一种多线激光雷达室外小范围导航算法设计

Outdoor Navigation Research Based on Multi-Line Lidar

对于在室外小范围(如公园、工业园区)运行的移动机器人,为控制成本,在不借助其他定位装置仅依靠激光雷达的情况下进行定位导航,单线激光雷达无法在未知的环境下对实际环境进行真实建图与定位,建立出的栅格地图无法描述真实的外部环境;适用于室外的多线激光雷达发出的点云数据虽可描述出复杂的室外环境,但其建立的点云地图无法直接用于导航。故本文提出一种借助多线激光雷达基于八叉树地图的室外导航算法,将点云地图压缩成八叉树地图,进一步转化成栅格地图,并提取栅格地图信息,为ROS中导航包提供相应的必要信息,从而实现室外导航避障功能。实验表明该算法可成功将周围环境以栅格地图形式进行描述,并发送出激光点信息完成室外导航。

For mobile robots operating in small outdoor areas(e.g.,park,industrial park),in order to control costs,positioning and navigation without relying on other positioning devices relying solely on laser radar,single-line lidar cannot be used in an unknown environment.Real mapping and positioning,the established raster map can not describe the real external environment;point cloud data issued by multi-line laser radar for outdoor,although it can describe the complex outdoor environment,but its established point cloud map Cannot be used directly for navigation.Therefore,this paper proposes an outdoor navigation algorithm based on octree map with multi-line lidar,which compresses the point cloud map into an octree map,further transforms it into a grid map,and extracts the raster map information as a navigation package in ROS.Provide necessary information to achieve outdoor navigation obstacle avoidance.Experiments show that the algorithm can successfully describe the surrounding environment in the form of a grid map and send out the laser point information to complete the outdoor navigation.