Research on 3D map reconstruction of coal mine underground roadway based on lidar

Xuebin Qin; Wei Tang; Yun Gao

doi:10.54097/74tn7017

Authors

Xuebin Qin
Wei Tang
Yun Gao

DOI:

https://doi.org/10.54097/74tn7017

Keywords:

Laser SLAM feature extraction, point cloud registration, loop detection.

Abstract

The underground coal mine environment is complex, the visual SLAM algorithm is not applicable in low light conditions, the underground roadway is narrow and long and the road surface is rugged. The traditional SLAM algorithm is prone to problems such as mismatching and cumulative errors, which affect the effect of mapping and positioning of inspection robots. Through the research of the A-LOAM algorithm of laser SLAM, combined with 16-line lidar, the 3D global map reconstruction of the underground crawler robot is studied. In the process of point cloud preprocessing, in order to solve the sparse point cloud of the A-LOAM algorithm in the coal mine scene, the Link 3D method is used to extract the point cloud features; in order to solve the cumulative error of SLAM mapping in the large-scale scene of the coal mine, Scan Context is used for loop detection, which improves the accuracy of the global map constructed and meets the requirements for mapping.

Downloads

Download data is not yet available.

References

[1] Zhang Qing Yu, Cui Lin Zhen, Du Xiu Duo, et al. 3D LiDAR SLAM algorithm for mapping and localization in mining environments[J]. Bulletin of Surveying and Mapping, 2023(5): 72-77.

[2] Cui Shao Yun,Bao Jiu Sheng,Hu De Ping,et al. The research status and development trend of SLAM technology and its application in the field of unmanned mine driving [J]. Industrial and mining automation,2024,50(10):38-52.

[3] Cui, Yunge, et al. "Link3d: Linear keypoints representation for 3d lidar point cloud." IEEE Robotics and Automation Letters (2024).

[4] Kim G , Kim A .Scan Context: Egocentric Spatial Descriptor for Place Recognition Within 3D Point Cloud Map[C]//2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).IEEE, 2018.

[5] [1] Yin J , Li A , Li T ,et al.M2DGR: A Multi-Sensor and Multi-Scenario SLAM Dataset for Ground Robots[J].IEEE Robotics and Automation Letters, 2022, 7(2).