关键词: 多自动导引小车, 冲突预测, 顶点属性, 改进粒子群优化算法

Abstract: To solve the path conflict and avoidance problem and improve the operational efficiency of multi-AGV system,an optimization method of multi-AGV conflict avoidance decision based on conflict prediction was proposed.A conflict prediction method based on vertex attributes and real-time pose was proposed.Considering the global state of the road network,the mathematical evaluation model of conflict avoidance decision was established,and an Improved Particle Swarm Optimization (IPSO) algorithm for multi-AGV system was proposed.By optimizing the speed and direction of particle motion,the premature convergence of optimization algorithm was avoided.In addition,the mutation operation was introduced into the particle by fusing the mutation idea of Genetic Algorithm (GA) to improve the global search ability of the optimization algorithm.The experimental results showed that the optimization method could solve the multi-AGV system path conflict problem effectively and reduce the total waiting time of AGVs in the process of conflict avoidance.Accordingly,the security and operation efficiency of multi-AGV system were improved.

Key words: multi-AGV, conflict prediction, vertex attributes, improved particle swarm optimization