Cooperative Algorithm for Multi-agent Foraging Task Based on Modified Hawk-Dove Game

Multi-agent foraging is a popular benchmark to verify the effectiveness of different cooperation algorithm. Markov game based approaches were wildly used although they could not select consistent equilibrium for the group. Using evolutionarily stable strategy as optimal solution, we build a modified hawk-dove game model to simulate the interaction between agents, and then proposed a ponder-replicator algorithm to find certain consistent maximal reward equilibrium for the group. The simulation verified the efficiency of the proposed algorithm.

[1]  J M Smith,et al.  Evolution and the theory of games , 1976 .

[2]  M. Niazi,et al.  Multi agent foraging - taking a step further Q-leaming with search , 2005, Proceedings of the IEEE Symposium on Emerging Technologies, 2005..

[3]  Xiaofeng Wang,et al.  Reinforcement Learning to Play an Optimal Nash Equilibrium in Team Markov Games , 2002, NIPS.

[4]  Andrew B. Williams,et al.  Lessons learned in single-agent and multiagent learning with robot foraging , 2003, SMC'03 Conference Proceedings. 2003 IEEE International Conference on Systems, Man and Cybernetics. Conference Theme - System Security and Assurance (Cat. No.03CH37483).

[5]  E. Akin,et al.  Dynamics of games and genes: Discrete versus continuous time , 1983 .

[6]  Per Svejstrup Hansen,et al.  Equilibrium selection in coordination games with simultaneous play , 2002 .

[7]  John C. Harsanyi,et al.  Общая теория выбора равновесия в играх / A General Theory of Equilibrium Selection in Games , 1989 .