Research On Multi-AGV Path Planning Problem Based On A* Algorithm

With the integration and development of informatization and industrialization,in the Flexible Manufacturing System(FMS),which has the characteristics of a wide variety of products and small production batches,the main factors affecting production efficiency is the time of material transportation.Automated Guided Vehicle(AGV),which has many advantages such as high flexibility,has become the main transportation tool of FMS.In order to solve the problem of multi-AGV path coordination,this paper carried out the following research:Aiming at the problem of solving the minimum total default time in multi-AGV path planning,this paper proposes a multi-AGV path planning method based on the A* algorithm.Since the AGV will default when performing tasks beyond the agreed time,first of all,the time and path length are comprehensively considered when calculating the actual cost of the expansion node,and then the AGV that plans the path of the next step is determined by comparing the maximum node schedule of each AGV,finally,the path of the AGV from the starting position to the position of the expansion node is stored in the OPEN table,and the OPEN table is used to simplify the path.This paper considers the situations where multiple AGVs plan paths from different starting positions and plan paths from the same starting position.The simulation results show that compared with other methods,the total default time,total time and total path length of multiple AGVs are reduced.Aiming at the problem of reducing the total default time for multiple AGVs after repeated planning of the path,this paper proposes a multi-AGV path planning method that combines game theory and A* algorithm.When multiple AGVs plan paths from different starting positions,the multi-AGV path planning method based on the A* algorithm does not consider repeated planning of the path,this paper introduces game theory.In the process of repeating the game,the probability of planning the path of the next step of each AGV is adjusted according to the result of each game.The simulation results show that the multi-AGV path planning method that combines game theory and A* algorithm reduces the total default time,total time and total path length of multiple AGVs.Aiming at the problem of expanding the search range of the multi-AGV path planning method based on the A* algorithm when simplifying the path,this paper proposes a multi-AGV path planning method based on public and private OPEN tables.When multiple AGVs plan paths from the same starting position,this paper divides the OPEN table into a public OPEN table and a private OPEN table.The public OPEN table stores the expansion nodes and paths of all AGVs.The AGV uses the public OPEN table to simplify the path,which expands the search range.The simulation results show that the multi-AGV path planning method based on public and private OPEN tables reduces the total default time,and total time of multiple AGVs.