Online path planner for mobile robots using particle swarm optimization

Abstract

An algorithm to find a collision free path, for an autonomous mobile robot, is proposed in this paper. The path between the starting point and the destination point, a priori established, is calculated for two methods: the local and the global path planner. The computation of this path, for both methods, is based on an evolutionary approach: Particle Swarm Optimisation (PSO). This approach uses a new fitness function for both path planning methods. The result of this algorithm would be a path planning control of the mobile robot in an environment with both fixed and mobile obstacles. The computational efficiency of this online path planning algorithm using PSO with the proposed fitness function is proved in simulation and real time implementation. Real time experiments using a mobile robot validate the local path planning method of the proposed algorithm. The mobile robot used in real time experiments is equipped with a laser scanner module to communicate accurate information about the fixed and mobile surroundings to the robot.