RoboCup 1998 Publications

From RoboCup Wiki
Jump to: navigation, search
Author Title Year Journal/Proceedings Reftype DOI/URL
Andou, T.

Andhill-98: A RoboCup Team Which Reinforces Positioning with Observation

1999 RoboCup-98: Robot Soccer World Cup II, pp. 338-345 inproceedings


Abstract: On reinforcement learning with limited exploration, an agent’s policy tends to fall into a worthless local optimum. This paper proposes Observational Reinforcement Learning method with which the learning agent evaluates inexperienced policies and reinforces it. This method provides the agent more chances to escape from a local optimum without exploration. Moreover, this paper shows the effectiveness of the method from experiments in the RoboCup positioning problem. They are advanced experiments described in our RoboCup-97 paper [1].
This work was mainly done when the author was in Dept. of Mathematical and Computing Sciences, Tokyo Institute of Technology.
Andre, D. & Teller, A.

Evolving Team Darwin United

1999 RoboCup-98: Robot Soccer World Cup II, pp. 346-351 inproceedings


Abstract: The RoboCup simulator competition is one of the most challenging international proving grounds for contemporary AI research. Exactly because of the high level of complexity and a lack of reliable strategic guidelines, the pervasive attitude has been that the problem can most successfully be attacked by human expertise, possibly assisted by some level of machine learning. This led, in RoboCup’97, to a field of simulator teams all of whose level and style of play were heavily influenced by the human designers of those teams. In contrast, our 1998 team was “designed” entirely by the process of genetic programming. Our evolved team placed in the middle of the pack at Robocup98, despite the fact that it was largely machine learned rather than hand coded. This paper presents our motivation, our approach, and the specific construction of our team that created itself from scratch.
Aparício, P., Ventura, R., Lima, P. & Pinto-Ferreira, C.

ISocRob — Team Description

1999 RoboCup-98: Robot Soccer World Cup II, pp. 434-439 inproceedings


Abstract: The SocRob project was born as a challenge for multidisciplinary research on broad and generic approaches for the design of cooperating society of robots, involving Control, Robotics and Artificial Intelligence researchers. A case study on Robotic Soccer played by a team of 3 robots is currently underway. The team competed at the World Cup of Robotic Soccer, RoboCup98, held in Paris, France, and qualified for the quarter-finals of the midlle-size league. In this paper we present the team description, the implemented behaviors and the results obtained.
This work has been supported by grants from the following Portuguese institutions: Fundação Calouste Gulbenkian, Fundação para a Ciência e a Tecnologia, PRAXIS XXI/BM/12937/97.
Asada, M., Veloso, M., Tambe, M., Noda, I., Kitano, H. & Kraetzschmar, G.K.

Overview of RoboCup-98

1999 RoboCup-98: Robot Soccer World Cup II, pp. 1-21 inproceedings


Abstract: RoboCup is an increasingly successful attempt to promote the full integration of AI and robotics research. Following the astonishing success of the first RoboCup-97 at Nagoya [1], the Second Robot World Cup Soccer Games and Conferences, RoboCup-98, was held at Paris during July 2nd and 9th, 1998 at the partly same place and period of the real world cup. There are three kinds of leagues: the simulation league, the real robot small-size league, and the real robot middle-size league. The champion teams are CMUnited-98 (CMU, USA) for both the simulation and the real robot small-size leagues, and CS-Freiburg (Freiburg, Germany) for the real robot middle-size league. The Scientific Challenge Award was given to three research groups (Electrotechnical Laboratory (ETL), Sony Computer Science Laboratories, Inc., and German Research Center for Artificial Intelligence GmbH (DFKI)) for their simultaneous development of fully automatic commentator systems for RoboCup simulator league. Over 15,000 spectators and 120 international media covered the competition worldwide. RoboCup-99, the third Robot World Cup Soccer Games and Conferences, will be held at Stockholm in conjunction with the Sixteenth International Joint Conference on Artificial Intelligence (IJCAI-99) at the beginning of August, 1999.
Binsted, K. & Luke, S.

Character Design for Soccer Commentary

1999 RoboCup-98: Robot Soccer World Cup II, pp. 22-33 inproceedings


Abstract: In this paper we present early work on an animated talking head commentary system called Byrne. The goal of this project is to develop a system which can take the output from the RoboCup soccer simulator, and generate appropriate affective speech and facial expressions, based on the character’s personality, emotional state, and the state of play. Here we describe a system which takes pre-analysed simulator output as input, and which generates text marked-up for use by a speech generator and a face animation system. We make heavy use of inter-system standards, so that future versions of Byrne will be able to take advantage of advances in the technologies that it incorporates.
Birk, A., Walle, T., Belpaeme, T., Parent, J., Vlaminck, T.D. & Kenn, H.

The Small League RoboCup Team of the VUB AI-Lab

1999 RoboCup-98: Robot Soccer World Cup II, pp. 410-415 inproceedings


Abstract: The paper describes the VUB AI-Lab team competing in the small robots league of RoboCup ’98 in Paris. The approach of this team targets for a longterm evolution of different robots, team-structures, and concepts. Therefore, the efforts for the ’98 participation focus on the provision of a exible architecture, which forms a basis for this goal. In doing so, the development of the so-called RoboCube constitutes a milestone on this road. The RoboCube is an extremely compact robot controller providing quite some computation power, memory, various I/O interfaces, and radio communication. It facilitates the use of many sensors and effectors, including their on-board processing, allowing to explore a large space of different robots and team set-ups. Accordingly, the ’98 VUB AI-Lab team consists of heterogeneous robots. The paper describes their important electromechanical features like drive units and shooting mechanisms. In addition, the radio communication, the basic control algorithms and the global vision system are described. Last but not least, first steps towards a general coordination scheme are explained, which is meant to cope with changing compositions of heterogeneous groups.
Boman, M., Åberg, H., Åhman, Å., Andreasen, J., Danielson, M., Jansson, C.-G., Kummeneje, J., Verhagen, H. & Walter, J.

UBU: Utility-Based Uncertainty Handling in Synthetic Soccer

1999 RoboCup-98: Robot Soccer World Cup II, pp. 352-357 inproceedings


Abstract: The UBU RoboCup team is described. Intuitive ideas and general objects for the participating researchers and their students are presented. The UBU team participates in the simulation league. The key idea is to repeatedly use the advice provided by a normative pronouncer (or decision module) when choosing what to do next. The pronouncer acts on input from each individual player, the basis of which is stored in a local information base. The team is under continuous development: At the time of writing, no version of UBU makes extensive use of pronouncer calls.
Bräunl, T.

CIIPS Glory Soccer Robots with Local Intelligence

1999 RoboCup-98: Robot Soccer World Cup II, pp. 416-421 inproceedings


Abstract: Our team was named after the new and successful Perth Glory soccer team. The heart of our robots are the EyeBot controllers, which we developed form scratch. We use a Motorola 68332 32-bit controller, which offers a variety of digital/analog I/O facilities. We developed our own operating system RoBIOS for these systems, which allows a great deal of flexibility. All image processing and planning is done locally on-board the EyeBot. We do not use any global sensor systems. The same EyeBot controller is also used for 6-legged and biped walking machines, and — as a boxed version — for undergraduate courses on assembly language programming.
Coradeschi, S. & Malec, J.

How to Make a Challenging AI Course Enjoyable Using the RoboCup Soccer Simulation System

1999 RoboCup-98: Robot Soccer World Cup II, pp. 120-124 inproceedings


Abstract: In this paper we present an AI programming organised around the RoboCup soccer simulation system. The course participants create a number of software agents that form a team, and participate in a tournament at the end of the course. The use of a challenging and interesting task, and the incentive of having a tournament has made the course quite successful, both in term of enthusiasm of the students and of knowledge acquired. In the paper we describe the structure of the course, discuss in what respect we think the course has met its aim, and the opinions of the students about the course.
A version of this paper in Japanese has appeared in the Journal of Robotic Society of Japan, special issue in Robotics and Education, May 98.
da Costa, A.C.P.L. & Bittencourt, G.

UFSC-Team: A Cognitive Multi-Agent Approach to the RoboCup’98 Simulator League

1999 RoboCup-98: Robot Soccer World Cup II, pp. 371-376 inproceedings


Abstract: The RoboCup Simulator League competition is a very interesting laboratory for open cognitive multi-agent systems. It presents an environment where two robot teams play soccer, with all the challenges that this task brings-up. In this environment each player has its own vision system, there is no single agent that has a global view of the field. Another restriction is that a very narrow communication band-width is allowed to inter-agent communication. The UFSC-Team adopts a cognitive multi-agent approach where most part of the cooperation process is based on visual information and where the exchanged messages are just used to decide the role each agent should play in a predefined strategy.
Driessens, K., Jacobs, N., Cossement, N., Monsieurs, P. & Raedt, L.D.

Inductive Verification and Validation of the KULRoT RoboCup Team

1999 RoboCup-98: Robot Soccer World Cup II, pp. 193-206 inproceedings


Abstract: As in many multi-agent applications, most RoboCup agents are complex systems, hard to construct and hard to check if they behave as intended. We present a technique to verify multi-agent systems based on inductive reasoning. Induction allows to derive general rules from specific examples (e.g. the inputs and outputs of software systems). Using inductive logic programming, partial declarative specifications of the software can be induced. These rules can be readily interpreted by the designers or users of the software, and can in turn result in changes to the software. The approach outlined was used to test the KULRoT RoboCup simulator team, which is briefly described.
Fujita, M., Zrehen, S. & Kitano, H.

A Quadruped Robot for RoboCup Legged Robot Challenge in Paris ’98

1999 RoboCup-98: Robot Soccer World Cup II, pp. 125-140 inproceedings


Abstract: One of the ultimate dream in robotics is to create life-like robotics systems, such as humanoid robot and animal-like legged robot. We choose to build pet-type legged robot because we believe that dog-like and cat-like legged robot has major potential for future entertainment robotics markets for personal robots. However, numbers of challenges exists before any of such robot to be fielded in the real world. Robots have to be reasonably intelligent, maintains certain level of agility, and be able to engaged in some collaborative behaviors. RoboCup is an ideal challenge to foster robotics technologies for small personal and mobile robotics system. This paper, we present Sony’s legged robots system which enter RoboCup-98 Paris as a special exhibition games.
González, E., Loaiza, H., Suárez, A. & Moreno, C.

Real MagiCol 98: Team Description and Results

1999 RoboCup-98: Robot Soccer World Cup II, pp. 440-445 inproceedings


Abstract: The hardware and software architectures of the Real MagiCol robots are presented. The robots were built for the RoboCup98 competition, but future research in other subjects has been also considered. Our programming methodology is called Behavior Oriented Commands (BOCs). Relevant aspects of the vision system, the BOC model of our goalie and the team strategy are presented.
Gugenberger, P., Wendler, J., Schröter, K. & Burkhard, H.-D.

AT Humboldt in RoboCup-98 (Team Description)

1999 RoboCup-98: Robot Soccer World Cup II, pp. 358-363 inproceedings


Abstract: The paper describes the scientific goals of the virtual soccer team “AT Humboldt 98”, which became vice champion in RoboCup-98 in Paris. It is the successor of the world champion “AT Humboldt” from RoboCup-97 in Nagoya.
Gutmann, J.-S., Hatzack, W., Herrmann, I., Nebel, B., Rittinger, F., Topor, A., Weigel, T. & Welsch, B.

CS Freiburg Robotic Soccer Team: Reliable Self-Localization, Multirobot Sensor Integration, and Basic Soccer Skills, The

1999 RoboCup-98: Robot Soccer World Cup II, pp. 93-108 inproceedings


Abstract: Robotic soccer is a challenging research domain because problems in robotics, artificial intelligence, multi-agent systems and real-time reasoning have to be solved in order to create a successful team of robotic soccer players. In this paper, we describe the key components of the CS Freiburg team. We focus on the self-localization and object recognition method based on using laser range finders and the integration of all this information into a global world model. Using the explicit model of the environment built by these components, we have implemented path planning, simple ball handling skills and basic multi-agent cooperation. The resulting system is a very successful robotic soccer team, which has not lost any game yet.
This work has been partially supported by Deutsche Forschungsgemeinschaft (DFG) as part of the graduate school on Human and Machine Intelligence, by Medien- und Filmgesellschaft Baden-Württemberg mbH (MFG), and by SICK AG, who provided the laser range finders.
Hugel, V., Bonnin, P., Bouramoué, J.C., Solheid, D., Blazevic, P. & Duhaut, D.

Quadruped Robot Guided by Enhanced Vision System and Supervision Modules

1999 RoboCup-98: Robot Soccer World Cup II, pp. 485-490 inproceedings


Abstract: Legged robots taking part in real multi-agent activities represent a very innovative challenge. This domain of research requires developments in three main areas. First the robot must be able to move efficiently in every direction in its environment. The faster the motion, the better it is. Special care must be taken when designing walking pattern transitions. Then, without any exteroceptive sensor to get information about its surroundings, the robot is blind. Fortunately, the quadruped prototype on which all experiments are carried out is equipped with a enhanced vision system and vision is the best means of getting a representation of the world that can be found in Nature. Finally the machine should be brought a minimum of intelligence since it has to manage vision information and its walking gaits by itself. When involved in cooperation, confrontation or both like in the soccer play, a high level supervision task is welcome. This paper presents detailed developments of these three points and describes how they are implemented on the real robot.
Igarashi, H., Kosue, S. & Miyahara, M.

Individual Tactical Play and Pass with Communication Between Players

1999 RoboCup-98: Robot Soccer World Cup II, pp. 364-370 inproceedings


Abstract: In this paper we describe our team, Miya2, which is to participate in the simulator league of RoboCup 98. Miya2 is characterized by soccer agents that make individual tactical plays and passes by using communication between players. In our experiments, 76.6% of passes made by Miya2 soccer agents were passes made with communication. More specifically, 53.6% of passes are quick passes made with communication in which the passers only used auditory information without looking around for receivers.
Ito, N., Nakagawa, K., Du, X. & Ishii, N.

A Description-Processing System for Soccer Agents and NIT Stones 98

1999 RoboCup-98: Robot Soccer World Cup II, pp. 221-236 inproceedings


Abstract: Many conventional object-oriented models suffer some problems in the representation of the multiple objects. A multiple object is an object with multiple aspects, autonomy, and pro-activeness. This paper proposes a new agent model called EAMMO, which applies the agent-oriented paradigm to represent multiple objects. EAMMO consists of three types of agents as follows: (1) an upper-agent describes autonomous objects, (2) a lower-agent describes reective objects, and (3) an environmental-agent describes the environment including agents. We design a description-processing system for EAMMO to confirm the efficiency of our model. We describe soccer games in EAMMO. A soccer player is a good example for EAMMO, because a soccer player is a kind of multiple objects. We define only basic skills and strategies in the soccer agent. However, we found that soccer agents can corporate in a more complex strategy generated by the system. As the result, we confirm that EAMMO is an efficient model for describing multiple objects.
Jung, C.G.

Layered and Resource-Adapting Agents in the RoboCup Simulation

1999 RoboCup-98: Robot Soccer World Cup II, pp. 207-220 inproceedings


Abstract: Layered agent architectures are particularly successful in implementing a broad spectrum of (sub-)cognitive abilities, such as reactive feedback, deliberative problem solving, and social coordination. They can be seen as special instances of boundedly rational systems, i.e., systems that trade off the quality of a decision versus the cost of invested resources. For sophisticated domains, such as the soccer simulation of RoboCup, we argue that a generalised framework that combines a layered design with explicit, resource-adapting mechanisms is reasonable. Based on the InteRRaP model, we describe a prototypical setting that is to guide and to evaluate the development of reasoning about abstract resources. These are representations of general interdependencies between computational processes. The realised soccer team, CosmOz Saarbrücken, participated successfully in the RoboCup-98 competition and confirmed that abstract resources are an appropriate modelling device in layered and resource-adapting agents.
supported by a grant from the “Deutsche Forschungsgemeinschaft” (DFG).
Klupsch, M., Lückenhaus, M., Zierl, C., Laptev, I., Bandlow, T., Grimme, M., Kellerer, I. & Schwarzer, F.

Agilo RoboCuppers: RoboCup Team Description

1999 RoboCup-98: Robot Soccer World Cup II, pp. 446-451 inproceedings


Abstract: This paper describes the Agilo RoboCuppers — the RoboCup team of the image understanding group (FG BV) at the Technische Universität München. With a team of five Pioneer 1 robots, equipped with a CCD camera and single board computer each and coordinated by a master PC outside the field we participated in the medium size RoboCup league in Paris 1998. 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 necessary data for the onboard scene interpretation. These features as well as the odometric data are checked on the master PC with regard to consistency and plausibility. The results are distributed to all robots as base for their local planning modules and also used by a coordinating global planning module.
The name is derived from the Agilolfinger, which were the first Bavarian ruling dynasty in the 8th century, with Tassilo as its most famous representative.
Kraetzschmar, G.K., Enderle, S., Sablatnög, S., Boß, T., Dettinger, M., Braxmayer, H., Folkerts, H., Klingler, M., Maschke, D., Mayer, G., Muller, M., Nubeck, A., Ritter, M., Seidl, H., Worz, R. & Palm, G.

The Ulm Sparrows: Research into Sensorimotor Integration, Agency, Learning, and Multiagent Cooperation

1999 RoboCup-98: Robot Soccer World Cup II, pp. 452-457 inproceedings


Abstract: We describe the motivations, research issues, current results, and future directions of The Ulm Sparrows, a project that aims at the design and implementation of a team of robotic soccer players.
Lubbers, J. & Spaans, R.R.

The Priority/Confidence Model as a Framework for Soccer Agents

1999 RoboCup-98: Robot Soccer World Cup II, pp. 162-172 inproceedings


Abstract: We propose the Priority/Confidence model as a reasoning model for agents. Decisions are made according to a confidence measure which is based on the importance of actions (priority) and the satisfaction of a priori preconditions. We implemented the Priority/Confidence model for a robotic soccer domain, namely the RoboCup. Our team, AIACS, has demonstrated the feasibility of this model by beating the world champion of 1997, AT-Humboldt in training matches and reaching a 9th place in the RoboCup’98 tournament.
This research is carried out as part of the authors’ Master’s thesis.
Lund, H.H. & Pagliarini, L.

Robot Soccer with LEGO Mindstorms

1999 RoboCup-98: Robot Soccer World Cup II, pp. 141-151 inproceedings


Abstract: We have made a robot soccer model using LEGO Mindstorms robots, which was shown at RoboCup98 during the World Cup in soccer in France 1998. We developed the distributed behaviour-based. approach in order to make a robust and high performing robot soccer demonstration. Indeed, our robots scored in an average of 75–80% of the periods in the games. For the robot soccer model, we constructed a stadium out of LEGO pieces, including stadium light, rolling commercials, moving cameras projecting images to big screens, scoreboard and approximately 1500 small LEGO spectators who made the “Mexican wave” as known from soccer stadiums. These devices were controlled using the LEGO Dacta Control Lab system and the LEGO CodePilot system that allow programming motor reactions which can be based on sensor inputs. The wave of the LEGO spectators was made using the principle of emergent behaviour. There was no central control of the wave, but it emerges from the interaction between small units of spectators with a local feedback control.
Maeda, K., Kohketsu, A. & Takahashi, T.

Ball-Receiving Skill Dependent on Centering in Soccer Simulation Games

1999 RoboCup-98: Robot Soccer World Cup II, pp. 152-161 inproceedings


Abstract: This paper describes an effective ball-receiving skill. When soccer games are played in real life, players generally must make consecutive actions in one play, for example, running, receiving, and shooting a ball. We believe that the same is true in the case of simulation soccer games. Therefore, we designed an experiment to check how changing ball-receiving methods which is dependent on the centering patterns influence scoring goals. The experiment shows that one ball-receiving method is more effective than the others. The result is embedded into our soccer team, Kasugai-bito II, and the effectiveness is discussed in games.
Marsella, S., Adibi, J., Al-Onaizan, Y., Erdem, A., Hill, R., Kaminka, G.A., Qiu, Z. & Tambe, M.

Using an Explicit Teamwork Model and Learning in RoboCup: An Extended Abstract

1999 RoboCup-98: Robot Soccer World Cup II, pp. 237-245 inproceedings


Abstract: The RoboCup research initiative has established synthetic and robotic soccer as testbeds for pursuing research challenges in Articial Intelligence and robotics. This extended abstract focuses on teamwork and learning, two of the multi- agent research challenges highlighted in RoboCup. To address the challenge of teamwork, we discuss the use of a domain-independent explicit model of team- work, and an explicit representation of team plans and goals. We also discuss the application of agent learning in RoboCup.
Matsubara, H., Frank, I., Tanaka-Ishii, K., Noda, I., Nakashima, H. & Hasida, K.

Automatic Soccer Commentary and RoboCup

1999 RoboCup-98: Robot Soccer World Cup II, pp. 34-49 inproceedings


Abstract: This paper suggests that automated soccer commentary has a key role to play within the overall RoboCup initiative. Firstly, we identify soccer commentary as allowing and requiring investigation of a wide variety of research topics, many of which could not be addressed by the simple development of teams for the RoboCup leagues themselves. Secondly, we highlight a key task of soccer commentary: the expert analysis of a game. We suggest that this expert analysis task has the potential to make a significant impact on RoboCup challenges such as learning, teamwork, and opponent modeling. We illustrate our arguments by discussing the progress on soccer commentary systems to date, in particular reviewing our own system, Mike.
Matsumura, T.

Description of Team Erika

1999 RoboCup-98: Robot Soccer World Cup II, pp. 377-383 inproceedings


Abstract: This paper introduces the learning of basic actions by Genetic Algorithm and the effective method for design of agents’ cooperation based on role and scenario model.
Mitsunaga, N., Asada, M. & Mishima, C.

BabyTigers-98: Osaka Legged Robot Team

1999 RoboCup-98: Robot Soccer World Cup II, pp. 498-506 inproceedings


Abstract: The Osaka Legged Robot Team, BabyTigers-98, attended the First Sony Legged Robot Competition and Demonstration which was held at La Cite La Villeta, a science and technology museum in Paris in conjunction with the second Robot Soccer World Cup, RoboCup-98, July, 1998. This article describes our approach to the competition. The main feature of our system is to apply a direct teaching method to behavior acquisition for four-legged robots as one of the robot learning issues. Since the robots recognize objects in the field by color information, the color calibration is another issue. To cope with color changes due to lighting conditions, we developed an interactive color calibration tool with graphic user interface. Based on the color calibration and the learning by teaching methods, the robot successfully exhibits basic skills such as ball approaching and kicking.
Nakamura, T., Terada, K., Shibata, A. & Takeda, H.

The RoboCup-NAIST: A Cheap Multisensor-Based Mobile Robot with Visual Learning Capability

1999 RoboCup-98: Robot Soccer World Cup II, pp. 326-337 inproceedings


Abstract: Our contribution is composed of two parts: one is development of a cheap multisensor-based mobile robot, the other is development of robust visual tracking system with visual learning capability. To promote robotic soccer research, we need a low cost and portable robot with some sensors and a communication device. This paper describes how to construct a robot system which includes a lightweight and low cost mobile robot with visual, tactile sensors, TCP/IP communication device, and portable PC where Linux is running. In real world, robust color segmentation is a tough problem because color signals are very sensitive to the slight changes of lighting conditions. In order to keep visual tracking systems with color segmentation technique running in real environment, a learning method for acquiring models for image segmentation should be developed. In this paper, we also describe a visual learning method for color image segmentation and object tracking in dynamic environment. An example of the developed soccer robot system and preliminary experimental results are also shown.
Nardi, D., Clemente, G. & Pagello, E.

ART Azzurra Robot Team

1999 RoboCup-98: Robot Soccer World Cup II, pp. 458-463 inproceedings


Abstract: Azzurra Robot Team is the result of a joint effort of seven Italian research groups from Univ. of Brescia, Univ. of Genoa, Politecnico of Milano, Univ. of Padua, Univ. of Palermo, Univ. of Parma, Univ. of Roma "La Sapienza", and the Consorzio Padova Ricerche which has provided resources and a set up of the soccer field in its Center in Padua. Our goal at Robocup 1998 has been to provide a exible and low-cost experimental team to make experience before undertaking a larger project. Our long term goal is to foster the development of research and education projects on autonomous mobile robots by exploiting the RoboCup challenge.
Pagello, E., Montesello, F., Garelli, F., Candon, F., Chioetto, P. & Griggio, S.

Getting Global Performance Through Local Information in PaSo-Team'98

1999 RoboCup-98: Robot Soccer World Cup II, pp. 384-389 inproceedings


Abstract: We illustrate new improvements in PaSo-Team (The University of Padua Simulated Robot Soccer Team), a Multi-Agent System able to play soccer game for participating to the Simulator League of RoboCup competition. PaSo-Team looks like a partially reactive system built upon a number of specialized behaviors, designed for playing a soccer game that generate actions accordingly with environmental changes. Major improvements have been obtained in the design of individual skills so that the general architecture of PaSo-Team and its coordination model are exploited in a better way.
Plagge, M., Diebold, B., Günther, R., Ihlenburg, J., Jung, D., Zahedi, K. & Zell, A.

Design and Evaluation of the T-Team of the University of Tuebingen for RoboCup'98

1999 RoboCup-98: Robot Soccer World Cup II, pp. 464-472 inproceedings


Abstract: In this paper we present the hard- and software architecture of the robots of the T-Team Tuebingen, which participated in the RoboCup’98. This paper describes how we try to accomplish the basic skills of our robot team capable of successfully playing robot soccer by designing our hard- and software and the experiences we made with our team at the RoboCup’98 in Paris.
Polani, D., Weber, S. & Uthmann, T.

A Direct Approach to Robot Soccer Agents: Description for the Team Mainz Rolling rains Simulation League of RoboCup ’98

1999 RoboCup-98: Robot Soccer World Cup II, pp. 390-395 inproceedings


Abstract: In the team described in this paper we realize a direct approach to soccer agents for the simulation league of the RoboCup ’98-tournament. Its backbone is formed by a detailed world model. Based on information which is reconstructed on the world model level, the rule-based decision levels chose a relevant action. The architecture for the goalie is different from the regular players, introducing heterogeneousness into the team, which combines the advantages of the different control strategies.
Price, A.R. & Jones, T.

An Innovative Approach to Vision, Localization and Orientation Using Omnidirectional Radial Signature Analysis

1999 RoboCup-98: Robot Soccer World Cup II, pp. 299-315 inproceedings


Abstract: The greatest risk of innovative design is that it may not prove successful. That does not mean, by any account, that it should not be tried. Nor does it make the idea irrelevant. What is important is that those that follow are made aware of the problems encountered. The Omnidirectional Radial Signature Analysis Network (ORSAN) was an attempt to overcome the problems associated with the Robocup environment that were evident at RoboCup 97. In particular, lighting inconsistencies and a steering problem with the Omnidirectional Ball Based Driving Mechanism developed by this team and presented at Robocup 97. Through a series of difficulties and setbacks following the successful Nagoya event, only 16 weeks were available to produce an entire team of robots for Paris. In the end, only the prototype was ready, and so the Deakin Black Knights attempt at Robocup-98 was really over before it began. This paper details the development of the Omnidirectional Radial Signature Analysis Network, the problems it was designed to solve and the eventual conclusions that were drawn about this innovative approach to Robocup 98.
Riekki, J., Pajala, J., Tikanmäki, A. & Röning, J.

CAT Finland: Executing Primitive Tasks in Parallel

1999 RoboCup-98: Robot Soccer World Cup II, pp. 396-401 inproceedings


Abstract: We present a novel representation for agent actions. An action map specifies the preferences for all the possible different actions. The key advantage of this representation is that it facilitates executing in parallel primitive tasks. We have utilized the action map representation in playing soccer. We describe here the system that controlled our players in the second RoboCup competition.
Rowstron, A., Bradshaw, B., Crosby, D., Edmonds, T., Hodges, S., Hopper, A., S. Lloyd, J.W. & Wray, S.

The Cambridge University Robot Football Team Description

1999 RoboCup-98: Robot Soccer World Cup II, pp. 422-427 inproceedings


Abstract: This paper describes the Cambridge University Robot Foot- ball entry and our experiences in the RoboCup’98 Small Robot League of the held in Paris competition.
In the competition we came top of our group, and fourth overall. We had the strongest group, with the team coming second in the group coming second overall in the end. This meant that we were able to play a number of good games, and we were able to evaluate our approach compared to others.
This paper presents an overview of our team, compares it to some of the other approaches used, and highlights the research issues of interest to us.
Scerri, P., Coradeschi, S. & Törne, A.

A User Oriented System for Developing Behavior Based Agents

1999 RoboCup-98: Robot Soccer World Cup II, pp. 173-186 inproceedings


Abstract: Developing agents for simulation environments is usually the responsibility of computer experts. However, as domain experts have superior knowledge of the intended agent behavior, it is desirable to have domain experts directly specifying behavior. In this paper we describe a system which allows non-computer experts to specify the behavior of agents for the RoboCup domain. An agent designer is presented with a Graphical User Interface with which he can specify behaviors and activation conditions for behaviors in a layered behavior-based system. To support the testing and debugging process we are also developing interfaces that show, in real-time, the world from the agents perspective and the state of its reasoning process.
Shen, W.-M., Adibi, J., Adobbati, R., Lanksham, S., Moradi, H., Salemi, B. & Tejada, S.

Integrated Reactive Soccer Agents

1999 RoboCup-98: Robot Soccer World Cup II, pp. 286-298 inproceedings


Abstract: Robot soccer competition provides an excellent opportunity for robotics research. In particular, robot players in a soccer game must perform real-time visual recognition, navigate in a dynamic field, track moving objects, collaborate with teammates, and hit the ball in the correct direction. All these tasks demand robots that are autonomous (sensing, thinking, and acting as independent creatures), efficient (functioning under time and resource constraints), cooperative (collaborating with each other to accomplish tasks that are beyond individual’s capabilities), and intelligent (reasoning and planing actions and perhaps learning from experience). To build such integrated robots, we should use different approaches from those employed in separate research disciplines. In the 1997 RoboCup competition, the USC/ISI robot team, called Dreamteam, fought hard and won the world championship in the middle-sized robot league. These robots all share the same general architecture and basic hardware, but they have integrated abilities to play different roles (goal-keeper, defender or forward) and utilize different strategies in their behavior. Our philosophy in building these robots is to use the least possible sophistication to make them as robust as possible. This paper describes our experiences during the competition as well as our new improvements to the team.
Siegberg, A., Bredenfeld, A., Guenther, H., Kobialka, H.U., Klaassen, B., Licht, U., Paap, K.L., Ploeger, P.G., Streich, H., Vollmer, J., Wilberg, J., Worst, R. & Christaller, T.

Team Description of the GMD RoboCup-Team

1999 RoboCup-98: Robot Soccer World Cup II, pp. 473-478 inproceedings


Abstract: The article describes the structure of the GMD robots developed for the RoboCup ’98. The hardware of these robots consists of an aluminum chassis with differential drives. They have low level sensors (odometry, distance sensors etc.) and a NewtonLab vision system. The software is organized in a layered structure using a uniform design pattern on each layer. A synchronous communication paradigm is adopted for the information exchange between the different layers.
Stone, P. & Veloso, M.

Team-Partitioned, Opaque-Transition Reinforcement Learning

1999 RoboCup-98: Robot Soccer World Cup II, pp. 261-272 inproceedings


Abstract: We present a novel multi-agent learning paradigm called team-partitioned, opaque-transition reinforcement learning (TPOT-RL). TPOT-RL introduces the use of action-dependent features to generalize the state space. In our work, we use a learned action-dependent feature space to aid higher-level reinforcement learning. TPOT-RL is an effective technique to allow a team of agents to learn to cooperate towards the achievement of a specific goal. It is an adaptation of traditional RL methods that is applicable in complex, non-Markovian, multi-agent domains with large state spaces and limited training opportunities. TPOT-RL is fully implemented and has been tested in the robotic soccer domain, a complex, multi-agent framework. This paper presents the algorithmic details of TPOT-RL as well as empirical results demonstrating the effectiveness of the developed multi-agent learning approach with learned features.
This research is sponsored in part by the DARPA/RL Knowledge Based Planning and Scheduling Initiative under grant number F30602-97-2-0250. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies or endorsements, either expressed or implied, of the U. S. Government.
Stone, P., Veloso, M. & Riley, P.

CMUnited-98 Champion Simulator Team, The

1999 RoboCup-98: Robot Soccer World Cup II, pp. 61-76 inproceedings


Abstract: The CMUnited-98 simulator team became the 1998 RoboCup simulator league champion by winning all 8 of its games, outscoring opponents by a total of 66-0. CMUnited-98 builds upon the successful CMUnited-97 implementation, but also improves upon it in many ways. This chapter describes the complete CMUnited-98 software, emphasizing the recent improvements. Coupled with the publicly-available CMUnited-98 source code, it is designed to help other RoboCup and multi-agent systems researchers build upon our success.
This research is sponsored in part by the DARPA/RL Knowledge Based Planning and Scheduling Initiative under grant number F30602-95-1-0018. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies or endorsements, either expressed or implied, of the U. S. Government.
Suzuki, S., Kato, T., Ishizuka, H., Takahashi, Y., Uchibe, E. & Asada, M.

An Application of Vision-Based Learning in RoboCup for a Real Robot with an Omnidirectional Vision System and the Team Description of Osaka University "Trackies

1999 RoboCup-98: Robot Soccer World Cup II, pp. 316-325 inproceedings


Abstract: This paper gives a team description of Osaka University “Trackies” for RoboCup-98, and related research issues. We focus on behavior learning of our goalie robot which has an omnidirectional vision system. A Q-learning method is applied by defining substates from visual information of the ball and the goal. To reduce the learning time, we propose an attention control method for an omnidirectional vision by means of an active zoom mechanism. We perform computer simulation and real robot experiments to show the validity of the proposed method.
Takahashi, T. & Naruse, T.

From Play Recognition to Good Plays Detection

1999 RoboCup-98: Robot Soccer World Cup II, pp. 187-192 inproceedings


Abstract: This paper describes an attempt to review the teams participating in RoboCup simulator leagues. RoboCup simulator games are played through communication between soccer player clients and the soccer server. The server simulates the player’s requirements and sends the result to the soccer monitor. We enjoy the games by seeing the players’ actions and the ball’s movement displayed on monitors. A method is proposed to recognize actions of a player such as shooting, kicking, etc., or the ball’s movement from the log files, which are equivalent to images displayed on the monitor. Action recognition is a necessary technique for scoring, commenting and judging a game. The games in RoboCup’ 97 are reviewed from log files and the analysis results are discussed.
Torterolo, F. & Garbay, C.

A Hybrid Agent Model, Mixing Short Term and Long Term Memory Abilities

1999 RoboCup-98: Robot Soccer World Cup II, pp. 246-260 inproceedings


Abstract: We present in this paper a novel approach for the modeling of agents able to react and reason under highly dynamic environments. A hybrid agent architecture is described, which allows to integrate the capacity to react rapidly to instantaneous changes in the environment with the capacity to reason more thoroughly about perceptions and actions. These capacities are implemented as independent processes running concurrently, and exploiting different memorizing abilities. Only a short-term memory is made available to reactive agents, whilst long-term memorizing abilities together with the possibility to reason about incomplete information is provided to cognitive agents. This model is currently experimented and tested under the framework of the RoboCup competition. An application example is provided to support the discussion.
Uchibe, E., Nakamura, M. & Asada, M.

Cooperative Behavior Acquisition in a Multiple Mobile Robot Environment by Co-Evolution

1999 RoboCup-98: Robot Soccer World Cup II, pp. 273-285 inproceedings


Abstract: Co-evolution has recently been receiving increased attention as a method for multi agent simultaneous learning. This paper discusses how multiple robots can emerge cooperative behaviors through co-evolutionary processes. As an example task, a simplified soccer game with three learning robots is selected and a, GP (genetic programming) method is applied to individual population corresponding to each robot so as to obtain cooperative and competitive behaviors through evolutionary processes. The complexity of the problem can be explained twofold: co-evolution for cooperative behaviors needs exact synchronization of mutual evolutions, and three robot co-evolution requires well-complicated environment setups that may gradually change from simpler to more complicated situations so that they can obtain cooperative and competitive behaviors simultaneously in a wide range of search area in various kinds of aspects. Simulation results are shown, and a discussion is given.
Veloso, M., Bowling, M., Achim, S., Han, K. & Stone, P.

The CMUnited-98 Champion Small-Robot Team

1999 RoboCup-98: Robot Soccer World Cup II, pp. 77-92 inproceedings


Abstract: In this chapter, we present the main research contributions of our champion CMUnited-98 small robot team. The team is a multiagent robotic system with global perception, and distributed cognition and action. We describe the main features of the hardware design of the physical robots, including differential drive, robust mechanical structure, and a kicking device. We briefly review the CMUnited-98 global vision processing algorithm, which is the same as the one used by the previous champion CMUnited-97 team. We introduce our new robot motion algorithm which reactively generates motion control to account for the target point, the desired robot orientation, and obstacle avoidance. Our robots exhibit successful collision-free motion in the highly dynamic robotic soccer environment. At the strategic and decision-making level, we present the role-based behaviors of the CMUnited-98 robotic agents. Team collaboration is remarkably achieved through a new algorithm that allows for team agents to anticipate possible collaboration opportunities. Robots position themselves strategically in open positions that increase passing opportunities. The chapter terminates with a summary of the results of the RoboCup-98 games in which the CMUnited-98 small robot team scored a total of 25 goals and suffered 6 goals in the 5 games that it played.
This research is sponsored in part by the Defense Advanced Research Projects Agency (DARPA) and the Air Force Research Laboratory (AFRL) under agreement numbers F30602-97-2-0250 and F30602-98-2-0135, and in part by the Department of the Navy, Office of Naval Research under contract number N00014-95-1-0591. The views and conclusions contained in this document are those of the authors and should not be interpreted as necessarily representing official policies or endorsements, either expressed or implied, of the Air Force, of the Department of the Navy, Office of Naval Research or the United States Government.
Veloso, M. & Uther, W.

The CMTrio-98 Sony Legged Robot Team

1999 RoboCup-98: Robot Soccer World Cup II, pp. 491-497 inproceedings


Abstract: Sony has provided a remarkable platform for research and development in robotic agents, namely fully autonomous legged robots. In this paper, we describe our work using Sony’s legged robots to participate in the RoboCup’98 legged robot demonstration and competition. The robots are fully autonomous with on-board vision, control, and navigation. The challenges we addressed in this framework include the color calibration of the vision hardware, the landmark-based robot localization on the playing field, and the development of robot behaviors for the actual play of the game. The paper presents our approach and contributions to these issues. We apply machine learning techniques for automated color calibration and we develop effective vision-servoed navigation. We present our Bayesian localization algorithm. Team strategy is achieved through pre-defined behaviors. Our team of the Sony legged robots, CMTrio-98, won all of its games in the RoboCup-98 competition, and was awarded the first place in the championship.
Verner, I.M.

The Survey of RoboCup' 98: Who, How and Why

1999 RoboCup-98: Robot Soccer World Cup II, pp. 109-119 inproceedings


Abstract: The need for educational research of robot competition programs is argued. A questionnaire for RoboCup ′98 team members is presented. The survey data about the participants, their experience in robot competitions, activities, subjects and motivation for participating in the program are reported and discussed.
Voelz, D., André, E., Herzog, G. & Rist, T.

Rocco: A RoboCup Soccer Commentator System

1999 RoboCup-98: Robot Soccer World Cup II, pp. 50-60 inproceedings


Abstract: With the attempt to enable robots to play soccer games, the RoboCup challenge poses a demanding standard problem for AI and intelligent robotics research. The rich domain of robot soccer, however, provides a further option for the investigation of a second class of intelligent systems which are capable of understanding and describing complex time-varying scenes. Such automatic commentator systems offer an interesting research perspective for additional integration of natural language and intelligent multimedia technologies.
In this paper, first results concerning the realization of a fully automated RoboCup commentator will be presented. The system called Rocco is currently able to generate TV-style live reports for arbitrary matches of the RoboCup simulator league. Based upon our generic approach towards multimedia reporting systems, step-by-step even more advanced capabilities are to be added with future versions of the initial Rocco prototype.
Wyeth, G., Browning, B. & Tews, A.

The UQ RoboRoos Small-Size League Team Description for RoboCup'98

1999 RoboCup-98: Robot Soccer World Cup II, pp. 428-433 inproceedings


Abstract: The UQ RoboRoos have been developed to participate in the RoboCup’ 98 robot soccer small size league. This paper describes the current level of implementation of the robots, including aspects of hardware design, as well as the software running on the robots and the controlling computer. Key features of the RoboRoos design include the agile and powerful mechanical frame, the robots’ navigational techniques and a coordinating planner system based on potential field methods.
Yokota, K., Ozaki, K., Watanabe, N., Matsumoto, A., Koyama, D., Ishikawa, T., Kawabata, K., Kaetsu, H. & Asama, H.

UTTORI United: Cooperative Team Play Based on Communication

1999 RoboCup-98: Robot Soccer World Cup II, pp. 479-484 inproceedings


Abstract: In order for multiple robots to accomplish a required mission together, they need to organize themselves, cooperate and share information. We regard such actions as “team play” and believe communication is the essential tool for team plays. This paper discusses communication in the distributed autonomous robotic system and development of cooperative actions for football playing robots. The discussed communication framework and cooperation are implemented in our omni-directional mobile robots which has vision for sensing and a wireless device for communication.
Zhang, Y. & Mackworth, A.K.

A Multi-Level Constraint-Based Controller for the Dynamo98 Robot Soccer Team

1999 RoboCup-98: Robot Soccer World Cup II, pp. 402-409 inproceedings


Abstract: Constraint Nets provide a semantic model for modeling hybrid dynamic systems. Controllers are embedded constraint solvers that solve constraints in real-time. A controller for our new softbot soccer team, UBC Dynamo98, has been modeled in Constraint Nets, and implemented in Java, using the Java Beans architecture. An evolutionary algorithm is designed and implemented to adjust the weights of constraints in the controller. The paper demonstrates that the formal Constraint Net approach is a practical tool for designing and implementing controllers for robots in multi-agent real-time environments.