Discovering Behavior Patterns in Muti-agent Teams
Three of the most important open research areas in multi-agent cooperative systems are the construction of models related with the communication between agents, the multiple interactions and the behaviors adopted by agents during a task  . This work deals with the discovery of behaviors in multi-agent teams, more precisely in soccer teams. It is an extension of a previous work presented in  . The extension is focused on the discovery of tactical plays adopted by soccer-agents during a match within the context of formations. Due to the nature of team work in soccer-agent domains, the discovery of tactical behaviors should be done within the context of team formations. Nevertheless, the dynamic nature and the multiple interactions between players at each instant of the game difficult the tracking of formations, which at the same time difficult the discovery of tactical plays. In this work is proposed an original and efficient way of discovering tactical plays supported by a robust tracking of formations, even though they are submitted to dynamic changes of the world, based on the construction of topological structures. Successful results, derived from the analysis of an important number of matches of the RoboCup Simulation league matches, valid the efficiency of the model presented in this work.
KeywordsSoccer-agents behavior pattern recognition tactics
Unable to display preview. Download preview PDF.
- 2.Berge, C.: Graphes. Guathier-Villars (1983)Google Scholar
- 4.Devaney, M., Ram, A.: Needles in a haystack: Plan recognition in large spatial domains involving multiple agents. In: AAAI/IAAI, pp. 942–947 (1998)Google Scholar
- 5.Ferber, J.: Multi-Agent System: An Introduction to Distributed Artificial Intelligence. Addison-Wesley Longman, Amsterdam (1999)Google Scholar
- 6.Freeman, H.: On the encoding of arbitrary geometric con?gurations. In: IRE Transactions, editor (1973)Google Scholar
- 7.Kaminka, G.A., Fidanboylu, M., Chang, A., Veloso, M.M.: Learning the sequential coordinated behavior of teams from observations. In: Kaminka, G.A., Lima, P.U., Rojas, R. (eds.) RoboCup 2002. LNCS (LNAI), vol. 2752, pp. 111–125. Springer, Heidelberg (2003)Google Scholar
- 8.Kitano, H., Tambe, M., Stone, P., Veloso, M.M., Coradeschi, S., Osawa, E., Matsubara, H., Noda, I., Asada, M.: The robocup synthetic agent challenge 97. In: Kitano, H. (ed.) RoboCup 1997. LNCS, vol. 1395, pp. 62–73. Springer, Heidelberg (1998)Google Scholar
- 9.Knuth, D.E.: The Art of Computer Programming. Addison-Wesley, Reading (1973)Google Scholar
- 10.Kuhlmann, G., Stone, P., Lallinger, J.: The ut austin villa 2003 champion simulator coach: A machine learning approach. In: Nardi, D., Riedmiller, M., Sammut, C., Santos-Victor, J. (eds.) RoboCup 2004. LNCS (LNAI), vol. 3276, pp. 636–644. Springer, Heidelberg (2005)Google Scholar
- 11.Lattner, A.D., Miene, A., Visser, U., Herzog, O.: Sequential pattern mining for situation and behavior prediction in simulated robotic soccer. In: Bredenfeld, A., Jacoff, A., Noda, I., Takahashi, Y. (eds.) RoboCup 2005. LNCS (LNAI), vol. 4020, pp. 118–129. Springer, Heidelberg (2006)CrossRefGoogle Scholar