基于多线激光雷达的无人车路径规划算法

Unmanned Vehicle Path Planning Algorithm Based on Multi-Line Lidar

针对在非结构路面条件下无人车路径规划存在建模难度大,计算量较大,算法运行的实时性较差等问题,提出一种基于多线激光雷达的无人车路径规划算法。对原始点云数据进行预处理,减少需要处理的点云数量,降低程序运行复杂度;将三维点云降维形成较低分辨率的二维栅格地图,降低路径规划程序的计算量;基于构建的局部栅格地图,使用改进的人工势场法完成路径规划。算法已在无人车上应用,实验证明了该方法的有效性和可行性。

Aiming at the problems of difficult modeling, large amount of calculation and poor practicality of the algorithm in path planning of unmanned vehicle on unstructured pavement,an unmanned vehicle path planning algorithm based on multi-line lidar was proposed. The original point cloud data were preprocessed to reduce the number of point clouds needed to be processed and the complexity of the program. The three-dimensional point cloud was reduced to form a two-dimensional raster map with lower resolution, which reduced the computational complexity of the path planning program. Based on the local raster map,the improved artificial potential field method was used to complete the path planning. The algorithm has been applied in unmanned vehicle,and the experiment proves the validity and feasibility of the method.