基于三维点云的变电站无人机巡检路径规划

Substation-inspection Path Planning for Quadrotors Based on 3D Point Cloud

面向变电站三维巡检应用需求,基于变电站的三维点云数据信息,开展无人机巡检路径规划方法研究。首先,根据三维激光雷达获取的点云数据,考虑环境障碍和安全距离,通过坐标映射和障碍膨胀,构建变电站三维空间的栅格地图。然后,以路径最短为目标、以障碍规避为约束,在栅格地图上运用改进A*搜索算法进行路径规划,得到一条可行的无人机巡检路径。改进A*搜索算法通过定制设计节点扩展策略和代价函数计算方法,减少节点搜索数量,提高路径规划效率。仿真试验结果表明,所提方法能够建立有效的环境地图、生成安全可行的巡检路径,改进A*算法能够以更短的时间生成最优性相当的路径结果。

To satisfy requirements of 3D inspection for power substations,the inspection path planning for quadrotors is conducted based on 3D point cloud of substations.First,based on the point cloud obtained by the 3D lidar,the 3D grid map is constructed through coordinates mapping and obstacles expansion considering obstacle avoidance and safe distance.Then,taking the path minimization as the goal and obstacle avoidance as the constraint,the modified A*is employed for path planning on the grid map to obtain a feasible inspection path for quadrotors.The modified A*could reduce the number of search nodes and improve the planning efficiency by tailoring the node-expansion strategy and cost computation method.The numerical experiments show that the proposed method can build an effective environment map and generate a safe inspection path,and the modified A*can generate comparable results on optimality with much less computation time.