A Novel Approach based on Evolutionary Game Theoretic model for Multi Player Pursuit Evasion
In this paper, we consider a novel Three Hierarchical decomposition approach for Multi-Player Pursuit Evaders (MPPE) game. In multi-player pursuit evasion game, hierarchical framework is applied widely in order to decompose the original complicated multiplayer game into multiple small scale games. In this paper, we first study the number of pursuers which necessitates; the capture condition and the time of all evaders have been captured. Then, describe the Distributed Task Assignment Stage Based on dynamic Coalition Formation. Last, a novel multi-agent Qlearning approach based on Evolutionary Game Theoretic model is used for pursue. Experimental results obtained on two different environments of a well-known pursuit domain show the effectiveness and robustness of the proposed Hierarchical architecture and learning approach.
Pursuit Evasion Q-learning, Evolutionary Game Theory Coalition Formation
Renping Liu Cai ze-su
College of Information Engineering, North China University of Technology, Beijing, 100144, China State Key Laboratory Robotics and System(HIT),Harbin Institute of Technology, Harbin 150001, China
国际会议
长春
英文
107-110
2010-08-24（万方平台首次上网日期，不代表论文的发表时间）