激光与光电子学进展, 2018, 55 (2): 022803, 网络出版: 2018-09-10   

基于多对点云匹配的三维激光雷达外参数标定 下载: 1713次

Calibration of Three-Dimensional Lidar Extrinsic Parameters Based on Multiple-Point Clouds Matching
作者单位
中国人民解放军军事交通学院研究生管理大队, 天津 300161
摘要
针对无人车三维激光雷达与全球定位系统/惯性导航系统组合导航系统安装位置关系难以准确测量及相对转角无法直接测量的问题,提出一种基于多对点云同时匹配迭代生成外参数的方法。首先选择车辆直线往返行驶中位置相近、方向相反的激光雷达点云对进行匹配;然后设定参数区域中心、初始步长及步数,遍历参数组合,寻找目标函数最小时的外参数组合并更新为最优迭代区域中心;最后不断缩减步长,直到得到满足精度要求的最优外参数。实验采集了两段环境不同的数据,分别采用较优和较差的参数初始迭代中心以及不同步数进行标定。结果表明,所提方法用时短,对于非理想参数初值也能够得到较好的标定结果,而且标定方法简单,无需专门的标定物即可达到需求的标定精度。
Abstract
To solve the problem that the exact position relation between three-dimensional lidar and global positioning system/inertial navigation system integrated navigation system cannot be obtained directly and the relative rotation angle between them cannot be directly measured, a method based on simultaneous matching of multiple-point clouds is proposed to calibrate the extrinsic parameters. Firstly, the lidar point clouds, which are similar in position and opposite in direction, are selected as the calibration data. Secondly, we set the parameters of the initial regional center, the initial step length and the step counts, then traverse the combination of parameters to find the optimal target function when the corresponding extrinsic parameters are updated to the optimal iterative regional center. We shorten the step length continuously, until optimal extrinsic parameter that satisfies accuracy requirement is obtained. The experiment collects two different data in different environments, and uses better and worse parameter initial iterative center and different step counts to calibrate. Results show that the proposed calibration method has a short time duration, and can get good calibration results for non-ideal initial parameters. The method is simple, and can achieve required calibration accuracy without special calibration object.

韩栋斌, 徐友春, 王任栋, 齐尧, 李华. 基于多对点云匹配的三维激光雷达外参数标定[J]. 激光与光电子学进展, 2018, 55(2): 022803. Dongbin Han, Youchun Xu, Rendong Wang, Yao Qi, Hua Li. Calibration of Three-Dimensional Lidar Extrinsic Parameters Based on Multiple-Point Clouds Matching[J]. Laser & Optoelectronics Progress, 2018, 55(2): 022803.

本文已被 5 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!