基于激光雷达的同时定位与地图构建方法综述 postprint

Abstract: Compared with visual based simultaneous localization and mapping (SLAM) technique, due to the robustness of localization, laser based SLAM become a research hotspot at home and abroad. The purpose is positioning itself pose in the unknown environment and build environment map at the same time, there are many applications such as autopilot, indoor and outdoor robot navigation, and 3D reconstruction. In this paper, firstly, it decomposed the basic framework of SLAM based on laser and summarized the mainstream algorithms of four modules from the angle of front-end scan match, back-end optimization, the closed loop detection and validation, and map generation; Then it analyzed and compared the representation scheme under two framework of SLAM, one is based on filter and the other is based on graph optimization; Finally, the paper point out the development trend of laser based SLAM.