RoboCup 2000 Publications

From RoboCup Wiki
Jump to: navigation, search
Author Title Year Journal/Proceedings Reftype DOI/URL
Abbaspour, A., Rahmani, M., Radjabalipour, B. & Nazemi, E.

SBC++ Simulator Team


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 397-400 inproceedings

DOI URL

Abstract: In this article the SBC++ simulator team and its overall design and scientific goals are described. Our main goal in this project is to achieve new methods of machine learning and deciding; also multi-agent behaviors such as strategy setting and cooperative learning with insistence on behavior networks. To achieve this, SBC++ uses MASM[1] model, designed by Maes, as its fundamental base. Our team uses the CMUnited-99[4] low level code with little modification. Major modification and improvements are in the high level code.
Adorni, G., Bonarini, A., Clemente, G., Nardi, D., Pagello, E. & Piaggio, M.

ART'00 - Azzurra Robot Team for the Year 2000


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 559-562 inproceedings

DOI URL

Abstract: Robotic soccer is a challenging research domain that can be used to explore new problems and test new techniques/solutions in the fields of Artificial Intelligence and Autonomous Robotics, as well as for training and educational purposes.
Adorni, G., Cagnoni, S., Mordonini, M. & Piaggio, M.

Team/goal-Keeper Coordination in the RoboCup Mid-Size League


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 279-284 inproceedings

DOI URL

Abstract: In this paper we describe the coordination strategies that were designed to achieve effective cooperation between a robot goal-keeper and the rest of the ART (Azzurra Robot Team) team that participates in the RoboCup F-2000 (mid-size) competitions.
The paper introduces the multi-robot environment on which cooperation in ART is based and, in particular, its communication sub-system. Some case studies and the results of their application in the ART team are then described.
Akiyama, H.

A Team


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 401-404 inproceedings

DOI URL

Abstract: Soccer has the problem of the extremely high complexity. This complexity can be considered a problem of the search space. When the player decides the action from the situation of the world, quite many choices exist and it is impossible to search for all the choices. This search space problem is essence of the problem in soccer.
Au, G., Goss, S., Heinze, C. & Pearce, A.R.

RescueModel: A Multi-Agent Simulation of Bushfire Disaster Management


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 285-290 inproceedings

DOI URL

Abstract: The RescueModel project is a vehicle for research into multiagent systems, architectures, and strategies. It builds on the theoretical, practical, and experimental base of a decade of beliefs-desires-intentions (BDI) agent systems development. This paper describes a project that will bring together the environmental richness found usually in large scale military operations research simulations with the architectural richness of agent models often researched in universities. Proposed applications of RescueModel include search and rescue and disaster response studies.
Baltes, J.

All Botz


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 515-518 inproceedings

DOI URL

Abstract: There are four features, which make the All Botz a very unique RoboCup team
– -There are four features, which make the All Botz a very unique RoboCup team:
– -The “robots” are cheap toy cars.
– -The “robots” are non-holonomic (car-like).
– -The robots do not use coloured markers, manual tagging, or bar codes.
– -The videoserver can view the playing field from any angle instead of only from directly overhead.
Baltes, J.

4 Stooges


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 519-522 inproceedings

DOI URL

Abstract: The 4 Stooges are a small sized RoboCup team, which grew out of the All Botz who competed at RoboCup-99.
One of the main differences is that the 4 Stooges are a team of fully autonomous robots. All sensors, actuators, and power supply (CMOS camera, and battery) as well as all processing is housed on the robot.
Baltes, J. & Hildreth, N.

Adaptive Path Planner for Highly Dynamic Environments


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 76-85 inproceedings

DOI URL

Abstract: This paper describes adaptive path planning, a novel approach to path planning for car-like mobile robots. Instead of creating a new plan from scratch, whenever changes in the environment invalidate the current plan, the adaptive path planner attempts to adapt the old plan to the new situation. The paper proposes an efficient representation for path that is easily amendable to adaptation. Associated with the path planner is a set of repair strategies. These repair strategies are local methods to fix a plan to compensate for object movement in the domain. The repair strategies are specific and have a high probability of being able to fix a plan. An empirical evaluation shows that adaptive path planning is suitable to highly dynamic domains, such as RoboCup. Adaptive path planning reduces the cumulative planning time by a factor of 2:7 compared to Bicchi’s planner. At the same time, the quality of the plans generated by the adaptive path planner were similar to those generated by Bicchi’s planner.
Bredenfeld, A., Christaller, T., Guenther, H., Hermes, J., Indiveri, G., Jaeger, H., Kobialka, H.-U., Plöger, P.-G., Schoell, P. & Siegberg, A.

GMD-Robots


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 579-582 inproceedings

DOI URL

Abstract: The overall research goal of GMD’s RoboCup team is to increase both, (1) the speed of mobile robots acting as team in a dynamic environment and (2) the speed of design for behavior-based robot control. Therefore, we started in 1998 to develop a proprietary fast robot platform and the integrated Dual Dynamics Design Environment.
Brusey, J., Makies, M., Padgham, L., Woodvine, B. & Fantone, K.

RMIT United


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 563-566 inproceedings

DOI URL

Abstract: The RMIT United team is entirely based on custom-made robots of two types, one goalie robot and three field player robots. The field players are distinguished by a powerful kicking device and are largely the same as our previous year’s entry. The goalie robot is designed to move from side to side, rather than forward and back, and has tank tracks rather than wheels. The robots mainly use vision for perception, and have a dedicated digital signal processor (DSP) to interpret what it sees. The goalie robot has two cameras and two DSPs.
Bräunl, T., Reinholdtsen, P. & Humble, S.

CIIPS Glory Small Soccer Robots with Local Image Processing


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 523-526 inproceedings

DOI URL

Abstract: CIIPS Glory is using a local intelligence approach to solve the task of robot soccer. In this respect our system is much closer to the F-2000 league (mid-size) than other approaches in the F180 (small-size) league. We are using EyeBot controllers, the RoBIOS operating system and the EyeCam digital onboard camera systems (CMOS). Our robot family also comprises 6-legged and biped walking machines, an omni-directional robot and an autonomous airplane.
Buck, S., Hanek, R. & Klupsch, M.

Agilo RoboCuppers: RoboCup Team Description


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 567-570 inproceedings

DOI URL

Abstract: This paper describes the Agilo RoboCuppers team of the image understanding group (FG BV) at the Technische Universität München. With a team of four Pioneer 1 robots, equipped with CCD camera and a single board computer each and coordinated by a master PC outside the field we participate in the Middle Size League of the fourth international RoboCup Tournament in Melbourne 2000. We use a multi-agent based approach to represent different robots and to encapsulate concurrent tasks within the robots. A fast feature extraction based on the image processing library HALCON provides the data necessary for the on-board scene interpretation. All robot observations are fused to one single consistent view. Decision making is done on this fused data.
Burkhard, H.-D., Bach, J., Schröter, K., Wendler, J., Gollin, M., Meinert, T. & Sander, G.

AT Humboldt 2000 (Team Description)


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 405-408 inproceedings

DOI URL

Abstract: Our agent team AT Humboldt 2000 is partly an extension of our former team AT Humboldt 99[2,3]. Again we used a BDI architecture. Especially the world model and some skills where revised. A new timing concept and a completely different architecture for the deliberation component were developed. The actual development was subject of an undergraduate course. Because of problems with the integration of components developed by different work groups, we were forced to start in RoboCup 2000 with a mixed team, consisting mainly of an extended version of the players used in EuRoboCup 2000. Only the goalie used all our new concepts.
This work has been partially supported by the German Research Society, Berlin- Brandenburg Graduate School in Distributed Information Systems (DFG grant no. GRK 316).
Butler, M., Prokopenko, M. & Howard, T.

Flexible Synchronisation Within RoboCup Environment: a Comparative Analysis


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 119-128 inproceedings

DOI URL

Abstract: Synchronisation between an agent and the environment it resides in, is without a doubt, an important aspect of a more generic problem of agent interaction with the environment. A systematic comparative analysis of alternative approaches to the synchronisation problem remains an open challenge, despite numerous successful implementations of RoboCup teams in the past. The underlying reasons appear to be a multiplicity of software platforms, implementation changes in the Simulator itself, and sometimes a methodological bias of designers driven by a particular agent architecture. In this paper we describe alternative methods of agent-environment synchronisation, introduce a simple software tool for analysing RoboCup games via server log files, and compare the proposed synchronisation alternatives with respect to certain quantitative metrics. This comparative analysis is conducted without varying situated, tactical or strategic agent skills, highlighting purely synchronisation features.
Candea, C. & Staicu, M.

AIRG Sibiu


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 409-412 inproceedings

DOI URL

Abstract: In this year the AIRG group prepares a new team for the RoboCup competitions based on new ideas and concepts. Based on the MAS(Multi Agent Systems) architecture implemented by us in 1999 for the Sibiu Team that took part at RoboCup 1999 competition hold in Stockholm, the paper describes a generic architecture for a RoboCup team with four holon types situated on three levels. This architecure tries to synergistically combine the characteristics of both domains: MAS and HMS(Holonic Manufacturing Systems).
Castelpietra, C., Iocchi, L., Nardi, D., Piaggio, M., Scalzo, A. & Sgorbissa, A.

Communication and Coordination Among Heterogeneous Mid-Size Players: ART99


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 86-95 inproceedings

DOI URL

Abstract: Distributed coordination among robotic soccer agents has been considered in the recent years within the framework offered by the RoboCup competitions, mostly in the simulation and F-180 leagues. In this paper we describe the methods and the results achieved in coordinating the players of the ART team participating in the F-2000 league. The team is formed by several heterogeneous robots having different mechanics, different sensors, different control software, and, in general, different abilities for playing soccer. The coordination framework we have developed has been successfully applied during the 1999 official competitions allowing both for a significant improvement of the overall team performance and for a complete interchangeability of all the robots.
Acknowledgments We are grateful to the members of the ART team and acknowledge the support of “Consorzio Padova Ricerche”, “Consiglio Nazionale delle Ricerche”, Universitàdi Roma “La Sapienza”, Politecnico di Milano, Università di Padova, Università di Genova, Università di Palermo, Università di Parma, AI IA, Vesta pneumatics,Sony Italia, Images, Tekno.
Chang, M.M., Browning1, B. & Wyeth, G.F.

ViperRoos 2000


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 527-530 inproceedings

DOI URL

Abstract: The ViperRoos are a team of soccer playing robots that made their debut in the F-180 league at RoboCup-2000. Each Viper robot is completely autonomous and relies on on-board vision, rather than an overhead camera, as its primary means of perceiving the world. In addition to participating in robot soccer competitions, the Viper robots have already been used successfully for other research investigations (for example [1]). The ViperRoos represent an innovative step towards cheap, autonomous robot hardware for use in both RoboCup competitions and general mobile robotics research.
Dahlström, A., Heintz, F., Jacobsson, M., Thapper, J. & Überg, M.

The NOAI Team Description


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 413-416 inproceedings

DOI URL

Abstract: The NOAI team is built on the RoboSoc framework, which is a system for developing RoboCup agents for educational use. The decision making is behavior based and is a continuation of the work of Paul Scerri et al. on the Headless Chickens III. We have extended their decision trees to directed acyclic decision graphs and made it possible to use full predicate expressions instead of just predicates. We have also added a language for describing the decision graphs. This is to be able to use their, highly successful, strategy editor to graphically design the team strategy.
Demura, K., Tachi, N., Kubo, N. & Miwa, K.

WinKIT


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 571-574 inproceedings

DOI URL

Abstract: This paper describes our research interests and technical information of our team for RoboCup-2000. Our robots have been developed to have a capability of pass-based tactics. That is, the capability of position estimations of the robots and the ball, the distinction between our team and opponent team, and passing the ball to a desired direction. To achieve the capability, robots have kicking devices and omnidirectional vision systems.
Dorer, K.

Improved Agents of the Magmafreiburg2000 Team


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 417-420 inproceedings

DOI URL

Abstract: The magmaFreiburg2000 team, as its predecessor, is based on Extended Behavior Networks [4]. This paper focuses on the improvements made compared to the magmaFreiburg1999 team of Stockholm. The main domain-independent improvements have been made by introducing concurrent behavior selection into the Extended Behavior Networks framework [3]. Domain-dependent improvements have mainly been made on server synchronization and game strategy.
Again the performance of the magmaFreiburg2000 has been very satisfying finishing the competition at an equal fifth place. In addition to the good performance of our team it should be mentioned that we also had a big portion of luck when proceeding to the next round by the first coin toss decision of the RoboCup history (sorry again to Sharif Arvand).
Douret, J., Dorval, T., Benosman, R., Bras, F., Surtet, G., Petit, T., Quedec, N., Philip, D., Cordurié, G., Rebello, M., Abraham, D., Couder, N. & Maroon, M.

RoboCup 2000 (F180) Team Description : UPMC-CFA Team (France)


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 531-534 inproceedings

DOI URL

Abstract: This paper describes the team UPMC-CFA that participated to the EuroCup2000 and RoboCup’2000 F180-league competition. We will start by presenting the mechanical and electrical design of our robots. We will then introduce the vision program explaining the colometric and geometric calibrations it needs. Finally we will explain the implemented behaviour and the way we controled our robots.
Drücker, C., Hübner, S., Schmidt, E., Visser, U. & Weland, H.-G.

Virtual Werder


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 421-424 inproceedings

DOI URL

Abstract: Virtual Werder is a new team in the simulation league. The team is the result of an initiative which has been formed from lectures on Artificial Intelligence at the Department of Mathematics and Computer Science at the University of Bremen.
The main focus of this team is to use the online coach for the detection of the opponents play system. The motivation and significance for this has been discovered during a structured interview with Thomas Schaaf, the manager of SV Werder Bremen.
D’Andrea, R., Kalmár-Nagy, T., Ganguly, P. & Babish, M.

The Cornell Robocup Team


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 41-51 inproceedings

DOI URL

Abstract: This paper describes the Cornell Robocup Team, which won the RoboCup-2000 F180 championship in Melbourne, Australia. The success of the team was due to electro-mechanical innovations (omnidirectional drive and a dribbling mechanism) and the control strategies that rendered them effective. As opposed to last years “role-based” strategy, a “play- based” strategy wasimplemented, which allowed us to make full use of the robot capabilities for cooperative control.
Acknowledgements The lead author would like to thank all of the students involved in the RoboCup project at Cornell University, past and present, for their dedication to the project, and for their numerous accomplishments. We would also like to thank our many sponsors for their continuing support, in particular Jim Morgan at Applied Materials, John Swanson at ANSYS, Lockheed Martin, and United Technologies.
Emery, R., Balch, T., Shern, R., Sikorski, K. & Stroupe, A.

CMU Hammerheads Team Description


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 575-578 inproceedings

DOI URL

Abstract: In this paper the design of the CMU Hammerhead middle-size robot soccer team is presented. The team consisted of 4 fully autonomous robots with wireless communication and color vision. The robots’ color segmentation algorithm, CMVision, provided reliable, real-time ball tracking and was tolerant to changes in lighting conditions. The robots’ software architecture was implemented using TeamBots, a Java-based environment for behavior based robot control, and Clay, a TeamBots module for creating motor-schema based control systems. The use of behavior-based control allowed to the robots to react quickly to the dynamic environment despite non-holonomic platforms. The team included some team- work by sharing information about the ball’s location and the states of other team members. The CMU Hammerheads made it to the quarter finals of the competition.
Enderle, S., Ritter, M., Fox, D., Sablatnög, S., Kraetzschmar, G. & Palm, G.

Vision-Based Localization in RoboCup Environments


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 291-296 inproceedings

DOI URL

Abstract: Knowing its position in an environment is an essential capability for any useful mobile robot. Monte-Carlo Localization (MCL) has become a popular framework for solving the self-localization problem in mobile robots. The known methods exploit sensor data obtained from laser range finders or sonar rings to estimate robot positions and are quite reliable and robust against noise. An open question is whether comparable localization performance can be achieved using only camera images, especially if the camera images are used both for localization and object recognition. In this paper, we discuss the problems arising from these characteristics and show experimentally that MCL nevertheless works very well under these conditions.
Enokida, S., Ohasi, T., Yoshida, T. & Ejima, T.

Extended Q-Learning: Reinforcement Learning Using Self-Organized State Space


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 129-138 inproceedings

DOI URL

Abstract: We propose Extended Q-learning. To accommodate continuous state space directly and to improve its generalization capability. Through EQ-learning, an action-value function is represented by the summation of weighted base functions, and an autonomous robot adjusts weights of base functions at learning stage. Other parameters (center coordinates, variance and so on) are adjusted at unification stage where two similar functions are unified to a simpler function.
Esaki, T., Sakushima, T., Futamase, S., Ito, N., Takahashi, T., Chen, W. & Wada, K.

Kakitsubata Team Description


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 425-428 inproceedings

DOI URL

Abstract: In a multi-agent system, it is important how an agent cooperate with the others. However, it is difficult for an agent to cooperate appropriately in a dynamic environment, such as the RoboCup soccer. Therefore, Our team has two main features which allow appropriate cooperative activity, a cooperative protocol and a coach-agent[1]. With these features, our team can cooperate with each other in such environment.
Ferraresso, M., Ferrari, C., Pagello, E., Polesel, R., Rosati, R., Speranzon, A. & Zanette, W.

Collaborative Emergent Actions Between Real Soccer Robots


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 297-302 inproceedings

DOI URL

Abstract: We discuss how to induce a set of collaborative emergent actions between two soccer robots. Cooperative abilities, like exchanging a ball, can be achieved through the use of efficient collision avoidance algorithms implemented on two players able to frequently swap their roles. These algorithms have been tested on Bart and Homer, designed at IAS Lab. of Padua Univ., that played quarter, semifinals, and finals with ART at RoboCup'99. The interaction with the ball was made easy by a directional kicker which allowed to hit the ball both frontally and laterally.
Ferrari, C., Garelli, F. & Pagello, E.

PaSo-Team 2000


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 429-432 inproceedings

DOI URL

Abstract: Following the experience done in previous competitions, it has been developed the 2000 version of PaSo-Team (The University of PAdua Simulated Robot SOccer Team), a reviewed release of Paso-Team99. During the RoboCup ’99 competition in Stockholm some teams suffered syncronization problems with the soccer server: these problems greatly influenced their performances and prevented them from playing successfully. While developing PaSo-Team 2000 the main efforts were dedicated to better understanding timing and syncronization techniques for real-time multi-agent systems. Following the interesting experience done by Kostiadis in developing the Essex Wizzard team [1] we redesigned the syncronization procedures using the multi-threading paradigm. Solving syncronization in a multi-threading environment gives important theoretical hints to approch the coordination for those multi-agent systems made by thousand of very simple concurrent interacting modules. During the Stockholm competition PaSo-Team99 suffered another major problem regarding the actions a player must take when the game is stopped (i.e. when the ball is outside or when a team is offside). For example if a player has to throw-in the ball, he must go outside the field, turn toward the field and eventually kick the ball, performing different actions even if the state of the game doesn’t change. As in a reactive architecture the current behaviour can change only when the game state changes, PaSo-Team99 introduced virtual states to ensure a change of behaviour. Instead in PaSo-Team 2000 we simplify the design of these actions introducing multi-step behaviours.
Frank, I., Kumiko, T.-I., Okuno, H.G., Nakagawa, Y., Nakadai, K., Kitano, H., Akita, J. & Maeda, K.

And the Fans Are Going Wild! SIG Plus MIKE


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 139-148 inproceedings

DOI URL

Abstract: We present an implemented commentary system for real- world robotic soccer. Our system uses an overhead camera to pass game information to the simulator league commentary program Mike, and uses the humanoid robot Sig to provide a physical embodiment for the commentary. The use of a physical robot allows us to direct audience attention to important events by looking at them, provide more realism to the interaction between Mike and other humans in the domain, and even set up a dialogue with the audience as part of the commentary itself. Our system combines the multi-modal input of audio and video information to generate a multi-modal commentary that brings the extra dimension of body language to the social interaction that characterises the overall commentary task.
Frank, I., Tanaka-Ishii, K., Arai, K. & Matsubara, H.

The Statistics Proxy Server


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 303-308 inproceedings

DOI URL

Abstract: We present a real-time statistical analysis tool for soccer. This system is designed to promote the advancement of RoboCup by facilitating fundamental research on issues such as learning and team evaluation and assessment. Analysis of a game is carried out by a central server, to which clients can connect to request data. We describe the operation of the system and give examples of its potential applications.
Habibi, J., Chiniforooshan, E., Khabbazian, M., Mirzazade, M., Safari, M. & Younesi, H.

Sharif-Arvand Simulation Team


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 433-436 inproceedings

DOI URL

Abstract: Like most teams SharifArvand consists of three parts: WorldModel, skills, and strategy. Duty of the WorldModel is getting received messages from the Soccer Sever (SServer[1]), creating a virtual model of the game field and updating the information about each object. Other layers of the program will take the neces sary information for playing soccer from theWorldModel in a convenient format. Also any command sent to the server should be reported to the WorldModel. Other parts of the player program are implemented in a multilevel manner. In other words, in the lowest level there are simple skills which only use the World Model for undertaking their duties, whereas more complex skills in higher levels use the WorldModel and the lower level skills to achieve their goal. Finally, a strategy uses these skills for playing a soccer game. So, a strategy is the highest level skill. Therefore, a strategy should create, control and destroy skills to play a successful soccer game.
Habibi, J., Foroughi, E., Motamed, M., Karimian, P., Hatami, H. & Fardad, H.

SharifII Soccer Simulation Team


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 437-440 inproceedings

DOI URL

Abstract: This paper describes ParsAI, a team developed and working at Sharif University of Technology. It is mainly concentrated on two distinct works, an idea for decision making components and a simple but powerful easy-to-debug implementation of clients.
Hanek, R., Schmitt, T., Klupsch, M. & Buck, S.

From Multiple Images to a Consistent View


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 169-178 inproceedings

DOI URL

Abstract: The approach presented in this paper allows a team of mobile robots to estimate cooperatively their poses, i.e. positions and orientations, and the poses of other observed objects from images. The images are obtained by calibrated color cameras mounted on the robots. Model knowledge of the robots’ environment, the geometry of observed objects, and the characteristics of the cameras are represented in curve functions which describe the relation between model curves in the image and the sought pose parameters. The pose parameters are estimated by minimizing the distance between model curves and actual image curves. Observations from possibly different view points obtained at different times are fused by a method similar to the extended Kalman filter. In contrast to the extended Kalman filter, which is based on a linear approximation of the measurement equations, we use an iterative optimization technique which takes non-linearities into account. The approach has been successfully used in robot soccer, where it reliably maintained a joint pose estimate for the players and the ball.
Heintzy, F., Kummenejez, J. & Scerriy, P.

Using Simulated RoboCup in Undergraduate Education


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 309-314 inproceedings

DOI URL

Abstract: We argue that RoboCup can be used to improve the teaching of AI in undergraduate education. We give some examples of how AI courses using RoboCup can be implemented using a problem based approach at two different Universities. To reduce the negative aspects found we present a solution, with the aim of easing the burden of grasping the domain of RoboCup for the students, RoboSoc which is a general framework for developing simulated RoboCup agents.
Hengst, B., Ibbotson, D., Pham, S.B., Dalgliesh, J., Lawther, M., Preston, P. & Sammut, C.

The UNSW RoboCup 2000 Sony Legged League Team


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 64-75 inproceedings

DOI URL

Abstract: We describe our technical approach in competing at the RoboCup 2000 Sony legged robot league. The UNSW team won both the challenge competition and all their soccer matches, emerging the outright winners for this league against eleven other international teams. The main advantage that the UNSW team had was speed. The robots not only moved quickly, due to a novel locomotion method, but they also were able to localise and decide on an appropriate action quickly and reliably. This report describes the individual software sub-systems and software architecture employed by the team.
Howard, A.

MuCows


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 535-538 inproceedings

DOI URL

Abstract: The University of Melbourne entered the RoboCup contest for the first time in 2000, fielding teams in both the F180 and Sony Legged Leagues. This paper describes certain aspects of our F180 team: the MuCows.
Hu, H., Gu, D. & Li, B.

Essex Rovers Team Description


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 611-614 inproceedings

DOI URL

Abstract: This article introduces our current research efforts to build a multi-agent system for cooperation and learning of multiple Sony legged robots in the RoboCup domain. A behaviour-based hierarchy is proposed for our Essex Rovers robot soccer team to achieve intelligent actions in real time, which includes both a neural network based color detection algorithm and a fuzzy logic controller.
Hu, H., Kostiadis, K., Hunter, M. & Kalyviotis, N.

Essex Wizards 2000 Team Description


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 441-444 inproceedings

DOI URL

Abstract: This article gives an overview of the Essex Wizards 2000 team participated in the RoboCup 2000 simulator league. A brief description of the agent architecture for the team is introduced. Both low-level behaviours and high-level behaviours are presented. The design issues regarding fixed planning and reinforcement learning are briefly outlined.
Hugel, V., Bonnin, P. & Blazevic, P.

French LRP Team's Description


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 615-618 inproceedings

DOI URL

Abstract: This paper describes the problems encountered in vision by the French team in RoboCup2000. Since the participation of LRP in 1998, the team has focused on the 3 following topics: locomotion, vision and strategy. In RoboCup 1999 in Stockholm, Sweden, we carefully designed the locomotion patterns [1][2][3] to make the robots walk as fast as possible. We also implemented some trajectory correction while walking. In RoboCup 2000 in Melbourne, Australia, our team tried to improve the vision system to see better and farther. We also focused on adding some special strategy features to improve the robots’ behavior.
Igarashi, H.

Path Planning of a Mobile Robot as a Discrete Optimization Problem and Adjustment of Weight Parameters in the Objective Function by Reinforcement Learning


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 315-320 inproceedings

DOI URL

Abstract: In a previous paper, we proposed a solution to path planning of a mobile robot. In our approach, we formulated the problem as a discrete optimization problem at each time step. To solve the optimization problem we, used an objective function consisting of a goal term, a smoothness term and a collision term. This paper presents a theoretical method using reinforcement learning for adjusting weght parameters in the objective functions. However, the conventional Q-learning method cannot be applied to a non-Markov decision process. Thus, we applied William’s learning algorithm, REINFORCE, to derive an updating rule for the weight parameters. This is a stochasic hill-climbing method to maximize a value function. We verified the updating rule by experiment.
Igarashi, H., Yamauchi, Y. & Iidoi, S.

RoboCup-2000 Simulation League: Team KU-Yam2000


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 445-448 inproceedings

DOI URL

Abstract: In this paper we described our KU-Yam2000 team that participated in the simulation league of RoboCup-2000 Melbourne. KU-Yam2000 is characterized by soccer agents that are controlled by a hierarchy of actions. Actions in each level of the hierarchy are separated into modules, which are arranged in a form of a C-language library to be called from a main program.
Iizuka, H., Hiramoto, M., Kawamura, H., Yamamoto, M. & Ohuchi, A.

Harmony Team


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 449-452 inproceedings

DOI URL

Abstract: In the simulation part of soccer server, there are many strong teams made by detailed designs of hand-coded program techniques based on top-down approaches. However, it is exciting if it can be possible to design strong soccer team by the computer program itself. It is difficult but interesting and challenging to let a computer program to design intelligent program by itself. The aim of us is to design the strong soccer agent team based on evolutionary technology and to overcome champion team by autonomous self-designed agents. As first step of such aim, we construct co-evolutionary soccer agent team with hand-coded primitive actions.
Itsuki, N.

Framework of Distributed Simulation System for Multi-Agent Environment


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 229-238 inproceedings

DOI URL

Abstract: Simulation systems are important tools for researches on Multi-agent systems. As a research tool, a simulation system should have features of: (1) light-weight and small requirement of computational resources, and (2) scalability to add new functions and features. I investigate experience of development of Soccer Server, the official soccer simulator used in RoboCup Simulation League, and propose a new framework for distributed simulation system for general purpose of multi-agent researches.
In the new framework, a simulation is divided into several modules and executed in parallel. These executions are combined by a kernel module via a computer network. Because of the modularity over networks, users easily maintain the development of simulation system. I also discuss about the relation to HLA, another framework for distributed military simulation system.
Jamzad, M., Foroughnassiraei, A., Aaghai, T.H., Mirrokni, V., Ghorbani, R., Noori, A.H., Kazemi, M., Chitsaz, H., Mobasser, F., Moghaddam, M.E., Gudarzi, M. & Ghaffarzadegan, N.

A Goal Keeper for Middle Size Robocup


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 583-586 inproceedings

DOI URL

Abstract: In real soccer goal keeper has a completely different behavior that the players. Thus we designed our goal keeper with some basic ideas taken from real soccer, which is moving mainly in goal area and using arms to take the balls. Our goal keeper has a moving mechanism based on 8 motors which enables it to move forward/backward, straight left/right and rotate around its geometrical center, and a sliding arm which moves toward the direction of ball faster than the robot body. Using 3 CCD cameras in front and rear left and right provides the goal keeper with a view of about 210 degrees.
Jonker, P., Caarls, J. & Bokhove, W.

Fast and Accurate Robot Vision for Vision Based Motion


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 149-158 inproceedings

DOI URL

Abstract: This paper describes the vision module from the soccer playing robots of the Dutch Team. Fast vision is necessary to get a close coupling with the motion software in order to allow fast turning and dribbling with the ball without loosing it. Accurate vision is necessary for the determination of the robot’s position in the field and the accurate estimation of the ball position. Both fast and accurate are necessary for the goalkeeper, but also when one robot passes the ball to another. While the Dutch team has pneumatic kicking devices that allows catching a ball smoothly, fast an accurate vision is mandatory. We use lens undistortion, a new color segmentation scheme and a shape classification scheme based on linear and circular Hough transforms in regions of Interest. We use a severe calibration procedure to get very good distance and angle measurements of the known objects in the field of view of the robot. For the keeper robot we use a Linear Processor Array in SIMD mode, that is able to execute the entire robust vision algorithm within 30ms. However the same software was programmed for the other robots with a WinTV framegrabber on the on-board Pentium of the robot. With optimizing for speed we also remained within 25ms, however, omitting the circular Hough transform for the ball and processing in a separate thread the Linear Hough transforms for self-localization on lower rate of about 50msec. The angular errors at 0 °, 20 ° and 30° heading are about 0.6 °, 0.5° and 0.4° up to 4,5 meters. The distance error at 0° heading is 5% up to 3 meters.
Jonker, P., van Geest, W. & Groen, F.

The Dutch Team


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 587-590 inproceedings

DOI URL

Abstract: This paper describes the layered and modular hard- and software architecture of autonomously soccer playing robots used in The Dutch Team. The schemes of the hard and software architectures are presented and the software functional modules are described with their relations and interfaces. With this software architecture we attempted to merge the distributed computing aspect of real-world intelligent autonomous robots, with the client-server approach of the soccer simulator.
Kaneda, T., Matsuno, F., Takahashi, H., Matsui, T., Atsumi, M., Hatayama, M., Tayama, K., Chiba, R. & Takeuchi, K.

Simulator Complex for Robocup Rescue Simulation Project - As Test-Bed for Multi-Agent Organizational Behavior in Emergency Case of Large-Scale Disaster


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 321-326 inproceedings

DOI URL

Abstract: In the RoboCup Rescue Simulation Project, several kinds of simulator such as Building-Collapse and Road-Blockage Simulator, Fire Spread Simulator and Traffic Flow Simulator are expected to provide a complicated situation in the case of the large-scale disaster through their synergistic effects. It is called Simulator Complex. This article addresses, first, system components of the prototype version of this Simulator Complex, then, explains each of the simulators and the Space-Time GIS(Geographical Information System) as DBMS(DataBase Management System). In the demonstrations, we have shown the performance enough for a test-bed for multi-agent system development.
Khessal, N.O.

Role-Based Strategy in TPOTs RoboCup Team


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 539-542 inproceedings

DOI URL

Abstract: This paper describes a role-based strategy implemented in TPOTs RoboCup team. A detailed work on the physical robots as well as the overall architecture can be found in [1,2]. Like most small size teams, TPOTs uses a global vision system. Ball position, opponent robots positions, team robots positions and orientations as well as landmarks on the field (e.g. goal keeper area) are captured using a global CCD camera.
Kiat, N.B., Ming, Q.Y., Hock, T.B., Yee, Y.S. & Koh, S.

LuckyStar II - Team Description Paper


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 543-546 inproceedings

DOI URL

Abstract: Our first robotic soccer team, LuckyStar, competed in Stockholm 1999 and was lucky enough to win third place. From that experience, we found that our team weaknesses were due to lack of vision reliability, smooth robot movement control, shooting capability and team cooperation. We decided to concentrate on areas with the most immediate impact on the game, i.e. vision, robot movement control and shooting mechanism.
Kobayashi, Y.

Team ARAIBO


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 619-622 inproceedings

DOI URL

Abstract: Last year, we introduced heading and diving action to the robots. But these actions were actually carried out few times. There were two main problems. First, the measurement of the ball position was not accurate enough to stop in front of the ball. Second, it took much time to actuate these actions because the motion arbitration was not successfully achieved. Thus, we focused on developing ball recognition, walking motions, and ball operations. We proceeded to the quarter final.
Koto, T.

TakAI


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 453-456 inproceedings

DOI URL

Abstract: The “Gullwing” developed by Tetsuhiko Koto since March 1999 is now assigned to the official name for the second strongest team of Takeuchi Lab. It gained the third position in the Japan Cup Open ’99 [1] held in May 1999. Since then, it has been thoroughly revised and it participated the Japan Autumn Camp[2] held in 1999, with the official name YowAI-2 at that time.
Kozhushkin, A.N.

PSI Team


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 457-460 inproceedings

DOI URL

Abstract: Team PSI was developed at Program Systems Institute. This paper is a description of the planning method and common principles of the architecture those we use to construct our software agents. Basic skills and roles of every agent are presented by means of the set of elementary plans. The purpose of the planning process is to compose the extended plan defining the behaviour of the agent from elementary ones. The planning system (or just a planner), built in the agent, modifies extended plans depending on external conditions and the internal state of the agent. It adds new elementary plans to the extended one refining it and controls the execution of elementary plans in a body of the extended plan.
Kubo, T.

Gnez: Adapting Knowledge to the Environment with GA


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 461-464 inproceedings

DOI URL

Abstract: This paper describes the Gnez RoboCup simulation team. we have investigated an evolutional acquisition method of the situation recognizer for a RoboCup soccer agent. We apply Hierarchical Fuzzy Intelligent Controller to decide soccer agents’ behavior, and implementing the behavior rules, which consist of condition and behavior. We try to acquire ball-kickable condition recognizer with Genetic Algorithm. On a single agent game, we get gene that can recognize ball-kickable condition actually. And the recognizer is effective enough to use as ball- kickable condition.
Kuwata, Y. & Shinjoh, A.

Design of Robocup-Rescue Viewers - Towards a Real World Emergency System -


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 159-168 inproceedings

DOI URL

Abstract: In this paper the design and implementation of a prototype viewer system for the RoboCup-Rescue framework (Version 0) is described. We discuss the requirements for the visualization of disaster information and propose a workflow model for the disaster mitigation process. A prototype of disaster information system has been built, based on this model, and here we discuss the system from the viewpoint of information visualizations.
Lenser, S., Bruce, J. & Veloso, M.

CMPack '00


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 623-626 inproceedings

DOI URL

Abstract: This is a description of Carnegie Mellon University’s entry in the Sony legged league of RoboCup 2000. See our web page for more details [4]. The main components of our system are: vision, localization, behaviors(including a basic world model), and motions. The main changes for this year are: basic world model, new behaviors/behavior architecture, walking and kicking motions. We placed third in the competition, losing only to the first place team.
Maire, F. & Taylor, D.

A Quadratic Programming Formulation of a Moving Ball Interception and Shooting Behaviour, and Its Application to Neural Network Control


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 327-332 inproceedings

DOI URL

Abstract: A desirable elementary behaviour for a robot soccer player is the moving ball interception and shooting behaviour, but generating smooth, fast motion for a mobile robot in a changing environment is a difficult problem. We address this problem by formulating the specifications of this behaviour as a quadratic programming optimisation problem, and by training a neural network controller on the exact solution computed off-line by a quadratic programming problem optimiser. We present experimental results showing the validity of the approach and discuss potential applications of this approach in the context of reinforcement learning.
Marceau, G.

The McGill's RedDogs Legged League System


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 627-630 inproceedings

DOI URL

Abstract: McGill University was back on the Robocup fields for a second year of four legged robot competition. As always, we preferred scalable algorithms and long term architectures.
This report will describe the aspects of our system which might inspire future work : section 2 discuss the support code, sections 3 and 4 review algorithms and give empirical data of performance. Finally, section 5 on behavior should give a good view at the style of decisions the system was able to make.
Marchese, F.M. & Sorrenti, D.G.

Omni-Directional Vision with a Multi-Part Mirror


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 179-188 inproceedings

DOI URL

Abstract: This paper presents an omni-directional sensor based on a camera and a mirror generated with a surface of revolution. The requirements the device must fulfill result from its use as the main perception system for the autonomous mobile robots used in F2000 RoboCup competitions. The more relevant requirements which have been pursued are: 1) range sensing in a quite wide region centered around the robot, with good accuracy; 2) sensing around the robot in a given vertical sector, in order to recognize team-mates and adversaries (all robots have a colored marker above a given height); 3) range sensing in a region very close around the robot, with the highest accuracy, to locate and kick the ball. Such requirements have been fulfilled by the design of a mirror built up of three different parts. Each part is devoted to the fulfillment of one requirement. Concerning the first requirement the approach developed is based on the design of a mirror’s profile capable to optically compensate the image distortion provided by the mirror profiles commonly used in previous literature. This approach resulted to be similar to a previous work by Hicks and Bajcsy, although independently developed by the authors.
Marques, C.F. & Lima, P.U.

A Localization Method for a Soccer Robot Using a Vision-Based Omni-Directional Sensor


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 96-107 inproceedings

DOI URL

Abstract: In this paper, a method for robot self-localization based on a catadioptric omni-directional sensor is introduced. The method was designed to be applied to fully autonomous soccer robots participating in the middle-size league of RoboCup competitions. It uses natural landmarks of the soccer field, such as field lines and goals, as well as a priori knowledge of the field geometry, to determine the robot position and orientation with respect to a coordinate system whose location is known. The landmarks are processed from an image taken by an omni-directional vision system, based on a camera plus a convex mirror designed to obtain (by hardware) the ground plane bird’s eye view, thus preserving field geometry in the image. Results concerning the method’s accuracy are presented.
This work was supported by grant PRAXIS XXI /BM /21091 /99 of the Portuguese Foundation for Science and Technology
Acknowledgements The authors would like to thank Luis Custodio, José Santos-Victor and Rodrigo Ventura for the fruitful discussions about the subject of this paper.
McAllester, D. & Stone, P.

Keeping the Ball from CMUnited-99


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 333-338 inproceedings

DOI URL

Abstract: This paper presents preliminary results achieved during our current development of a team for simulated robotic soccer in the RoboCup soccer server [2]. We have constructed a team that plays a simplified “keepaway” game. Playing keepaway against the 1999 RoboCup champion CMUnited-99 team, our new program holds the ball for an average of 25 second with an average distance of 24 meters from the opponents end of the field. CMUnited-99 playing against itself holds the ball for an average of only 6 seconds. Here we describe the design of the keepaway team. The principal technique used is the vector sum of force-fields for governing player motion when they are not in possession of the ball.
Mitsunaga, N. & Asada, M.

Observation Strategy for Decision Making Based on Information Criterion


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 189-198 inproceedings

DOI URL

Abstract: Self localization seems necessary for mobile robot navigation. The conventional method such as geometric reconstruction from landmark observations is generally time-consuming and prone to errors. This paper proposes a method which constructs a decision tree and prediction trees of the landmark appearance that enable a mobile robot with a limited visual angle to observe efficiently and make decisions without global positioning in the environment. By constructing these trees based on information criterion, the robot can accomplish the given task efficiently. The validity of the method is shown with a four legged robot.
Mitsunaga, N., Nagai, Y. & Asada, M.

BabyTigers: Osaka Legged Robot Team


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 631-634 inproceedings

DOI URL

Abstract: Our interests are learning issues such as action selection, observation strategy without 3D-reconstruction, and emergence of walking. This year we focused our development on embodied trot walking and behavior of goal keeper.
We consider that our embodied walking showed the fastest movement in the all twelve teams since we got 2nd place in the RoboCup Challenge 1 and 2, also achieved shortest time in the RoboCup Challenge 3 in spite of our stop-observe- act approach.
Morishita, T., Kawarabayashi, T., Shimora, H., Nishino, J. & Ogura, H.

Zeng00: The Realization of Mixed Strategy in Simulation Soccer Game


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 465-468 inproceedings

DOI URL

Abstract: We propose and examine a time mixed strategy method. We can obtain the optimal mixed strategy in a simulation soccer by means of this method, even if we don’t have the payoff matrix which consists of both teams’ strategy. We prepared two symmetric strategys, “side attack strategy” and “center attack strategy” and examine. The result is that proposed approach could obtain optimal mixed strategy in soccer game.
Murphy, R.R., Casper, J. & Micire, M.

Potential Tasks and Research Issues for Mobile Robots in RoboCup Rescue


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 339-344 inproceedings

DOI URL

Abstract: Previous work [5] has summarized our experiences working with the Hillsborough Fire Rescue Department and FEMA documents pertaining to Urban Search and Rescue. This paper discusses the lessons learned and casts them into four main categories of tasks for the physical agent portion of RoboCup-Rescue: 1) reconnaissance and site assessment, 2) rescuer safety, 3) victim detection, and 4) mapping and characterizing the structure.
Murray, J., Obst, O. & Stolzenburg, F.

Towards a Logical Approach for Soccer Agents Engineering


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 199-208 inproceedings

DOI URL

Abstract: Building agents for a scenario such as the RoboCup simulation league requires not only methodologies for implementing high-level complex behavior, but also the careful and efficient programming of low-level facilities like ball interception. With this hypothesis in mind, the development of RoboLog Koblenz has been continued. As before, the focus is laid on the declarativity of the approach. This means, agents are implemented in a logic- and rule-based manner in the high-level and flexible logic programming language Prolog. Logic is used as a control language for deciding how an agent should behave in a situation where there possibly is more than one choice.
In order to describe the more procedural aspects of the agent’s behavior, we employ state machines, which are represented by statecharts. Because of this, the script language for modeling multi-agent behavior in [8] has been revised, such that we are now able to specify plans with iterative parts and also reactive behavior, which is triggered by external events. In summary, multi-agent behavior can be described in a script language, where procedural aspects are specified by statecharts and declarative aspects by logical rules (in decision trees). Multi-agent scripts are implemented in Prolog. The RoboLog kernel is written in C++ and makes now use of the low-level skills of the CMUnited-99 simulator team.
Murray, J., Obst, O. & Stolzenburg, F.

RoboLog Koblenz 2000


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 469-472 inproceedings

DOI URL

Abstract: RoboCup-2000 was the second world championship, RoboLog Koblenz participated in. To the best of our knowledge, RoboLog Koblenz is the only team in the simulation league, whose agents are programmed in Prolog. The main goal of our research is to make declarative agent programming possible, by writing the agents’ control programs in SWI–Prolog [6]. Procedural aspects of the agents’ behavior can be specified by statecharts, that have become part of the Unified Modeling Language (UML) [4]. See also the paper Towards a logical approach for soccer agents engineering in this volume.
Nagasaka, Y., Murakami, K., Naruse, T., Takahashi, T. & Mori, Y.

Potential Field Approach to Short Term Action Planning in RoboCup F180 League


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 345-350 inproceedings

DOI URL

Abstract: We propose a potential field approach to represent a game situation. In a potential field, a ball should be moved according to the gradient of the potential field. There are three kinds of potential fields. One is defined for a game field, and another is defined for each robot. The third field is defined as the combination of these two. The combined field is used for evaluation of a situation. We applied this method to our robot control program. Potential values are used to determine the direction in which a robot kicks a ball. We compared the potential field based strategy and an usual “if then” type rule based strategy. The potential field based strategy makes better decisions in several cases and no worse decisions than the rule based strategy.
Nakagawa, Y., Okuno, H.G. & Kitano, H.

Bridging Gap Between the Simulation and Robotics with a Global Vision System


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 209-218 inproceedings

DOI URL

Abstract: A wide gap lies between the simulation league and real robot leagues of RoboCup, in particular in applications. In the simulation league, the soccer server maintains all data on the pitch and feeds such data to applications via networks. Therefore, several applications such as 2D- and 3D-viewers, commentary systems have been developed. In this paper, we propose that using a global vision system can bridge the gap of applications. A global vision system, which is usually used in the small size league, is extended to serve as a data feeder. Its main issues include missing object identification (due to occlusion and feature extraction failure), errors in player identification (due to mis-assignment of objects among continuous frames), detection of stoppage, and elimination of unnecessary scenes. To cope with these issues, the proposed system uses estimation and scene analysis. We also present an extension of the soccer server protocols to feed state specific to the small size league. The resulting system succeeded in feeding data of the final game of RoboCup’99 to a 2D-viewer and a commentary system.
Nakamura, T., Oohara, M., Ebina, A., Imai, M., Ogasawara, T. & Ishiguro, H.

Real-Time Estimating Spatial Configuration Between Multiple Robots by Triangle and Enumeration Constraints


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 219-228 inproceedings

DOI URL

Abstract: In the multi-agent environment, it is important to identify position and orientation of multiple robots for accomplishing a given task in cooperative manner. This paper proposes a method for estimating position and orientation of multiple robots using multiple omnidirectional images based on geometrical constraints. Our method reconstruct not only relative configuration between robots but also absolute one using the knowledge of landmarks in the environment. Even if there are some obstacles in the environment, our method can estimate absolute configuration between robots based on the results of self-localization of each robot.
Nakamura, T., Takeda, H., Terada, T., Oohara, M., Yamamoto, T. & Takeda, M.

The RoboCup-NAIST


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 591-594 inproceedings

DOI URL

Abstract: Through robotic soccer issue, we focus on “perception” and “situation and behavior” problem among RoboCup physical agent challenges [1]. So far, we have implemented some behaviors for playing soccer by combining four primitve processes (motor control, camera control, vision, and behavior generation processes)[2]. Such behaviors were not sophisticated very much because they were fully implemented by the human programmer. In order to improve the performance of such behaviors, we have applied a kind of learning algorithm during off/on-line skill development phase. For example, to acquire purposive behavior for a goalie, we have developed a robot learning method based on system identification approach. We also have developed the vision system with on-line visual learning function [3]. This vision system can adapt to the change of lighting condition in realtime.
Nardi, D., Castelpietra, C., Guidotti, A., Salerno, M. & Sanitati, C.

S.P.Q.R.


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 635-638 inproceedings

DOI URL

Abstract: The SPQR (Soccer Player Quadruped Robots, but also Senatus PopulusQue Romanus) team was a newcomer of the Sony Legged League in RoboCup 2000. The work started in April 2000, based on the previous experience gained in the Middle Size League [3] and on the collaboration of Artificial Intelligence and Robotics researchers of our Department. Due to the very short time to prepare the team for the competition, we decided to focus on the previously developed software architecture, based on the explicit representation of the knowledge of the robotic agent [1, 2] and on the effective realization of some control primitives, the kick in particular. Given the above constraints, the performance of the team in Melbourne was very satisfactory. In fact, SPQR classified fourth, winning games with more experienced teams, playing very tightly with the 99 winners in the semi-final, and generally showing a good level of play.
Nishino, J., Morishita, T. & Kubo, T.

Open Zeng: an Open Style Distributed Semi-Cooperative Team Development Project from Japan


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 473-476 inproceedings

DOI URL

Abstract: The team Open Zeng is a quite interesting and unique project to study subjects on heterogeneous cooperative teamwork behavior. Twelve players from seven different university and institutions were gatherd and played soccer well against ordinary teams. We introduced the Evolutionary development model. The project provides a distributed heterogeneous team development using WWW and eMail.
Dept. of Human and Artificial Intelligent Systems, Fukui University JAPAN
Ohashi, T., Enokida, S., Minatodani, J., Kawamoto, K., Shigeoka, Y., Yoshida, T. & Ejima, T.

KIRC: Kyutech Intelligent Robot Club


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 595-598 inproceedings

DOI URL

Abstract: Autonomous soccer robots should recognize the environment from the captured image from a video camera and plan to a proper behavior. Furthermore, when some robots play cooperatively, communication system between robots is important. Because of simple vision and actuator system, the gap between the real world and simulation environment is considered to be small. Our research target is to accomplish multi-agent system using reinforcement learning based on a simulation environment.
Ohta, M.

Gemini in RoboCup-2000


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 477-480 inproceedings

DOI URL

Abstract: We implemented “Gemini” a client program for the SoccerServer. The objective of this program is testing a lot of learning methods on multi-agent environments. In the current implementation,Gemini can select the most effective strategy for an enemy, using reinforcement learning. Furthermore, we are trying to implement a meta-level learning, which turn each learning function on or off according to whether the learning succeed or not.
Ohta, M., Takahashi, T. & Kitano, H.

Robocup-Rescue Simulation: in Case of Fire Fighting Planning


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 351-356 inproceedings

DOI URL

Abstract: RoboCup-Rescue project was proposed, to examine disaster prevention andmitigation using technology from RoboCup. We have implementeda disaster simulator for RoboCup-Rescue, and use it to select the optimal distribution of fire brigades. We found the “Concentrate Strategy” is the best in this case.
Ostrowski, J.P., McIsaac, K.A., Das, A., Chitta, S. & Neiling, J.

The University of Pennsylvania RoboCup Legged Soccer Team


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 639-642 inproceedings

DOI URL

Abstract: This was a “building” year for the UPennalizers, where we focused on developing a solid foundation of interconnected modules that could be easily customized and upgraded as our team progressed.We suffered two hard fought losses in early competition, and so did not make it to the playoff rounds. We worked on developing our own walking routines- at first, relying on quasi-static stability, and more recently moving towards fully dynamic walking gaits. We also were the only team to develop an audio communication protocol that could be used to transfer small pieces of information (3-4 bits per second). Lastly, we developed new visual tracking routines that allowed for greater flexibility in how we track the ball and perform localizations.
Oswald, N., Becht, M., Buchheim, T., Burger, P., Hetzel, 1]G., Kindermann, G., Lafrenz, R., Schanz, M., Schulé, M. & Levi, P.

CoPS-Team Description


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 599-602 inproceedings

DOI URL

Abstract: This paper presents the hardware and software design principles of the medium size RoboCup Team CoPS Stuttgart which are developed by the image understanding group at the Institute for Parallel and Distributed High Performance Systems (IPVR) of the University of Stuttgart. By adapting already successfully tested multiagent software concepts by our group to the domain of robotic soccer we intend to improve those concepts at the field of realtime applications with uncertain sensory data.
de Pascalis, P., Ferraresso, M., Lorenzetti, M., Modolo, A., Peluso, M., Polesel, R., Rosati, R., Scattolin, N., Speranzon, A. & Zanette, W.

Golem Team in Middle-Sized Robots League


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 603-606 inproceedings

DOI URL

Abstract: Golem is an holonomic robot designed to be compliant with Robocup regulations for the middle-sized league. The project consists in three parts: mechanics and hardware, vision system and software.We adopted the universal three wheels model to achieve a great freedom of movement. In order to take advantage of this particular feature, the vision system has to give a full sight of the environment. This objective is achieved by a mirror mounted on the mobile base. Decision making and planning are based on the knowledge of the relative prosition of all objects in the field (ball, walls, robots, goals) and especially on the recognition of teammates and opponents. Every robot can play as goalkeeper, defender or attacker.
Polani, D. & Martinetz, T.

Team Description for Lucky Lübeck — Evidence-Based World State Estimation


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 481-484 inproceedings

DOI URL

Abstract: This paper describes the important features of the team Lucky LÜbeck Melbourne 2000 which is based on the Mainz Rolling Brains 1999 agent code. In Lucky LÜbeck, a new method, Evidence- based World state Estimation, has been introduced, by which player and ball position can be estimated about an order of magnitude more accurately than before.
Prokopenko, M., Butler, M. & Howard, T.

On Emergence of Scalable Tactical and Strategic Behaviour


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 357-366 inproceedings

DOI URL

Abstract: The principle of behavioral programming [1] suggests to derive low-level controllers from symbolic high-level task descriptions in a predictable way. This paper presents an extension of the principle of behavioral programming — by identifying a feedback link between emergent behaviour and a scalable Deep Behaviour Projection (DBP) agent architecture. In addition, we introduce a new variant of the RoboCup Synthetic Soccer, called Circular Soccer. This variant simulates matches among multiple teams on a circular field, and extends the RoboCup Simulation towards strategic game-theoretic issues. Importantly, the Circular Soccer world provides a basis for an architecture scale-ability evaluation, and brings us closer to the idea of meta-game simulation.
This paper subsumes the team description paper “Cyberoos2000: Experiments with Emergent Tactical Behaviour”.
Reis, L.P. & Lau, N.

FC Portugal Team Description: RoboCup 2000 Simulation League Champion


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 29-40 inproceedings

DOI URL

Abstract: FC Portugal is the result of a cooperation project between the Universities of Aveiro and Porto in Portugal. The project started in February 2000 and only three months later, in Amsterdam, FC Portugal became the first European Champion of RoboCup scoring a total of 86 goals without conceding a single goal. Three months later, in Melbourne, FC Portugal became RoboCup Simulation League World Champion scoring 94 goals, again without conceding any goal. This paper briefly describes some of the most relevant research developments and innovations that lead to FC Portugal team success.
Acknowledgments The authors would like to thank to Luis Seabra Lopes for having this (wonderful) idea of participating in RoboCup 2000 and for his excellent contribution on all administrative, promotional and funding work of FC Portugal as well as fruitful discussions on several topics. Our thanks goes also to Peter Stone, Patrick Riley and Manuela Veloso for making available the CMUnited99 low-level source code that saved us a huge amount of time at the beginning of the project. We would also like to thank the financial support from FCT — Portuguese Foundation for Science and Technology, Compaq Portugal, PT Innovation, LIACC, University of Aveiro and APPIA — Portuguese Association for Artificial Intelligence.
Riedmiller, M., Merke, A., Meier, D., Hoffmann, A., Sinner, A. & Thate, O.

Karlsruhe Brainstormers 2000 Team Description


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 485-488 inproceedings

DOI URL

Abstract: The main motivation behind the Karlsruhe Brainstormer’s effort in the robotic soccer simulation league of the RoboCup project is to develop and to apply Reinforcement Learning (RL) techniques in complex domains. Our long term goal is to have a learning system which is able to learn by itself the best winning behaviour. The soccer simulation domain allows more than (108x50)23 different positionings of the 22 players and the ball - the complete state space considering object velocities and player’s stamina is magnitudes larger. In every cycle, an agent can choose between more than 300 basic commands (parametrized turns and dashes), which makes a choice of 30011 joint actions for the team per cycle! A problem of such complexity is a big challenge for today’s RL methods; in the Brainstormers project we are investigating methods to practically handle learning problems of such size.
Riedmiller, M., Merke, A., Meier, D., Hoffmann, A., Sinner, A., Thate, O. & Ehrmann, R.

Karlsruhe Brainstormers - A Reinforcement Learning Approach to Robotic Soccer


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 367-372 inproceedings

DOI URL

Abstract: Our long-term goal is to build a robot soccer team where the decision making part is based completely on Reinforcement Learning (RL) methods. The paper describes the overall approach pursued by the Karlsruhe Brainstormers simulator league team. Main parts of basic decision making are meanwhile solved using RL techniques. On the tactical level, first empirical results are presented for 2 against 2 attack situations.
Riley, P., Stone, P., McAllester, D. & Veloso, M.

ATT-CMUnited-2000: Third Place Finisher in the Robocup-2000 Simulator League


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 489-492 inproceedings

DOI URL

Abstract: The ATT-CMUnited-2000 simulator team finished in third place at Robocup- 2000. It was one of only three teams to finish ahead of the previous year’s champion, CMUnited-99. In controlled experiments, ATT-CMUnited-2000 beats CMUnited-99 by an average score of 2.6-0.2.
Rojas, R., Behnke, S., Knipping, L. & Frötschl, B.

FU-Fighters 2000


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 547-550 inproceedings

DOI URL

Abstract: Our F180 team, the FU-Fighters, participated for the second time at the RoboCup Competition. Our main design goal for RoboCup’2000 was to build a heterogeneous team of robots with different shapes and characteristics, and to use new kicking devices. Different sets of robots can be selected to adapt to the strategies of other teams. Our team finished second, as the runner-up to Big Red.
de la Rosa, J.L., Innocenti, B., Montaner, M., Figueras, A., Muñoz, I. & Ramon, J.A.

RoGi Team Description


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 551-554 inproceedings

DOI URL

Abstract: RoGi team started up in 1996 at the first robot-soccer competition as the result of a doctorate course in multi-agent systems. In 1997, 1998 and 1999 it has participated at the international workshops held in Japan, Paris and Stockholm. The main goal has been always the implementation and experimentation on dynamical physical agents and autonomous systems. Through these years the platform has become a stable system, where experiments can be repeated and new theories regarding unstructured, dynamic and multi-agent world can be tested. This paper focuses on decision making and vision improvements carried out during this last year. This year results at the competition were: 0-4, 10-0, 2-0 and 0-14 at quarterfinals.
dela Rosa, J.L., Muñoz, I., Innocenti, B., Figueras, A., Montaner, M. & Ramon, J.A.

Preliminary Studies of Dynamics of Physical Agents Ecosystems


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 373-378 inproceedings

DOI URL

Abstract: This paper is a step forward from the agent ecosystems that Hogg studied [6]. We plan to extend these agents ecosystems to physical agents that interact with the physical world. The aim is to conceive algorithms for the choice of resources and to expand this work. Dynamics of choice in such ecosystems depends on pay-off functions that contain information about the real physical world. One contribution here is to formalise the choice of knowledge resources by including a consensus technique. The second contribution is to include diversity by means of physical agents and to analyse the emergent impact in terms of diversity and performance. Simulated soccer robots exemplifies all this.
Saffiotti, A., Boman, M., Buschka, P., Davidsson, P., Johansson, S. & Wasik, Z.

Team Sweden


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 643-646 inproceedings

DOI URL

Abstract: “Team Sweden” is the Swedish national team that entered the Sony legged robot league at the RoboCup ’99 and RoboCup 2000 competitions. We had two main requirements in mind when preparing our entries:
1. The entry should effectively address the specific challenges present in this domain; in particular, it should be able to tolerate errors and imprecision in perception and execution; and
2. it should illustrate our research in autonomous robotics, by incorporating general techniques that can be reused in different robots and environments. While the first requirement could have been met by writing some ad hoc competition software, the second one led us to develop principled solutions that drew upon our current research in robotics, and that pushed it further ahead.
Scerri, P., Reed, N., Wiren, T., Lönneberg, M. & Nilsson, P.

Headless Chickens IV


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 493-496 inproceedings

DOI URL

Abstract: The Headless Chickens IV (HCIV) were developed using a prototype end-user agent development environment called EASE (End-user Actor Specification Environment). The experience provided a great deal of data about general and domain specific properties of the system. A secondary research goal was to generalize the strategy editor used to specify the team strategies of last years team to specify team strategies for agent of different types via the use of an XML interface. The poor performance of the team (3 losses and a draw in the competition) is attributed to the early stage of the research as well as the very high computational complexity of the agent runtime architecture.
Schappel, B. & Schulz, F.

Mainz Rolling Brains 2000


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 497-500 inproceedings

DOI URL

Abstract: The team Mainz Rolling Brains participated in RoboCup world championships for the third time this year. The team is based on the improved technical level from last year but has a completly new decision level. The latter level was modularized and we introduced communication for world model and tactical information.
Shinoda, K. & Kunifuji, S.

Team Description of Spatial-Timer


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 501-504 inproceedings

DOI URL

Abstract: Our team aims at the improvement of soccer agent’s ability of situaltion recognition and action dicition by saving environmental information as a spatial image. In real time simulation, such as soccer game, the agent needs to have all environmental informaion in order to make a right judgement of situation and make decition own action[Ito99, Nitta88]. However, soccer agents are limited the sight information are acquired form soccer server[SS98]. Then, our team aims at the improvement of agent’s ability of situation judgement with saving of the spatial image in the spatio-temporal databese[Erwing98]. As a result, soccer agent can make judement own environment with not only sight information from soccer server but enviromnet outside the sight range used by this spatiotemporal databese. However, the inforamtion, which is given from socer server, is uncertainty and incompletement. Therfore, it is difficult for agent to make the spatio-temporal database. Then, when a spatial information was saved in the database, the agent made to complete the lack information used by the difference of these items of information in our team. As a result, our agent can make a judgement own environment with virtual spatial information of the spatio-temporal databese, whose information is wider than the range of normal range of the agent. The following, this paper explains our spatio-temporal database and the methods complement of lack information in this database.
Sim, R., Russo, P., Graham, A., Blair, A., Barnes, N. & Blair, A.

RoboMutts


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 647-650 inproceedings

DOI URL

Abstract: A software control system for an autonomous mobile robot must be capable of performing many complex processing tasks in real time. Such tasks may include localising within the environment, recognising objects or safely avoiding static and dynamic obstacles. The focus for our team this year was to build a solid structural base, to be used in future RoboCup competitions. This led us to concentrate on development of effective reusable components such as a robust vision system. The RoboMutts team was a joint venture between the University of Melbourne and RMIT University. This paper focuses on the parts of the system developed at the University of Melbourne.
Simon, M., Behnke, S. & Rojas, R.

Robust Real Time Color Tracking


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 239-248 inproceedings

DOI URL

Abstract: This paper describes the vision system that was developed for the RoboCup F180 team FU-Fighters.
The system analyzes the video stream captured from a camera mounted above the field. It localizes the robots and the ball predicting their positions in the next video frame and processing only small windows around the predicted positions. Several mechanisms were implemented to make this tracking robust. First, the size of the search windows is adjusted dynamically. Next, the quality of the detected objects is evaluated, and further analysis is carried out until it is satisfying. The system not only tracks the position of the objects, but also adapts their colors and sizes. If tracking fails, e.g. due to occlusions, we start a global search module that localizes the lost objects again. The pixel coordinates of the objects found are mapped to a Cartesian coordinate system using a non-linear transformation that takes into account the distortions of the camera. To make tracking more robust against inhomogeneous lighting, we modeled the appearance of colors in dependence of the location using color grids. Finally, we added a module for automatic identification of our robots. The system analyzes 30 frames per second on a standard PC, causing only light computational load in almost all situations.
Stankevitch, L. & Akhapkin, S.

Polytech100


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 505-508 inproceedings

DOI URL

Abstract: Soccer makes a good example of the problem of the real world, which is moderately abstracted. This play has being chosen as one of standard problems for study on multi-agent systems. We are developing the soccer agent basing on the cognitive approach. In this year, we focused on synchronization problems and correct field representation into agent memory.
Stone, P., Asada, M., Balch, T., Fujita, M., Kraetzschmar, G., Lund, H., Scerri, P., Tadokoro, S. & Wyeth, G.

Overview of Robocup-2000


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 1-29 inproceedings

DOI URL

Abstract: The Fourth Robotic Soccer World Championships was held from August 27th to September 3rd, 2000, at the Melbourne Exhibition Center in Melbourne, Australia. In total, 83 teams, consisting of about watched the events. RoboCup-2000 showed dramatic improvement over past years in each of the existing robotic soccer leagues (legged, smallsize, mid-size, and simulation), while introducing RoboCup Jr. competitions and RoboCup Rescue and Humanoid demonstration events. The RoboCup Workshop, held in conjunction with the championships, provided a forum for exchange of ideas and experiences among the different leagues. This article summarizes the advances seen at RoboCup-2000, including reports from the championship teams and overviews of all the RoboCup events.
Acknowledgements RoboCup-2000 was sponsored by Sony, SGI, FujiXerox, Australon, and RMIT University. The sections of this article reect the work of the indicated authors along with Tomoichi Takahashi and the RoboCup-Rescue team (rescue) and Dominique Duhaut (humanoid).
Stone, P., Sutton, R.S. & Singh, S.

Reinforcement Learning for 3 Vs. 2 Keepaway


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 249-258 inproceedings

DOI URL

Abstract: As a sequential decision problem, robotic soccer can benefit from research in reinforcement learning. We introduce the 3 vs. 2 keepaway domain, a subproblem of robotic soccer implemented in the RoboCup soccer server. We then explore reinforcement learning methods for policy evaluation and action selection in this distributed, real-time, partially observable, noisy domain. We present empirical results demonstrating that a learned policy can dramatically outperform hand-coded policies.
Suzuki, T., Asahara, S., Kurita, H. & Takeuchi, I.

Team YowAI-2000 Description


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 509-512 inproceedings

DOI URL

Abstract: Team YowAI-2000 is an improved version of YowAI-1999 which was originally developed by Takashi Suzuki. It won the RoboCup Japan Open 2000 Championship. However, it does only slightest dynamic cooperation among agents. For example, say and hear commands are not used. It was developed as an experiment to examine how far player’s individual skill can go without cooperation.
Takahashi, T., Takeuchi, I., Koto, T., Tadokoro, S. & Noda, I.

RoboCup-Rescue Disaster Simulator Architecture


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 379-384 inproceedings

DOI URL

Abstract: RoboCup-Rescue project aims to simulate large urban disasters and rescue agents’ activities. The simulator must support both simulation of heterogeneous agents such as fire fighters, victims’ behaviors and interface to disaster’s environments in the real world. RoboCup- Rescue simulator is a comprehensive urban disaster simulator into which a new disaster simulator or rescue agents can be easily plugged. In this paper, the simulator’s specification, based on the Hanshin-Awaji Earthquake, and the simulator’s architecture are described.
Takahashi, Y., Takeda, M. & Asada, M.

Improvement Continuous Valued Q-Learning and Its Application to Vision Guided Behavior Acquisition


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 385-390 inproceedings

DOI URL

Abstract: Q-learning, a most widely used reinforcement learning method, normally needs well-deffined quantized state and action spaces to con- verge. This makes it difficult to be applied to real robot tasks because of poor performance of learned behavior and further a new problem of state space construction. We have proposed Continuous Valued Q-learning for real robot applications, which calculates contribution values to estimate a continuous action value in order to make motion smooth and effective
This paper proposes an improvement of the previous work, which shows a better performance of desired behavior than the previous one, with roughly quantized state and action. To show the validity of the method, we applied the method to a vision-guided mobile robot of which task is to chase a ball.
Takahashi, Y., Uchibe, E., Tamura, T., Yanase, M., Ikenoue, S., Inui, S. & Asada, M.

Osaka University "Trackies 2000


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 607-610 inproceedings

DOI URL

Abstract: This is the team description of Osaka University “Trackies” for RoboCup2000. The hardware and software architecture are presented.
Tanaka, N. & Yamamoto, M.

11monkeysII


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 513-514 inproceedings

DOI URL

Abstract: Constructing a system that can cope with a dynamically changing environment is one of the greatest interests in the field of software engineering.And, biologically inspired systems, such as brain-nervous system, genetic system are already modeled as neural networks and genetic algorithms(GAs in short), and its usefulness is reported in various fields.Ho wever, despite of its advanced information processing system, immune system and endocrine system still has a long way to go before it is modeled like other ones.This may due to their complexity and randomness of the system.Nev ertheless, the nature of the biological immune system, it is dedicated to self-preservation under hostile environment and enables the creature to maintain its life.Imm une system is also capable of learning, memory, and pattern recognition.
Visser, U., Drücker, C., Hübner, S., Schmidt, E. & Weland, H.-G.

Recognizing Formations in Opponent Teams


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 391-396 inproceedings

DOI URL

Abstract: The online coach within the simulation league has become more powerful over the last few years. Therefore, new options with regard to the recognition of the opponents strategy are possible. For example, the online coach is the only player who gets the information of all the objects on the £eld. This leads to the idea determine the opponents play system by the online coach and then choose an effective counter-strategy. This has been done with the help of an arti£cial neural network and will be discussed in this paper. All soccer-clients are initialized with a speci£c behavior and can change their behavior to an appropriate mode depending on the coach’s commands. The result is a ¤exible and effective game played by the eleven soccer-clients.
Weigel, T., Auerbach, W., Dietl, M., Dumler, B., Gutmann, J.-S., Marko, K., Müller, K., Nebel, B., Szerbakowski, B. & Thiel, M.

CS Freiburg: Doing the Right Thing in a Group


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 52-63 inproceedings

DOI URL

Abstract: The success of CS Freiburg at RoboCup 2000 can be attributed to an effective cooperation between players based on sophisticated soccer skills and a robust and accurate self-localization method. In this paper, we present our multiagent coordination approach for both, action and perception, and our rich set of basic skills which allow to respond to a large range of situations in an appropriate way. Furthermore our action selection method based on an extension to behavior networks is described. Results including statistics from CS Freiburg final games at RoboCup 2000 are presented.
This work has been partially supported by Deutsche Forschungsgemeinschaft (DFG), by Medien- und Filmgesellschaft Baden-Württemberg mbH (MFG), and by SICK AG, who donated a set of laser range finders and supported us in constructing a new kicker.
Wendler, J., Brüggert, S., Burkhard, H.-D. & Myritz, H.

Fault-Tolerant Self Localization by Case-Based Reasoning


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 259-268 inproceedings

DOI URL

Abstract: In this article we present a case-based approach for the selflocalization of autonomous robots based on local visual information of landmarks. The goal is to determine the position and the orientation of the robot sufficiently enough, despite some strongly incorrect visual information. Our approach to solve this problem makes use of case-based reasoning methods.
Werner, M., Myritz, H., Düffert, U., Lötzsch, M. & Burkhard, H.-D.

Humboldt Heroes


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 651-654 inproceedings

DOI URL

Abstract: Our last years approach to the Sony legged league suffered from a number of drawbacks:
- The motion system was ways to slow and imprecise.
- The different tasks were not well coordinated. The robot got quite a number of deadlock situations and was not real-time at all.
- The controlling based on a pure reactive approach, that did not allow any real planing.
With this year’s code, we tried to overcome some of these disadvantages.
We partly succeed. We get second in group and went to the final round. However, we never had any chance in the quarter final against the cup-dominating UNSW team.
Wyeth, G., Tews, A. & Browning, B.

UQ RoboRoos: Kicking on to 2000


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 555-558 inproceedings

DOI URL

Abstract: The UQ RoboRoos have been developed to participate in the RoboCup robot soccer small size league. This paper overviews the history of the team, and provides details of some key factors to the teams success in 2000: a new goalkeeper design, robust communications and smooth, fast navigation. The paper concludes with some thoughts on the future of the RoboRoos.
Wünstel, M., Polani, D., Uthmann, T. & Perl, J.

Behavior Classification with Self-Organizing Maps


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 108-118 inproceedings

DOI URL

Abstract: We describe a method that applies Self-Organizing Maps for direct clustering of spatio-temporal data. We use the method to evaluate the behavior of RoboCup players. By training the Self-Organizing Map with player data we have the possibility to identify various clusters representing typical agent behavior patterns. Thus we can draw certain conclusions about their tactical behavior, using purely motion data, i.e. logfile information. In addition, we examine the player-ball interaction that give information about the players’ technical capabilities.
Yamasaki, F., Matsui, T., Miyashita, T. & Kitano, H.

PINO The Humanoid : A Basic Architecture


2001 RoboCup 2000: Robot Soccer World Cup IV, pp. 269-278 inproceedings

DOI URL

Abstract: In this paper, we present a basic architecture and design principle behind Pino, a humanoid for RoboCup Humanoid League. Pino is designed to be a platform for research in robotics and AI. There are four major issues in Pino’s design; (1) high DOF system to realize various behaviors, (2) exterior design, (3) cheap mechanical components, and (4) a basic behavioral control systems. These issues are specifically addressed in this paper in order to illustrate basic architecture and components for humanoid platform that can be widely used for many RoboCup researchers in the world.