基于融合地图的自动驾驶激光雷达定位算法  

Autonomous driving LiDAR localization algorithm based on fusion map

作　　者：黄伟杰[1] 张希[1] 朱旺旺 赵柏暄 陈浩[1] HUANG Weijie;ZHANG Xi;ZHU Wangwang;ZHAO Baixuan;CHEN Hao(National Engineering Laboratory for Automotive Electronic Control Technology,Shanghai Jiao Tong University,Shanghai 200240,China)

机构地区：[1]上海交通大学汽车电子控制技术国家工程实验室,上海200240

出　　处：《传感器与微系统》2023年第5期139-142,共4页Transducer and Microsystem Technologies

基　　金：政府间国际科技创新合作重点专项资助项目(2019YFE0100200)。

摘　　要：针对环境变更对自动驾驶激光雷达(LiDAR)定位产生不可预估影响的问题,提出一种基于融合地图的LiDAR定位方法。在线定位中,根据里程计的结果将激光点云融入到离线地图中,使用融合后的地图有效地解决在线定位跳变。定位过程中,根据位姿滑窗维护和离线地图格式一致的在线地图,通过求解在线地图和离线地图间的差异度,自动识别环境更新区域,协助离线地图更新。实验结果表明:算法可以有效降低环境变更导致的定位跳变,在环境出现变更时,可以及时地识别出环境的更新区域。Aiming at the problem that environmental changes will have unpredictable effects on autonomous driving LiDAR localization,a LiDAR localization method based on fusion map is proposed.In online localization,the laser point cloud is fused into the offline map according to the result of odometer,and the fused map is used to effectively solve the online positioning jump.During the localization,the online map that is consistent with the offline map format is maintained according to the pose sliding window,and the environment update area is automatically identified by solving the difference between the online map and the offline map to assist in the offline map update.Experimental results show that the algorithm can effectively reduce the location jump caused by environmental changes.When the environment changes,the algorithm can identify the update area of the environment in time.

关 键 词：自动驾驶 激光雷达定位 里程计 融合地图 环境变更检测 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]