一种融合IMU的手持旋轴激光雷达定位建图方法  

A Handheld Rotary LiDAR SLAM Method Integrated with IMU

作　　者：赵宇轩 贾克斌 陈嘉平 ZHAO Yuxuan;JIA Kebin;CHEN Jiaping(Faculty of Information,Beijing University of Technology,Beijing 100124,China;Beijing Laboratory of Advanced Information Network,Beijing 100124,China)

机构地区：[1]北京工业大学信息学部,北京100124 [2]先进信息网络北京实验室,北京100124

出　　处：《测控技术》2023年第10期38-43,81,共7页Measurement & Control Technology

基　　金：北京市自然科学基金项目(4212001)。

摘　　要：同步定位与建图(Simultaneous Localization and Mapping,SLAM)是移动机器人与智能车辆实现环境感知、实时定位的关键技术。随着科技的不断发展,具有便携、扫描范围广等优势的手持旋轴激光雷达应用愈加广泛。为了解决手持旋轴激光雷达在运行过程中由于旋转引起的特征点稀疏导致定位建图质量差的问题,提出一种改进的SLAM算法。在LIO-SAM算法的基础上计算了一种表征点云强度信息的特征点,在提取几何特征边缘点与平面点时,将周围强度值变化较大的点作为一种新的特征点引入点云匹配。同时,通过实时判断系统是否存在退化风险,从而动态调整滤波数值设置,保证了系统运行的稳定性。实验结果表明,在手持旋轴激光雷达采集的户外场景数据下,经过改进的SLAM算法相比LIO-SAM算法有更好的定位与建图效果。Simultaneous localization and mapping(SLAM)is a key technology for mobile robots and intelligent vehicles to realize environment awareness and real-time localization.With the continuous development of technology,handheld rotary LiDAR,which has the advantages of portability and wide scanning range,is more and more widely used.To solve the problem of sparse feature points caused by rotation during the operation of handheld rotary LiDAR,which leads to poor quality of localization and mapping,an improved SLAM algorithm is proposed.Based on the LIO-SAM algorithm,a feature point characterizing the intensity information of the point cloud is calculated,and when extracting the geometric feature edge points and plane points,the points with large changes in surrounding intensity values are introduced as a new feature cloud for point cloud matching.At the same time,the system operation stability is ensured by dynamically adjusting the filtering parameter settings by judging whether the system has degradation risk in real-time.The experimental results show that the improved SLAM algorithm has better localization and map-building effects than the LIO-SAM algorithm under the outdoor scene data collected by handheld rotary LiDAR.

关 键 词：同步定位与建图 激光雷达 惯性测量单元 特征提取 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]