摘要
为准确、高效地获取多个激光雷达配准结果,提出一种基于坐标转换的多激光雷达配准算法。将全球定位系统(global positioning system,GPS)数据转换到地心空间直角坐标系下表示,求解基于空间直角坐标系的激光雷达数据与地心空间直角坐标系数据间的映射关系,根据总体最小二乘法原理进行转换矩关系的优化,基于转换关系,将多个激光雷达数据配准到大地坐标系下,实现多个激光雷达数据在大地坐标下的融合。试验对比4种不同型号、不同线束的激光雷达数据转换效果,结果表明各型号激光雷达经配准后均可以GPS数据形式呈现,通过转换数据与测量数据相比,精度可达到5 cm。本算法可较准确快速地获取多激光雷达配准结果,将多个激光雷达数据统一配准到大地坐标系下,配准后的数据在定位、安全评估等领域具有较高的使用价值。
In order to obtain multi-LiDAR registration models accurately and efficiently,a data fusion method of roadside LiDAR based on coordinate transformation was proposed.The data of global positioning system(GPS) was to convert into spatial rectangular coordinate system.The mapping relationship between geocentric space rectangular coordinates and LiDAR data presented in Cartesian coordinate system was solved.We utilized the principle of total least square to optimize transformation matrix.The multiple LiDAR data was convert into Geodetic coordinate system based on the optimized transformation matrix for the realization of multi-LiDAR data fusion.This paper compared the results of four different LiDAR,test results showed that all types of LiDAR could be presented in GPS form after registration.Compared the calculated data and measured data,the error could be controlled within 5 cm.The multi-LiDAR registration results could be obtained through this algorithm accurately and quickly.By registering the multi-LiDAR data to the geodetic coordinate system.Data of LiDAR after registration possesses high application value in positioning,traffic safety analysis and other fields.
作者
周勇
吕琛
侯福金
郭鑫铭
宋修广
ZHOU Yong;LÜ Chen;HOU Fujin;GUO Xinming;SONG Xiuguang(Shandong Hi-Speed Construction Management Group Co.Ltd.,Jinan 250014,Shandong,China;School of Qilu Transportation,Shandong University,Jinan 250002,Shandong,China;Suzhou Research Institute,Shandong University,Suzhou 215123,Jiangsu,China)
出处
《山东大学学报(工学版)》
CAS
CSCD
北大核心
2022年第6期41-49,共9页
Journal of Shandong University(Engineering Science)
基金
国家自然科学基金青年科学基金项目(52002224)
江苏省自然科学基金项目(BK20200226)
苏州市重点产业技术创新项目(SYG202033)
山东省重点研发计划资助项目(2020CXGC010118)。