基于双目视觉和单线激光雷达的三维场景重建系统和算法

doi:10.13474/j.cnki.11-2246.2024.0509

摘要/Abstract

摘要： 双目视觉和激光雷达是实现三维场景重建的两种有效方法,但它们都具有自身的局限性。将视觉传感器和激光传感器的数据相融合,可克服其各自的缺陷,具有重要意义。本文针对单线旋转激光雷达设备的独特性,提出了一种基于双目图像和激光雷达数据的模块化双目视觉-激光雷达SLAM(Stereo-LiDAR SLAM)系统和算法。在该方法中,双目视觉信息被用于消除激光雷达点云的畸变,并支撑双目视觉定位(VO)模块,为整个系统提供初始的位姿估计。然后,独立于VO模块的激光雷达SLAM(L-SLAM)模块对位姿参数进行优化,从而得到高精度的三维场景重建结果。试验表明,本文设计的系统和算法可以有效提高大规模低成本三维场景重建的精度和环境适应能力。

关键词: 双目视觉, 激光雷达, 三维重建

Abstract: Stereo vision and LiDAR are two effective methods for 3D scene reconstruction, but they both have some limitations. As a result, it is meaningful to fuse visual sensor data and LiDAR data in order to conquer their weaknesses. In this paper, we address the uniqueness of the single-line spinning LiDAR device, and propose a modular visual-LiDAR SLAM algorithm based on the integration of image and range data. In the method, visual information is used to undistort the LiDAR point cloud and provide an initial pose estimation from the visual odometry (VO) module. After that, pose refinement is performed by a LiDAR SLAM (L-SLAM) module which is independent from the VO module, and then we obtain highly accurate 3D scene reconstruction results. Experiments show that our system and algorithm could increase the accuracy and adaptation of low-cost large-scale 3D scene reconstruction tasks.

Key words: stereo vision, LiDAR, 3D reconstruction

中图分类号:

P237

钟雷声, 夏辉, 陈佳林. 基于双目视觉和单线激光雷达的三维场景重建系统和算法[J]. 测绘通报, 2024, 0(5): 48-52,59.

ZHONG Leisheng, XIA Hui, CHEN Jialin. 3D scene reconstruction system and algorithm based on stereo vision and single-line LiDAR[J]. Bulletin of Surveying and Mapping, 2024, 0(5): 48-52,59.

参考文献

[1] 李玉美,郭庆华,万波,等. 基于激光雷达的自然资源三维动态监测现状与展望[J]. 遥感学报,2021,25(1): 381-402.
[2] URMSON C,ANHALT J,BAGNELL D,et al. Autonomous driving in urban environments: boss and the urban challenge[J]. Journal of Field Robotics,2008,25(8): 425-466.
[3] 周宗锟,姜卫平,唐健,等. LiDAR地图匹配与二维码融合的AGV室内定位与导航[J]. 测绘通报,2021(1): 9-12.
[4] 王晨捷,罗斌,李成源,等. 无人机视觉SLAM协同建图与导航[J]. 测绘学报,2020,49(6): 767-776.
[5] 段敏燕. 机载激光雷达点云电力线三维重建方法研究[J]. 测绘学报,2016,45(12): 1495.
[6] 张清宇,崔丽珍,杜秀铎,等. 矿山环境三维激光雷达SLAM算法建图与定位[J].测绘通报,2023(5):72-77.
[7] GOMEZ-OJEDA R,MORENO F A,ZUÑIGA-NOËL D,et al. PL-SLAM: a stereo SLAM system through the combination of points and line segments[J]. IEEE Transactions on Robotics,2019,35(3): 734-746.
[8] 李旺,牛铮,高帅,等.机载激光雷达数据分析与反演青海云杉林结构信息[J].遥感学报,2013,17(6):1612-1626.
[9] COLE D M,NEWMAN P M. Using laser range data for 3D SLAM in outdoor environments[C]//Proceedings of 2006 IEEE International Conference on Robotics and Automation.Orlando,FL,USA:IEEE,2006:1556-1563.
[10] 王铉彬,李星星,廖健驰,等. 基于图优化的紧耦合双目视觉/惯性/激光雷达SLAM方法[J]. 测绘学报,2022,51(8): 1744-1756.
[11] ZHANG J,SINGH S.LOAM: LiDAR odometry and mapping in real-time[C]//Proceedings of 2014 Science and Systems Conference on Robotics.Berkeley:[s.n.],2014.
[12] ZHANG J,SINGH S. Laser-visual-inertial odometry and mapping with high robustness and low drift[J]. Journal of Field Robotics,2018,35(8): 1242-1264.
[13] 彭梦,万琴,陈白帆,等. 基于观测概率有效下界估计的二维激光雷达和摄像机标定方法[J]. 电子与信息学报,2022,44(7): 2478-2487.
[14] YAN Li,DAI Jicheng,TAN Junxiang,et al. Global fine registration of point cloud in LiDAR SLAM based on pose graph[J]. Journal of Geodesy and Geoinformation Science,2020,3(2): 26-35.
[15] RUSINKIEWICZ S,LEVOY M. Efficient variants of the ICP algorithm[C]//Proceedings of the 3rd International Conference on 3D Digital Imaging and Modeling. Quebec City: IEEE,2002: 145-152.
[16] SHAN Tixiao,ENGLOT B. LeGO-LOAM: lightweight and ground-optimized lidar odometry and mapping on variable terrain[C]//Proceedings of 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems. Madrid: IEEE,2018: 4758-4765.