Loop closure detection method for LiDAR SLAM supported by stable static point cloud clusters

doi:10.11947/j.AGCS.2025.20250252

Abstract

Abstract:

In dynamic, degenerate, and large-scale cluttered environments, loop closure detection methods based solely on point cloud processing exhibit poor robustness. Moreover, existing methods generally suffer from weak translation sensitivity and low computational efficiency. To address these challenges, this paper proposes a bag-of-words with stable static point cloud clusters-based loop closure detection method. Firstly, the degradation of the preprocessed point cloud is evaluated from the environmental structure perspective, and a robust point cloud cluster classification scheme is designed to obtain the stable static point cloud clusters to weaken the interference of dynamic targets. Subsequently, to reduce the redundancy in loop closure information, the fuzzy comprehensive evaluation algorithm is used to adaptively filter the key frames. Finally, based on the stable static point cloud cluster and keyframe selection results, a bag-of-words with point cloud cluster local descriptors-based loop closure detection algorithm is proposed. The relative spatial relationship and attribute relationship between stable point cloud clusters are used to improve the translation and rotation sensitivity of bag information, so as to ensure the actual performance of loop closure detection in degenerate and cluttered scenes. Experimental results demonstrate that the proposed method can robustly detect the correct loop closure relationship in the measured scene, and the non-loop closure frame error detection rate is only 5.56%, with a single-keyframe processing time of 0.052 8 s. Compared with three similar methods BoW3D, ISC, and SGLC, the average improvement in the loop closure frame correct detection rate reaches 75.73%, the average reduction in the non-loop closure frame error detection rate is 81.93%, the processing has strong real-time performance, and it exhibits stronger robustness and applicability.

Key words: point cloud degradation evaluation, point cloud cluster classification, keyframe selection, bag-of-words model, loop closure detection

CLC Number:

P228

Jiaxin GAO, Xin SUI, Changqiang WANG, Aigong XU, Zhengxu SHI. Loop closure detection method for LiDAR SLAM supported by stable static point cloud clusters[J]. Acta Geodaetica et Cartographica Sinica, 2025, 54(12): 2194-2205.

Figures/Tables 9

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Tab. 1

Tab. 2

Fig. 5

Fig. 6

Tab. 3

References 30

[1]	YAROVOI A, CHO Y K. Review of simultaneous localization and mapping (SLAM) for construction robotics applications[J]. Automation in Construction, 2024, 162: 105344-105371.
[2]	蔺小虎, 杨鑫, 姚顽强, 等. 面向复杂地下空间的多源传感器数据动态加权融合SLAM方法[J]. 测绘学报, 2025, 54(3): 523-535. DOI: . doi: 10.11947/j.AGCS.2025.20230586
	LIN Xiaohu, YANG Xin, YAO Wanqang, et al. A dynamic weighted fusion SLAM method using multi-source sensor data in complex underground spaces[J]. Acta Geodaetica et Cartographica Sinica, 2025, 54(3): 523-535. DOI: . doi: 10.11947/j.AGCS.2025.20230586
[3]	AL-TAWIL B, HEMPEL T, ABDELRAHMAN A, et al. A review of visual SLAM for robotics: evolution, properties, and future applications[J]. Frontiers in Robotics and AI, 2024, 11: 1347985-1348002.
[4]	李帅鑫, 李九人, 田滨, 等. 面向点云退化的隧道环境的无人车激光SLAM方法[J]. 测绘学报, 2021, 50(11): 1487-1499. DOI: . doi: 10.11947/j.AGCS.2021.20210248
	LI Shuaixin, LI Jiuren, TIAN Bin, et al. A laser SLAM method for unmanned vehicles in point cloud degenerated tunnel environments[J]. Acta Geodaetica et Cartographica Sinica, 2021, 50(11): 1487-1499. DOI: . doi: 10.11947.AGCS.2021.20210248
[5]	YUE Xiangdi, ZHANG Yihuan, CHEN Jiawei, et al. LiDAR-based SLAM for robotic mapping: state of the art and new frontiers[J]. Industrial Robot: the International Journal of Robotics Research and Application, 2024, 51(2): 196-205.
[6]	WU Haoyu, CHEN Yinglong, YANG Qiming, et al. A review of underwater robot localization in confined spaces[J]. Journal of Marine Science and Engineering, 2024, 12(3): 428.
[7]	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 (IROS). Madrid: IEEE, 2018: 4758-4765.
[8]	SHAN Tixiao, ENGLOT B, MEYERS D, et al. LIO-SAM: tightly-coupled LiDAR inertial odometry via smoothing and mapping[C]//Proceedings of 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems. Las Vegas: IEEE, 2020: 5135-5142.
[9]	KIM G, KIM A. Scan context: egocentric spatial descriptor for place recognition within 3D point cloud map[C]//Proceedings of 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems. Madrid: IEEE, 2018: 4802-4809.
[10]	KIM G, CHOI S, KIM A. Scan Context++: structural place recognition robust to rotation and lateral variations in urban environments[J]. IEEE Transactions on Robotics, 2022, 38(3): 1856-1874.
[11]	WANG Ying, SUN Zezhou, XU Chengzhong, et al. LiDAR Iris for loop-closure detection[C]//Proceedings of 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems. Las Vegas: IEEE, 2020: 5769-5775.
[12]	CUI Yunge, CHEN X, ZHANG Yinlong, et al. BoW3D: bag of words for real-time loop closing in 3D LiDAR SLAM[J]. IEEE Robotics and Automation Letters, 2023, 8(5): 2828-2835.
[13]	JI Xingliang, ZUO Lin, ZHANG Changhua, et al. LLOAM: LiDAR odometry and mapping with loop-closure detection based correction[C]//Proceedings of 2019 IEEE International Conference on Mechatronics and Automation. Tianjin: IEEE, 2019: 2475-2480.
[14]	WANG Han, WANG Chen, XIE Lihua. Intensity scan context: coding intensity and geometry relations for loop closure detection[C]//Proceedings of 2020 IEEE International Conference on Robotics and Automation. Paris: IEEE, 2020: 2095-2101.
[15]	潘锋, 蒋林, 胡雨欣, 等. 基于点云强度与高度信息的回环检测算法[J]. 武汉科技大学学报, 2022, 45(6): 464-471.
	PAN Feng, JIANG Lin, HU Yuxin, et al. Loop closure detection algorithm based on point cloud intensity and height information[J]. Journal of Wuhan University of Science and Technology, 2022, 45(6): 464-471.
[16]	WANG Neng, CHEN Xieyuanli, SHI Chenghao, et al. SGLC: semantic graph-guided coarse-fine-refine full loop closing for LiDAR SLAM[J]. IEEE Robotics and Automation Letters, 2024, 9(12): 11545-11552.
[17]	NGUYEN T M, YUAN Shenghai, CAO Muqing, et al. Viral SLAM: tightly coupled camera-IMU-UWB-LiDAR SLAM[EB/OL]. [2025-04-01]. https://arxiv.org/pdf/2105.03296.
[18]	LIN Jiarong, ZHENG Chunran, XU Wei, et al. R2LIVE: a robust, real-time, LiDAR-inertial-visual tightly-coupled state estimator and mapping[J]. IEEE Robotics and Automation Letters, 2021, 6(4): 7469-7476.
[19]	LEVINSON J, THRUN S. Unsupervised calibration for multi-beam lasers[M]//Experimental robotics. Berlin: Springer, 2014: 179-193.
[20]	HONG S, KO H, KIM J. VICP: velocity updating iterative closest point algorithm[C]//Proceedings of 2010 IEEE International Conference on Robotics and Automation. Anchorage: IEEE, 2010: 1893-1898.
[21]	FAN Weiwei, LIU Xinyi, ZHANG Yongjun, et al. Airborne LiDAR point cloud filtering algorithm based on supervoxel ground saliency[J]. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2024, X-2-2024: 73-79.
[22]	HIMMELSBACH M, HUNDELSHAUSEN F V, WUENSCHE H J. Fast segmentation of 3D point clouds for ground vehicles[C]//Proceedings of 2010 IEEE Intelligent Vehicles Symposium. La Jolla: IEEE, 2010: 560-565.
[23]	徐爱功, 高佳鑫, 隋心, 等. 退化环境下LiDAR里程计定位方法[J]. 地球信息科学学报, 2024, 26(10): 2316-2334.
	XU Aigong, GAO Jiaxin, SUI Xin, et al. LiDAR odometry positioning method for degenerate environment[J]. Journal of Geo-information Science, 2024, 26(10): 2316-2334.
[24]	XU Aigong, GAO Jiaxin, SUI Xin, et al. LiDAR dynamic target detection based on multidimensional features[J]. Sensors, 2024, 24(5): 1369.
[25]	刘康, 仇培元, 刘希亮, 等. 利用词向量模型分析城市道路交通空间相关性[J]. 测绘学报, 2017, 46(12): 2032-2040. DOI: . doi: 10.11947/j.AGCS.2017.20170166
	LIU Kang, QIU Peiyuan, LlU Xiliang, et al. Measuring traffic correlations in urban road system using word embedding model[J]. Acta Geodaetica et Cartographica Sinica, 2017, 46(12): 2032-2040. DOI: . doi: 10.11947/j.AGCS.2017.20170166
[26]	OZENDI M, AKCA D, TOPAN H. A point cloud filtering method based on anisotropic error model[J]. The Photogrammetric Record, 2023, 38(184): 460-497.
[27]	王俊森, 金绍华, 边刚, 等. 结合不确定度与密度聚类算法的多波束异常值自动滤波算法[J]. 测绘学报, 2023, 52(10): 1669-1678. DOI: . doi: 10.11947/j.AGCS.2023.20220579
	WANG Junsen, JIN Shaohua, BIAN Gang, et al. A multi-beam outlier automatic filtering algorithm combining uncertainty and density clustering method[J]. Acta Geodaetica et Cartographica Sinica, 2023, 52(10): 1669-1678. DOI: . doi: 10.11947/j.AGCS.2023.20220579
[28]	SONG Dong, WANG Bukang, PANG Jifang, et al. Feasibility evaluation of highwall mining in open-pit coal mine based on method of integrated analytic hierarchy process-fuzzy comprehensive evaluation-variable weight theory[J]. Electronics, 2023, 12(21): 4460.
[29]	GUO Lei, XU Jihui, HE Weihui. Health assessment of mission-oriented TT&C antenna[J]. Journal of Physics: Conference Series, 2023, 2489(1): 012019.
[30]	杨文娟, 王文明, 王全玉, 等. 基于感知哈希和视觉词袋模型的图像检索方法[J]. 图学学报, 2019, 40(3): 519-524.
	YANG Wenjuan, WANG Wenming, WANG Quanyu, et al. Image retrieval method based on perceptual hash algorithm and bag of visual words[J]. Journal of Graphics, 2019, 40(3): 519-524.

方法	场景	ηr	η_e	t
BoW3D	1	1.000 0	0.166 7	0.053 8
	2	0.733 3	1.000 0	0.056 7
	3	1.000 0	0.250 0	0.073 4
	均值	0.911 1	0.472 2	0.061 3
ISC	1	0.000 0	0.000 0	0.071 0
	2	0.266 7	0.500 0	0.072 6
	3	0.692 3	0.000 0	0.096 2
	均值	0.319 7	0.166 7	0.079 9
SGLC	1	1.000 0	0.333 3	0.045 6
	2	0.866 7	1.000 0	0.046 1
	3	1.000 0	0.500 0	0.050 2
	均值	0.955 6	0.611 1	0.047 3
本文方法	1	1.000 0	0.166 7	0.046 7
	2	1.000 0	0.000 0	0.048 9
	3	1.000 0	0.000 0	0.062 8
	均值	1.000 0	0.055 6	0.052 8

算法	RMSE/m	ME/m	单帧耗时/s
LeGO-LOAM	13.456 9	86.224 1	0.037 2
U-LeGO-LOAM	1.249 9	4.528 8	0.047 6