Mobile robot path planning in environments cluttered with non-convex obstacles using particle swarm optimization

Abstract

Generally workspaces of mobile robots are cluttered with obstacles of different sizes and shapes. Majority of the path planning algorithms get stuck in non-convex obstacles pertaining to local minima. Particle Swarm Optimization (PSO) is by comparison simple and readily intelligible yet a very powerful optimization technique which makes it an apt choice for path finding problems in complex environments. This paper presents a particle swarm optimization based path planning algorithm developed for finding a shortest collision-free path for a mobile robot in an environment strewed with non-convex obstacles. The proposed method uses random sampling and finds the optimal path while avoiding non-convex obstacles without exhaustive search. Detailed simulation results show the functionality and effectiveness of the proposed algorithm in different scenarios.