Improved A*-DWA fusion path planning algorithm with ideal path area constraints

摘要

Abstract Navigation of mobile robots in unknown environments continues to be a common and challenging problem in robotics. In order to improve the mobile robot to obtain the global optimal path and real-time obstacle avoidance capability, this paper proposes a new fusion algorithm based on A* algorithm and DWA (Dynamic Window Approach) algorithm. First, compared with the classical Dijkstra algorithm and BFS(Best First Search) algorithm, the A* algorithm is optimized in path search. Although adding heuristic functions to the A* algorithm can reduce a large number of useless search nodes, experiments show that there are still many useless search nodes, and the obtained global path is not a feasible path in some cases or will increase the movement time significantly, and unsafe paths. In this paper, we add the ideal path area constraint to the A* algorithm, and then integrate the DWA algorithm to address the problem that the A* algorithm is not suitable for real-time obstacle avoidance in dynamic obstacle environments. Simulation results show that the improved A* algorithm can significantly reduce the number of nodes to be searched and decrease the running time of the DWA algorithm compared with the traditional Dijkstra, BFS, and A* algorithms.