Abstract
In this paper, a real-time adaptive learning system from observation by teammates’ behaviors and results from RoboCup soccer agents is proposed. The agent is required to adapt to its opponents in real time in a RoboCup simulated soccer game.
By only learning to adapt from its own behavior and results, the agent would limit its chances of learning during a game. Therefore, the proposed adapted learning system covers self and teammates’ behaviors and results, which improves and increases its learning chances and learning ability about other opponents.
The agent tries to adapt to recognize possible scoring situations by learning and responding to opponents’ intercept ability according to a learnt parameter. Compared to other systems (a “Non-Learning” system and a “Self-Behavior Learning” system), with the proposed system the score rate was improved from 0.04 (the non learning system) and 0.06 (“Self-Behavior Learning”) to 0.10.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
H. Kitano, M. Asada, Y. Kuniyoshi, I. Noda, E. Ozawa.(1995) RoboCup: The Robot World Cup Initiative, Proceedings of IJCAI-95 Workshop on Entertainment and AI/Alife.
RoboCup official site, http://www.robocup.org/
I. Noda, H. Matsubara, K. Hiraki, I. Frank.(1998) Soccer server: A tool for research on multi-agent systems. Journal of Applied Artificial Intelligence, 12, 233–250.
Soccer Server web site, http://sserver.sourceforge.net/
P. Stone, M. Veloso.(1998) Layered approach to learning client behaviors in the robocup soccer server. Journal of Applied Artificial Intelligence, 12, 165–188.
M. Riedmiller, A. Merke, D. Meier, A. Hofmann, A. Sinner, O. Thate, and Ch. Kill. (2001). Kerlsruhe Brainstormers 2000. RoboCup2000:RobotSoccer WorldCup IV, Springer-Verlag, Berlin.
R. S. Sutton, A. G. Barto. (1998) Reinforcement Learning. The MIT Press.
T. Andou. (1998) Refinement of Soccer Agents’ Positions Using Reinforcement Learning. RoboCup-97:RobotSoccer World Cup 1, 371–388.
K. Kostiadis, H. Hu (1999) Reinforcement learning and co-operation in a simulated multi-agent system. Proceedings of 1999 IEEE/RSJ Int. International Conference on Intel. Robots and Systems.
H. Akiyama, T. Nagao(2001)An Optimization Method of Soccer Player’s Behavior using Genetic Programming, Proceeding of JSAI SIG-Challenge.
Y. Kumada, K. Ueda. (2001) Acquisition of Cooperative Tactics by Soccer Agents with Ability of Prediction and learning. Journal of Japanese Society for Artificial Intelligence, 16, 120–127.
J. Rasmussen. (1983) Skills, rules, and knowledge; signals, signs, and symbols, and other distinctions in human performance models. Journal of IEEE Trans. SMC, 13, 257–266
A. Bandura. (1971)Psychological modeling: Conflicting theories. Aldine Atherton, Chicago.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer-Verlag Tokyo
About this paper
Cite this paper
Kawarabayashi, T., Kubo, T., Morisita, T., Nishino, J., Odaka, T., Ogura, H. (2002). Real-time Adaptive Learning from Observation for RoboCup Soccer Agents. In: Asama, H., Arai, T., Fukuda, T., Hasegawa, T. (eds) Distributed Autonomous Robotic Systems 5. Springer, Tokyo. https://doi.org/10.1007/978-4-431-65941-9_21
Download citation
DOI: https://doi.org/10.1007/978-4-431-65941-9_21
Publisher Name: Springer, Tokyo
Print ISBN: 978-4-431-65943-3
Online ISBN: 978-4-431-65941-9
eBook Packages: Springer Book Archive