基于改进A^*算法的无人配送车避障最优路径规划

Optimal Path Planning for Obstacle Avoidance of Unmanned Distribution Vehicles Based on Improved A^* Algorithm

针对无人配送车在特定应用场景下的全局路径规划问题,提出了一种改进的A^*算法.首先,根据无人配送车的运行环境,建立栅格地图,并使用基于八邻域搜索的A^*算法在栅格地图上规划起始点到目标点的路径;其次,在基于八邻域搜索A^*算法的基础上,融入了一种适应无人配送车运行环境的多启发式搜索方法,使改进后的算法能够快速精确地规划出路径距离最优的初始路径;并进一步优化初始路径的转折线段为圆弧,最大限度地平滑路径;最后,改进A^*算法的路径生成机制,使规划生成的路径不会斜线经过障碍物的边界线,保证无人配送车安全有效地避开障碍物,到达目标点.仿真结果表明,该算法缩短了路径规划时间,同时转折线段优化为圆弧,生成的路径与障碍物始终保持至少半个栅格的最小安全距离,该研究对于无人配送车快速实现安全配送具有重要的理论价值和应用意义.

Aiming at the global path planning problem of unmanned delivery vehicles in specific operating scenarios,an improved A^* algorithm is proposed.Firstly,a grid map is established according to the operating environment of the unmanned delivery vehicle,and the A^* algorithm based on the eight-neighborhood search is used to plan the path from the starting point to the target point on the grid map.Secondly,based on the eight-neighborhood search A^* algorithm,a multi-heuristic search method adapted to the operating environment of unmanned delivery vehicles is incorporated so that the improved algorithm can plan the initial path with the optimal path distance quickly and accurately;and the turning line segment of the initial path is further optimized as arc,smoothing the path to the maximum extent.Finally,the path generation mechanism of the A^* algorithm is improved so that the planned path does not obliquely cross the boundary line of the obstacle to ensure that the unmanned delivery vehicle safely and effectively avoids the obstacle and reaches the target point.The simulation results show that the algorithm shortens the path planning time,and the turning line segment is optimized to be an arc.The generated path and obstacles always maintain a minimum safety distance of at least half a grid.The study has important theoretical value and application significance for quick and safe delivery practice of unmanned delivery vehicles..