Path Planning Algorithms For AGV Systems

With the rapid development of the logistics industry,logistics systems have become multi-tasking,process changes are large,dynamic,and ordinary automation equipment is not competent.Automatic guided vehicles(AGV)have been widely used in the logistics industry because of their automation,high flexibility,high efficiency,and high reliability.The normal operation of the AGV system is inseparable from proper path planning,and the quality of the path directly affects the operational efficiency of the AGV.How to design efficient path planning algorithms has always been a hotspot in the research of AGV systems.Nowadays,for the path planning problem of the AGV system,it has been extensively studied.Many path planning algorithms have been proposed.As one of the earliest heuristic algorithms studied,the ant colony algorithm can effectively plan paths in practical applications,however,the ant colony algorithm is easy to fall into local optimization during the calculation process,which reduces the AGV's operating efficiency.In this paper,for the AGV path planning problem under the grid map,an improved ant colony algorithm is used to reduce the probability of the traditional ant colony algorithm falling into the local optimum,and a genetic algorithm is combined to accelerate the algorithm's convergence speed.The main work and results are as follows:(1)In order to solve the ant colony algorithm falling into the local optimal defect and realize the speed of the search path,this article adds an elite ant strategy to the ant colony algorithm,the ant with the best path is selected from each iteration,the elite ant pheromone update strategy is improved,and a faster convergence speed can be obtained.In order to solve the defect that the ant colony algorithm is easy to fall into the local optimum,this paper adds a tabu search algorithm to the ant colony algorithm to do local optimization processing,so that the algorithm jumps out of the local optimum.(2)Aiming at the multi-AGV scheduling system,this paper uses an improved time window algorithm to plan the shortest path for multiple AGVs.Different from the traditional time window algorithm,this paper adds a soft time window.When the AGV passes within the soft time window of a node,it needs to receive a certain penalty.In this way,the conflict problem caused by the early or late arrival of AGV can be effectively avoided.At the same time,this article also uses two different avoidance methods,namely the re-planning method and the waiting method.Different avoidance strategies are selected according to the characteristics of the path,which improves the operating efficiency of the AGV.(3)This paper proposes the concept of node busy index.According to the situation that each node is occupied by AGV,calculate the busy index of all nodes in the map.The higher the node busy index,the more AGVs passing through the node,and the higher the probability of conflicts at the node,based on the node busy index,the algorithm avoids busy nodes,reduces the probability of conflicts caused by the unstable operation of AGV,and enhances the anti-interference of the path.