Probability map partitioning for multi-player pursuit-evasion game

In this paper, we consider a probabilistic pursuit-evasion game, in which a team of pursuers try to capture multiple evaders in a 2-D environment. We apply map partition and assignment algorithm to overcome the concentration of agents caused by global-max pursuit policy. To this effect, k-Means clustering is employed. From the simulation results, we analyze the characteristics of the proposed algorithm.