Switching pursuit strategies for multi-pursuer single-evader games

Abstract

This paper introduces a new family of pursuit strategies for multi-pursuer single-evader games in a planar environment. The main idea is to exploit conditions under which capture of the evader in minimum time can be achieved by only two pursuers. The first contribution is to characterize such conditions in terms of the agent positions. Then, new pursuit strategies are proposed in which the multi-pursuer team aims to meet such conditions, switching to a two-pursuer game once they are satisfied. The benefit of this approach is twofold. First, it is shown that naive strategies that are in general unsuccessful can be turned into winning strategies by switching to the appropriate two-pursuer game. Second, the switching mechanism significantly enhances the performance of existing pursuit algorithms, like those based on Voronoi partitions. This is demonstrated by means of extensive numerical simulations.