Path Planning Method for Mobile Robot Based on Multiple Improved PSO

Abstract

The problem of easy to fall into the local optimum for the traditional particle swarm optimization (PSO) in robot path planning has gained attention constantly. For solving the problems such as discrete and optimization problems, the path is not smooth, etc., multiple improved PSO algorithm is introduced in this paper. The improved algorithm adapts to the adjustment, optimizes the inertia weight coefficients, and learns factors, thus increase the optimal value of the path. It constructs a penalty function and proposes constraint indexes that can be optimized for evaluation, the robot can avoid obstacles autonomously, and get a smooth path by the cubic spline. The improved algorithm was verified by experiments. The experimental results show that: the optimal value of the path length obtained by the improved algorithm is 18.2% less than that of the classic PSO. Comparing with other improved algorithms mentioned in this paper, the path length is reduced by 6.2%, but it saves at least 5.04 seconds in convergence speed.