AGV Path Planning in Ammunition Cavern Based on Improved A* Algorithm and DWA
Junlin Zeng, Bin Xiao, F. Liu, Zhijie Guo · 2024
This paper studies the environmental modeling method of the narrow ammunition cavern based on the geometric method. By fusing optimal A* algorithm and dynamic window approach(DWA), A path planning algorithm for ammunition cavern is designed and the path planning efficiency and dynamic obstacle avoidance capability of AGV are improved. Firstly, the storage space of the cavern is constructed with rectangular polygons and rings. And the position relationship between AGV and the obstacles can be solved by vertex coordinates for later path analysis and solution. Secondly, the traditional A* algorithm is improved by adding search neighborhoods, optimizing the evaluation function and deleting redundant nodes strategy, so as to improve the global path planning efficiency of AGV. Then, the obstacle avoidance ability of AGV in the narrow passage is improved by optimizing the evaluation function of DWA. Finally, by combining the advantages of the two optimal algorithms, the proposed algorithm can achieve both global path optimality and dynamic obstacle avoidance capability. The simulation results show that compared with the traditional A* algorithm, the inflection point of the optimal A* algorithm is reduced by 95.4%, the planning time is reduced by 79%, and the path is shortened by 4.4%. The proposed algorithm has the advantages of high fitness and strong security in dynamic environment, and can meet the requirements of real-time AGV path planning in ammunition cavern.