Soldiers’ Group Behaviors Simulation Based on Improved MAPRM

Soldiers’ group behaviors are a class of flocking behaviors in which a squad of soldiers march forward according to a specified formation and try to keep it when encountering narrow corridors. It has a leading application in deterring the behavior and activities of a potentially hostile crowd and bringing a mob engaged in a riot under control in social stable maintenance. In this paper, we focus on investigating soldier’s group behaviors based on an improved MAPRM algorithm. Firstly, we use MAPRM algorithm to sample configurations from the medial axis of the simulation space. Secondly, in the construction and query phase of MAPRM, clearance information of the space is introduced to local planner and used as heuristic information to A* algorithm which improves the MAPRM algorithm. Thirdly, when the soldiers pass through narrow corridors, their formation transitions are achieved by sampling the desired formation shape. The simulation results show that our approach is effective and feasible.