汽车工程 ›› 2021, Vol. 43 ›› Issue (8): 1228-1237.doi: 10.19562/j.chinasae.qcgc.2021.08.015

• • 上一篇    下一篇

激光雷达的内参建模与点云修正方法

张晓龙1,江昆1,孙恺2,刘茂林1,邵振雷2,肖鹏川2,杨殿阁1()   

  1. 1.清华大学车辆与运载学院,汽车安全与节能国家重点实验室,北京 100084
    2.上海禾赛科技股份有限公司,上海 201702
  • 收稿日期:2021-02-25 修回日期:2021-04-24 出版日期:2021-08-25 发布日期:2021-08-20
  • 通讯作者: 杨殿阁 E-mail:ydg@mail.tsinghua.edu.cn
  • 基金资助:
    国家重点研发计划(2018YFB0105000)

Intrinsic Parameters Modeling for LiDAR and Point Cloud Correction Method

Xiaolong Zhang1,Kun Jiang1,Kai Sun2,Maolin Liu1,Zhenlei Shao2,Pengchuan Xiao2,Diange Yang1()   

  1. 1.School of Vehicle and Mobility,Tsinghua University,State Key Laboratory of Automotive Safety and Energy,Beijing 100084
    2.Hesai Technology Co. ,Ltd. ,Shanghai 201702
  • Received:2021-02-25 Revised:2021-04-24 Online:2021-08-25 Published:2021-08-20
  • Contact: Diange Yang E-mail:ydg@mail.tsinghua.edu.cn

摘要:

本文中建立了一种针对机械式激光雷达参数较少的内参模型;考虑到偏心结构对于点云的影响,提出了角度修正和距离修正两种点云修正方法,对两种方法的修正效果进行仿真,并通过实际点云加以验证。最后还用位姿估计算法验证修正算法的有效性。

关键词: 激光雷达, 内参模型, 点云修正, 位姿估计算法

Abstract:

In this paper, an intrinsic parameter model with less parameters and dedicated to mechanical LiDAR is established, and in view of the effects of eccentric structure on point cloud, two point?cloud correction methods (angle and distance corrections) are proposed. A simulation is conducted on the effects of two correction methods, with its results verified by real point cloud. Finally, a pose estimation algorithm is also used to validate the effectiveness of the correction methods proposed.

Key words: LiDAR, intrinsic model:simultaneous localization and mapping, LeGO?LOAM