Advertisement

Analysis of Agents’ Cooperation in RoboCupRescue Simulation

  • Kazunori Iwata
  • Nobuhiro Ito
  • Kuniyoshi Toda
  • Naohiro Ishii
Part of the Studies in Computational Intelligence book series (SCI, volume 149)

Abstract

In this paper, we present a method to evaluate agents’ cooperation in a Multi-Agent System (MAS). In the MAS research area, a MAS can be evaluated like according to total points, games won, targets reached and so on. The results do not, however, give an evaluation of the agents’ cooperation due to the difficulty of investigating and evaluating the role of each agent, but instead give an total evaluation of the entire MAS. Against this background, we focus on the RoboCupRescue Simulation. The RoboCupRescue Simulation is used as the testbed environment and simulates, on a network of computers, a great earthquake and various kinds of disaster-relief activities by multi-agents in a virtual city. The evaluation of the MAS in this case is given by a “score”, which is defined to reflect the disaster damage and does not give an evaluation of the agents’ cooperation. Hence, we investigate a new indicator that can evaluate the agents’ cooperation in the system. We consider and define 5 kinds of agents’ cooperation dependent on Joint Intention Theory, Joint Responsibility Theory, the COM-MTDP Model and Coordination Theory. We also define cooperation in the RoboCupRescue Simulation by separating the definition into 10 definitions. Finally, we analyze the results of the elimination round in the RoboCupRescue Simulation League 2006 and consider the results of this analysis.

Keywords

Multi-Agent Systems Cooperation RoboCupRescue 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Robocup rescue home, http://www.robocuprescue.org/
  2. 2.
    Cohen, P.R., Levesque, H.J.: Teamwork. Nous 25, 487–521 (1991)Google Scholar
  3. 3.
    Jenings, N.R.: Controlling cooperative problem solving in industrial multi-agent systems using joint intentions. Artificial Intelligence 75(2), 87–240 (1995)CrossRefGoogle Scholar
  4. 4.
    Malone, T.W., Crowston, K.: The interdisciplinary study of coordination. ACM 26(1), 87–119 (1994)Google Scholar
  5. 5.
    Morimoto, T.: edited by RoboCupRescue Technical Commitee: How to develop a robocuprescue agent, http://ne.cs.uec.ac.jp/~morimoto/rescue/manual/index.html
  6. 6.
    Pynadath, D.V., Tambe, M.: Multiagent teamwork: Analyzing the optimality and complexity key theories and models. In: AAMS, pp. 873–880 (2002)Google Scholar
  7. 7.
    Tadokoro, S., Kitano, H.: RoboCup-Rescue Technical Committee, The RoboCup Federation. In: RoboCup Japanese National Comittee (eds.) RoboCup Rescue (in Japanese), Kyoritsu Shuppan Co. Lt. (2000)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Kazunori Iwata
    • 1
  • Nobuhiro Ito
    • 2
  • Kuniyoshi Toda
    • 3
  • Naohiro Ishii
    • 2
  1. 1.Dept. of Business AdministrationAichi UniversityJapan
  2. 2.Dept. of Applied Information ScienceAichi Institute of TechnologyToyotaJapan
  3. 3.Dept. of Compute Science and EngineeringNagoya Institute of TechnologyNagoyaJapan

Personalised recommendations