Research on Loading and Unloading Path Optimization for AGV at Automatic Container Terminal Based on Improved Particle Swarm Algorithm

Abstract

This article investigates the optimal path problems of Automatic guided vehicle (AGV) during the loading and unloading process of the automated container terminal. The problem under consideration has been solved by employing an improved version of particle swarm optimization (IPSO) with crossover and mutation. In this paper, the map model of AGV is established by grid method, and the fitness function mathematical model is established according to the actual situation of the automated container terminal. Meanwhile, the particles are crossed over and mutated by drawing on the idea of genetic algorithm (GA) crossover and mutation, which has increased the diversity of the population and the ability of jumping out of the local area. Compared to particle swarm (PSO) algorithm, the analysis shows that the IPSO has some improvements in reducing iteration times and increasing convergence speed. The IPSO algorithm shortened the driving distance of AGV by 1.13m. The feasibility of the IPSO algorithm has been verified through MATLAB.