Improved PRM Path Planning in Narrow Passages Based on PSO

Abstract

Probabilistic roadmap (PRM) has been successfully applied in mobile robot path planning. However, the task of path planning is still a challenging job when narrow passages exist in the configuration space. In this paper, we propose a new approach that combines the technique of particle swarm optimization (PSO) with PRM. By sharing the information of free space with the sampling points which were originally deployed in the area obstacles, these sampling points can explore the free area in the later move and thus increase connectivity of the undirected graph without increasing the total sampling time. Based on the simulation results, the proposed method improves both the utilization rate of sampling points and the success rate of narrow passages path planning.