An intelligence-based hybrid PSO-SA for mobile robot path planning in warehouse

Abstract

Mobile robots play crucial roles in industry and commerce, and automatic guided vehicles (AGV) are one of the primary parts of smart manufactory and intelligent logistics. Path planning is the core task for the AGV system, and it generates the path from origin to destination. The motivation of the study is to improve the scalability, flexibility, adaptability, and performance of the robot path planning systems. We propose the hybrid PSO-SA algorithm for the optimization of AGV path planning. Compared with other heuristic algorithms by benchmark functions, including HS, FA, ABC and GA, the proposed algorithm shows excellent performance in dealing with optimization problems. It reduces the possibility of getting trapped in one local optimum and enhances the efficiency to get the best global solution with faster convergence and less time consumption. It is evaluated with multiple cost functions and path planning with simulations and experiments. The objective of the proposed algorithm is to minimize the path length and produce a smooth path without collision. The proposed PSO-SA algorithm is compared with PSO in the path planning application, and the mean runtime and iteration times are usually significantly lower than PSO.