ⎞ the mobile robot base is constrain to the ⎜ ⎟ ⎜ ⎟ ⎛ vC ⎞ q = ⎜ yC ⎟; q& = ⎜vCy ⎟ , v = ⎜ ⎟ and horizontal plane. That means the system ⎜ ⎟ ⎜ ⎟ ⎝ωCz ⎠ can’t change its vertical position. ⎝ ϕ ⎠ ⎝ ϕ& ⎠ After the calculation of kinetic energy, the ⎛cos ϕ − b⋅sin ϕ⎞ ⎜ ⎟ dynamical equations of the mobile base in J(q) = ⎜ sin ϕ b ⋅cos ϕ ⎟ . Fig. 1 can be expressed in the matrix form: ⎜ 0 1 ⎟ ⎝ ⎠ D(q)⋅q&&+ H (q,q&)⋅q& = B(q)⋅τ + A(q)⋅λ ...... (4) where: where: v , v - are translation velocities of the Cx Cy ⎡ M + mw 0 mw ⋅b⋅sin ϕ ⎤ ⎢ ⎥ robot body, D(q) = ⎢ 0 M + mw − mw ⋅b⋅cosϕ⎥ , ω Cz - is the robot z-rotational velocity, ⎣⎢mw ⋅b⋅sin ϕ − mw ⋅b⋅cosϕ I ⎦⎥

ω w1 , ω w2 - are wheel rotational velocities, ⎡0 0 mw ⋅b ⋅ϕ& ⋅cos ϕ⎤ ⎢ ⎥ r - is actuated wheel radius, H (q, q&) = ⎢0 0 mw ⋅b ⋅ϕ& ⋅sin ϕ⎥ , a, b - are distances of wheels from ⎣⎢0 0 0 ⎦⎥ robot axes and ⎡cos ϕ cos ϕ⎤ 1 J (q) - is Jacobian matrix. B(q) = ⎢sin ϕ sin ϕ⎥ , r ⎢ ⎥ Constants a, b and r should be set according ⎣⎢ a − a ⎦⎥ to the robot proportions. Implicit values are: ⎡−sinϕ⎤ a =20 cm, b = 5 cm and r =10 cm. ⎢ ⎥ C A(q) = ⎢ cosϕ ⎥ and ⎣⎢ −b ⎦⎥ trajectory a R λ = −mw (vCx cos ϕ + vCy sin ϕ)⋅ϕ& , where: b C’ M – mass of mobile robot platform, mw – mass of wheel, all three wheels has same masse.

2 2 I – moment of inertia for mobile robot. Rmin = a +b = 20.616 [cm] The mass and inertia of the wheels are Figure 2: Minimal trajectory radius during considered explicitly. turning left or right. INVERSE DYNAMIC DYNAMIC MODELING As we know this consists on that for known The Lagrange formulations are used to inputs (torques on wheels), find outputs derive the dynamic equations of the mobile (position and velocity of mobile robot). robot. Based on equation (2), equation (4) become: d ⎛ ∂L ⎞ ∂L D(q)⋅(J&(q) ⋅v + J (q)⋅v&) + H (q, q&) ⋅ J (q) ⋅v = .... (5) ⎜ ⎟ − =τ ...... (3) = B(q) ⋅ τ + A(q)⋅λ dt ⎝ ∂q& ⎠ ∂q where: where: J&(q) - derivate of Jacobian. ⎛ xC ⎞ ⎜ ⎟ Based on equation (5), is needed to find v& : q = ⎜ yC ⎟ - generalized coordinates, [D(q) ⋅ J (q)] ⋅ v& = −D(q) ⋅ J&(q) ⋅ v − H (q, q&) ⋅ ⎜ ϕ ⎟ (6) ⎝ ⎠ ⋅ J (q) ⋅ v + B(q) ⋅ τ + A(q) ⋅ λ L = Ek + E p - Lagrange function,

45 As we know, mobile robot is redundant and The proposed control scheme for the mobile have positions of singularities, equation (6) robot trajectory tracking is presented in Fig. become: 3. T −1 T Computed torque input τ(t) is defined as v& = {[D(q)⋅ J(q)] ⋅[D(q)⋅ J(q)]}[D(q)⋅ J (q)] follow. [−D(q)⋅ J&(q)⋅v − H (q, q)⋅ J (q)⋅v + (7) & Output from mobile robot is his velocity: + B(q)⋅ τ + A(q)⋅λ] v = [v ω]T . Based on equation (2) we have:

CONTROL DESIGN q& = J ⋅v and with integration is founded q. Trajectory tracking error is: a. Conventional PD Controller e = Te (qd − q) , where ⎡ cos ϕ sin ϕ 0⎤ ⎢ ⎥ The trajectory tracking problem for mobile Te = ⎢− sin ϕ cos ϕ 0⎥ , and the auxiliary robot is passed as follows. Given is a desired ⎣⎢ 0 0 1⎦⎥ trajectory: velocity control input achieves tracking is xd , yd , ϕd respectively velocities: given by: x , y , ϕ . &d & d & d ⎡ vd cos e1 + k1e1 ⎤ vc = ⎢ ⎥ ...... (9) Based on equation (2): ⎣ωd + k2vd e2 + k3vd sin e3 ⎦ x&d = vd cos ϕd , y& d = vd sin ϕd , ϕ& d = ωd . Velocity tracking error is defined as: or ...... (8) ec = vc − v . T T qd = [xd yd ϕd ] , vd = [vd ωd ] .

qd yd (xd) y (x) d Fuzzy ϕ vd Logic Controller

T vc PD MOBILE v e ROBOT J

∫q&

Figure 3. Fuzzy Logic trajectory tracking control scheme.

This error now is input to conventional PD b. Fuzzy Algorithm controller and output from that is: u(t) = [K p Kv ]⋅ec ...... (10) The fuzzy controller in our approach is u(t) represent torque on wheels generated by based on the fuzzy control principles. Each conventional PD controller. input space is partitioned by fuzzy sets as Conventional PD controller is used both as shown in Figure 3. To obtain the desired an ordinary feedback controller to guaranty response during the tracking-error asymptotic stability during motion period, convergence movement by compensating for and as a reference model for responses of the nonlinear model dynamics, a Fuzzy Logic controlled mobile robot. Controller is designed to become a nonlinear controller.

46 Inputs for FLC are: Where for output 1 – additional torque to - trajectory tracking-error or distance error: wheel 1: d = yd (xd ) − y(x) T1-Z – additional torque is zero, T1-S – additional torque is small, T1-M – additional torque is large, T1-VL – additional torque is very large.

Figure 4.1. Input variable 1 – Distance Error (d) Where for input 1- distance error: NVL – distance is negative very large, Figure 5.2. Output 2 – additional torque for NL – distance is negative large, wheel 2. NM – distance is negative medium, Where for output 2 – additional torque to Z – distance is zero, wheel 2: PM – distance is positive medium, T2-Z – additional torque is zero, PL – distance is positive large, T2-S – additional torque is small, PVL – distance is positive very large, T2-M – additional torque is large, Angle error: T2-VL – additional torque is very large. e = ϕ − ϕ . Here, asymmetrical triangular function ϕ d which allow a fast computation, essential under real-time conditions, are utilized to describe each fuzzy set. The fuzzy sets of the two inputs dj, and ϕj is calculated thru 27 rules and the final output of the unit is given by a weighted average over 27 rules.

Tabele 1. Rules proposed by Authors Figure 4.2. Input variable 2– Distance Angle Error Angle Error (eϕ) Error a-NL a-NM a-Z a-PM a-PL Where for input 2 – angle error: NVL T1-Z T1-Z T1-Z - - a-NL – angle error is negative large, T2-VL T2-VL T2-VL a-NM – angle error is negative medium, NL T1-Z T1-Z T1-Z - - a-Z – angle error is zero, T2-L T2-L T2-L a-PM – angle error is positive medium, NM T1-Z T1-Z T1-Z - - a-PL – angle error is positive large. T2-M T2-M T2-M NS T1-Z T1-Z T1-Z - - T2-M T2-S T2-S Outputs from FLC is additional torque on Z T1-Z T1-Z T1-Z - - wheels Fig. 5. T2-M T2-S T2-Z PS - - T1-S T1-S T1-S T2-Z T2-Z T2-Z PM - - T1-M T1-M T1-M T2-Z T2-Z T2-Z PL - - T1-L T1-L T1-L T2-Z T2-Z T2-Z PVL - - T1-VL T1-VL T1-VL T2-Z T2-Z T2-Z Figure 5.1. Output 1 – additional torque for wheel 1.

47 SIMULATIONS 10-2 10-3 1.8 0.4

1.6 0.35

1.4

0.3 1.2

1 0.25

] ] d d 0.8 a a r r 0.2 e [ e [ gl gl

n 0.6 n

A A

0.4 0.15

0.2 0.1

0

0 20 40 60 80 100 120 140 160 180 200 0 20 40 60 80 100 120 140 160 180 200 Time (Seconds) Time (Seconds) Figure 6.1. Angle Error with PD without FLC Figure 6.2. Angle Error with PD and FLC

0.5 0.8

0 0.7

-0.5

0.6

-1

] ] m m c c [ [ 0.5 ce ce

n -1.5 n a a st st i i D D

-2 0.4

-2.5 0.3

-3 0 20 40 60 80 100 120 140 160 180 200 0 20 40 60 80 100 120 140 160 180 200 Time (Seconds) Time (Seconds) Figure 7.1. Distance Error with PD without FLC Figure 7.2. Distance Error with PD and FLC

Figure 8 shows using FLC and PD designed for path-planning, where mobile robot have e sensor which is giving information’s about distance and angle of “obstacles-green lines”.

Figure 8. Simulated Trajectory of mobile robot for detecting a Area

48 CONCLUSIONS AND FUTURE WORK FLC have decrease error. The simulation results carried out showed the effectiveness This paper has compared the design of the proposed control strategy. procedure and performance of a PD The disadvantage of having a Fuzzy Logic controller and a Fuzzy Logic Controller used Controller is number of rules and for trajectory tracking a constraint robot combinations to have effective trajectory system. The paper has achieved the tracking. How would the controller react to objective of constructing a simple controller the additional systems and trajectories? that mimics the handling a task that requires This and other points such as robustness will coordination of the limbs from a human be considered in future work. motor control perspective. The rules for the

REFERENCE

[1] David Kortenkamp, R.Peter Bonasso, and Robin Murphy, Artificial Intelligence and Mobile Robot, AAAI Press/The MIT Press. [2] Dei-Jeung Huh, Woo-Young Lee, Uk-Youl Huh, "A Study on Intelligence Navigation for Autonomous Mobile RobotUsing Fuzzy Logic Control" ICCAS 2001. [3] Young Ki Kim, “A Method of Obstacle Avoidance for a Mobile Robot using Fuzzy Reasoning", Inha University, 2002. [4] Using MATLAB Graphics. The MathWorks, Natick, MA, USA, 2000. [5] Getting Started with MATLAB. The MathWorks, Natick, MA, USA, 2000.

49 UTILISATION OF BURMESTER THEORY IN DESING OF PLANAR MECHANISM/MANIPUTALOR

I. Gojani, A. Pajaziti, B. Pira Mechanical Engineering Faculty, University of Prishtina, Kosova P. Marango Politechnical University of Tirana, Albania

Abstract: A lot has been done so far in the use of Burmester Notice that the vectors defined above form a closed loop, curves in designing four-bar mechanisms that prescribe three including the first and j-th positions: and four positions but not so much for five positions and even iβj iαj more in design of six-bar mechanisms prescribing four and five We +Ze -δj-Z-W=0 (1) positions. This paper briefly describes a way to utilize Burmester curves in design of planar mechanism/manipulator For four positions the above formula becomes: prescribing five precision points as a part of a mobile robot. In this paper we would like to present an application of Burmester iβ2 iα2 W(e -1)+Z(e -1)=δ2 Theory in practice in design planar mechanisms/manipulator iβ3 iα3 that prescribes five positions using. W(e -1)+Z(e -1)=δ3 (2)

iβ4 iα4 1. INTRODUCTION W(e -1)+Z(e -1)=δ4

After the compatible sets of β2 are obtained, then any method Looking at the world around, approximately every few seconds for solving simultaneous complex equations can be used to find one encounters a mechanism of some kind without being aware Z and W. Then the circle point k and the center point m are of its presence. Although the wheel is one of the greatest given by the following expressions (figure 1). inventions of the man, manipulating with it was just as well challenging. For any object that’s needs moving from one k = R - Z position to another a mechanism of some sort is required. In 1 m = k – W other words, mechanisms ore nothing more than a matter combined with a simple geometry. Although mechanisms can The above pair of these two points, is called Burmester Point be tree dimensional, most of them are limited in two Pair (B.P.P.), and the locus of these pairs for different values dimensions. of β2 form the Burmester Curves The objectives of this paper are to utilize the Burmester Theory in a design of a four-bar mechanism that will prescribe five 3. APPLICATION OF BURMESTER THEORY precision points as well as keep the desired angle of the element moved by the mechanism. An example where the Burmester Theory utilized by Chan Y.M. in 1994 with a Burmester Theory is utilized will be presented with this paper. specially written computer program [2]. It calculated circle- point and centre-point circles (specific for Burmester Theory) 2. BURMESTER THEORY for four positions, figure 2 2. This program was later modified initially by Noussas D. and finally by the coauthor of this When a body moves through two and three prescribed positions paper, Pira B., for mechanisms that prescribe five precision there are three respectively two infinities of solutions in the points [3, 4]. design of the mechanism; however, when the body passes Using the Burmester curves for four prescribed positions, a set through four and five prescribed positions then there is one- of centre-point and circle-point circles, show the solution, infinity respectively few possible designs of a mechanism. figure 2. For five positions, if a solution exists, the solutions Burmester found the solutions to problem of four prescribed lies by superimposing two sets of four positions, position 1-2-3- positions of the plane using complex numbers [1], figure 1. 4 and 1-2-3-5. These two solutions can yield the solution to the

5-position problem by finding the intersections of the centre and circle curves, figure 3.

Centre points Circle points

Figure 2: Burmester Curves for 4 positions

Figure 1: The unknown dyad’s W,Z which can guide the moving plane π from the first to the j-th position 50

Centre points Circle points Centre points Circle points

Figure 6: Mechanism/manipulator in its first position Figure 3: Burmester Curves for 5 positions To make sure that the above mechanism prescribes all five 4. CASE STUDY positions, DeMech computer software was used to animate its movements. Figure 7 shows the mechanism in its five Figure 4 shows a small bucket/gripper containing various positions. explosives (inactive or even active) that is to be dumped into a large container. There is a need to synthesize a four-bar mechanism, as a part of mobile robot, which will lift the bucket/griper and dump its content into the container. Ground points of the mechanism are attached to the container. The container has to go through five prescribed precision points at a specific angle for every position.

4

3 5

2

1 x Figure 7: The mechanism/manipulator in its 5 prescribed Figure 4: Case study positions

Once the computer program is executed, two sets of curves are 5. CONCLUSION drawn. At the intersection of two centre-point curves (two intersection) two ground points of the mechanism are identified There have been several approached to design four-bar while at the intersection of the of two circle-point curves two mechanisms that prescribe five precision points using moving points of the mechanism are also identified, figure 5. Burmester Theory. This method shows how useful and how easy is to design planar mechanisms using Burmester curves.

Although it looks as if it is very strait forward to design four- Moving bar mechanisms using Burmester curves, six-bar mechanism Points seem to be a different story all together. Regardless of that, we Centre points are going to try to use Burmester curves to design six-bas Ground Circle points mechanisms in the future. Centre points Points Circle points First Position 6. REFERENCES Second Position Third Position Fourth Position [1] Sandor G. N. & Erdman A. G. “Advanced Fifth Position Mechanism Design: Analysis and Synthesis Volume 2”, PRENTICE HALL 1984, pp263-299. [2] Chan Y. M. “Some Case Studies in Mechanism Design MSc thesis” Imperial College Mechanical Engineering Department 1994. Figure 5: Ground points and moving points [3] Noussas D. “Case Studies in Mechanism Design MSc thesis” Imperial College Mechanical Engineering By connecting these points together a four-bar mechanism is Department 1996, designed that will prescribe all five precision points, figure 6, [4] Pira B. “The use of Burmester Curves in Design of Planar Mechanisms” Imperial College Mechanical with A and B as ground points and A1 and B1 as moving points. Engineering Department 1999.

B

51

B1 A