A pursuit-evasion problem-solving strategy based on probability estimation in a planar region (Q391730)

Summary: The pursuit-evasion problem in planar regions is presented, and a strategy based on probability estimation is proposed. The instant evasion routines of the evader are abstracted and described by linear and nonlinear probabilistic models. The pursuer computes the capture probabilities of the instant evasion trajectories dynamically to achieve the behaviour process and moving strategy. The proposed strategy takes advantages of evolutionary ideas and estimation of probabilistic distribution approach, and the efficiency and convergence of the strategy in this paper are verified. The simulation experiments show that the proposed approach has better performances on the success rates of capture and the capture time, and the experimental results show that the pursuers can capture the evaders successfully in limited time.