融合改进A^(*)算法和动态窗口法的自动驾驶路径规划

Fusing improved A^(*)algorithm and dynamic window approach for autonomous driving path planning

针对自动驾驶汽车路径规划全局最优、耗时最优和避障的需求,提出一种改进A^(*)算法和动态窗口法的融合算法。A^(*)算法主要从启发函数、权重系数、搜索邻域和搜索策略4个方面进行改进,动态窗口法主要改进评价函数。利用改进后的A^(*)算法和双向A^(*)算法完成栅格地图上的全局路径规划,去除冗余节点并平滑处理优化全局路径,利用融合动态窗口算法进行局部路径规划,完成避障。与传统的A^(*)算法相比,改进的A^(*)算法和双向A^(*)算法搜索全局路径耗时和节点显著减少,优化的A^(*)算法与动态窗口法的融合算法具有更高的效率、更好的路径规划能力和避障能力。

This paper proposes a fusion algorithm that combines the improved A^(*)algorithm and dynamic window approach to address the needs of global optimal,time optimal,and obstacle avoidance in autonomous driving path planning.The A^(*)algorithm is improved from four aspects:heuristic function,weight coefficient,search neighborhood,and search strategy while the dynamic window approach mainly improves the evaluation function.The improved A^(*)algorithm and bidirectional A^(*)algorithm are employed to complete global path planning on the grid map.Then,redundant nodes are removed,and the optimized global path is smoothed.Finally,the dynamic window approach is integrated to perform local path planning and obstacle avoidance.Compared with the traditional A^(*)algorithm,the improved A^(*)algorithm and bidirectional A^(*)algorithm significantly reduce the time consumption and nodes in searching for global paths.The fusion algorithm of the optimized A^(*)algorithm and dynamic window approach achieves a higher efficiency and an improved ability in path planning and obstacle avoidance.