In view of the fact that the path planned by the traditional A* algorithm in the complex grid map is not smooth,there are redundant turning points and redundant collinear nodes,and when the traditional octree search strategy is used,it is easy to cause the AGV to collide with obstacles.Facing complex environments obstacles randomly appearing on the global path cannot achieve dynamic obstacle avoidance and other problems,an improved A* algorithm was proposed.Since the search efficiency of the traditional A* algorithm mainly depended on the design of the evaluation function,the weight coefficient of the heuristic function was introduced to improve the search efficiency of the A* algorithm.The obstacle safety distance was set as the judgment obstacle,which provided a reference for whether the current obstacles in the object area affected AGV passage,the original octree search strategy was improved again to improve the obstacle avoidance performance,and then path optimization was performed on the obtained collision-free path to retain key turning points,and finally the fusion of A* and DWA algorithms was realized,further the path was optimized,and global dynamic path planning was realized.The experimental results show that the fusion algorithm makes the path smoother,improves the obstacle avoidance performance of the algorithm,and shows the feasibility of the fusion algorithm in the robot path planning.