Negative Obstacle Detection for Ground Unmanned Vehicles Using Multiple Types of LiDAR in Unstructured Environments

doi:10.12382/bgxb.2025.0197

Abstract

Abstract:

The negative obstacles pose a significant threat to the driving safety of unmanned ground vehicles (UGVs) when they operate in unstructured environments.However,in the existing research on negative obstacle detection,the camera-based detection methods are not well suited for the detection of negative obstacles in unstructured environments with poor lighting conditions and complex backgrounds; and LiDAR(light detection and ranging)-based detection methods are mostly designed based on the characteristics of mechanical LiDAR harnesses and have poor universality.Therefore,a highly compatible and scalable negative obstacle detection method is proposed,which can adapt to different types of LiDAR and achieve high-precision negative obstacle detection.The proposed method involves preprocessing the point cloud data,extracts the ground regions of interest,and performs point cloud pose correction.The adaptive resolution polar coordinate rasterization technology is used to enhance the spatial representation capability of point cloud data.A negative obstacle grid feature descriptor is designed,and the potential negative obstacle regions are extracted from multiple features such as the hollow characteristics,height differences,and minimum height of point clouds.A multi-frame fusion strategy is introduced,and the high-precision negative obstacle surface occupancy range is outputed through map reprojection and Bayesian rule-based probability updates.The experimental results show that the proposed method is applicable to LiDARs with different scanning modes,and can effectively identify the negative obstacle areas in complex unstructured environments.

Key words: unmanned ground vehicle, liDAR, negative obstacle, unstructured environment

CLC Number:

TP242.6

WU Danfeng, CHEN Tongzhou, KUANG Minchi, SONG Chunsen, ZHOU Fenfen, ZHANG Xueyan. Negative Obstacle Detection for Ground Unmanned Vehicles Using Multiple Types of LiDAR in Unstructured Environments[J]. Acta Armamentarii, 2025, 46(11): 250197-.

Figures/Tables 28

Fig.1 System framework

Fig.2 The jolting of UAV

Fig.3 Point cloud filtering

Fig.4 Adaptive resolution polar grid construction

Fig.5 Imaging principle of negative obstacle point clouds

Fig.6 Change in an amount of negative obstacle point clouds

Fig.7 Change in height difference of negative obstacle point clouds

Fig.8 Change in minimum height of negative obstacle point clouds

Fig.9 Multi-feature fusion

Fig.10 Multi-feature thermal distribution

Fig.11 Negative obstacle area detection algorithm based on feature descriptor

Fig.12 Map reprojection

Fig.13 Bayesian grid probability evolution map

Fig.14 Negative obstacle clustering based on two-dimensional grid connectivity

Table 1 Negative obstacle clustering

聚类	场景1	场景2	场景3
无聚类
有聚类

Fig.15 Experimental scenario

Table 2 Experimental results on validation of adaptive resolution polar grid

栅格类型	5~10m	10~15m	15~20m
固定分辨率极坐标栅格
直角坐标栅格
自适应分辨率极坐标栅格

Table 3 Experimental results on validation of adaptive resolution polar grid

栅格类型	指标	5~10m	10~15m	15~20m
直角坐标栅格	IoU	0.71	0.69	0.51
固定分辨率极坐标栅格	IoU	0.79	0.68	0.55
自适应分辨率极坐标栅格	IoU	0.82	0.71	0.58

Table 4 Experimental results on the effectiveness of negative obstacle grid feature descriptor

特征类型	5~10m	10~15m	15~20m
融合特征
去除密集空洞特征
去除高度差特征
去除最低高度特征

Table 5 Experimental results on the effectiveness of negative obstacle grid feature descriptor

特征类型	指标	5~10m	10~15m	15~20m
去除密集空洞特征	FPR	0.16	0.11	0.17
去除高度差特征	FPR	0.15	0.16	0.12
去除最低高度特征	FPR	0.09	0.13	0.20
融合特征	FPR	0.07	0.11	0.10

Table 6 Experimental results on the effectiveness of multi-frame data fusion

策略	5~10m	10~15m	15~20m
多帧融合
单帧检测

Table 7 Experimental results on the effectiveness of multi-frame data fusion

检测类型	指标	5~10m	10~15m	15~20m
单帧检测	IoU	0.70	0.63	0.44
多帧融合	IoU	0.79	0.70	0.55

Table 8 Comparative experimental results of hybrid solid-state LiDAR

场景	本文方法	文献[17]方法	文献[18]方法
场景1
场景2
场景3

Table 9 Comparative experimental results of hybrid solid-state LiDAR

场景	指标	本文方法	文献[17]方法	文献[18]方法
场景1	IoU	0.80	0.76	0.71
场景2	IoU	0.79	0.78	0.75
场景3	IoU	0.74	0.72	0.71

Table 10 Comparison results of hybrid solid-state LiDARs performances

计算对象及平均耗时	本文方法	文献[17]方法	文献[18]方法
计算对象	栅格	三维点云	点云对
平均耗时/ms	6	9	11

Table 11 Experimental results of mechanical LiDARs

场景	本文方法	文献[17]方法	文献[18]方法
场景1
场景2

Table 12 Experimental results of mechanical LiDAR

场景	指标	本文方法	文献[17]方法	文献[18]方法
场景1	IoU	0.83	0.73	0.80
场景2	IoU	0.84	0.69	0.85

Table 13 Performance comparison of mechanical LiDARs

计算对象及平均耗时	本文方法	文献[17]方法	文献[18]方法
计算对象	栅格	三维点云	点云对
平均耗时/ms	4	6	8

References 23

[24]	NAWALE S, KHUT D, DAVE D, et al. Potholeguard:a pothole detection approach by point cloud semantic segmentation[C]// Proceedings of the 2023 International Conference on Modeling & E-Information Research,Artificial Learning and Digital Applications.Karawang,Indonesia:IEEE, 2023:138-143.
[25]	LIU Z J, FAN G Y, RAO L, et al. Positive and negative obstacles detection based on dual-lidar in field environments[J]. IEEE Robotics and Automation Letters, 2024, 9(8):6768-6775. doi: 10.1109/LRA.2024.3414256 URL
[1]	THAKKER R, ALATUR N, FAN D D, et al. Autonomous off-road navigation over extreme terrains with perceptually-challenging conditions[C]// Proceedings of the Experimental Robotics:The 17th International Symposium.Cham,Switzerland:Springer, 2021:161-173.
[2]	周梦如, 陈慧岩, 熊光明, 等. 越野环境下无人履带平台的道路可通行性分析[J]. 兵工学报, 2022, 43(10):2485-2496. doi: 10.12382/bgxb.2021.0824
	ZHOU M R, CHEN H Y, XIONG G M, et al. Road traversability analysis of unmanned tracked platform in off-road environment[J]. Acta Armamentarii, 2022, 43(10):2485-2496. (in Chinese) doi: 10.12382/bgxb.2021.0824
[3]	KARUNASEKERA H, ZHANG H D, XI T, et al. Stereo vision based negative obstacle detection[C]// Proceedings of the 2017 13rd IEEE International Conference on Control & Automation.Ohrid,Macedonia:IEEE, 2017:834-838.
[4]	RANKIN A L, HUERTAS A, MATTHIES L H. Night-time negative obstacle detection for off-road autonomous navigation[J]. Proceedings of SPIE-the International Society for Optical Engineering, 2007, 6561:656103.
[5]	HU T B, NIE Y M, WU T, et al. Negative obstacle detection from image sequences[C]// Proceedings of the 3rd International Conference on Digital Image Processing.Bellingham,WA,US:SPIE, 2011:80090Y.
[6]	FENG Z, FENG Y C, GUO Y N, et al. Adaptive-mask fusion network for segmentation of drivable road and negative obstacle with untrustworthy features[C]// Proceedings of the 2023 IEEE Intelligent Vehicles Symposium.Anchorage,AK,US:IEE, 2023:1-6.
[7]	HUO G L, CAO C Q, LI Y X, et al. Research on negative road obstacle detection based on multimodal feature enhancement and fusion[J]. Applied Sciences, 2025, 15(3):2076-3417. doi: 10.3390/app15042076 URL
[8]	HAN J Z, ZHANG Z H, GAO X S, et al. Research on negative obstacle detection method based on image enhancement and improved anchor box yolo[C]// Proceedings of the 2022 IEEE International Conference on Mechatronics and Automation.Guilin,China:IEEE, 2022:1216-1221.
[9]	BUČKO B, LIESKOVSKÁ E, ZÁBOVSKÁ K, et al. Computer vision based pothole detection under challenging conditions[J]. Sensors, 2022, 22(22):8878. doi: 10.3390/s22228878 URL
[10]	RUAN S L, LI S B, LU C W, et al. A real-time negative obstacle detection method for autonomous trucks in open-pit mines[J]. Sustainability, 2022, 15(1):120. doi: 10.3390/su15010120 URL
[11]	HEO D H, CHOI J Y, KIM S B, et al. Image-based pothole detection using multi-scale feature network and risk assessment[J]. Electronics, 2023, 12(4):826. doi: 10.3390/electronics12040826 URL
[12]	KHARE O M, GANDHI S, RAHALKAR A M, et al. Yolov8-based visual detection of road hazards:potholes,sewer covers,and manholes[C]// Proceedings of the 2023 IEEE Pune Section International Conference.Pune,India:IEEE, 2023:1-6.
[13]	ZHAO T, YANG L, XIE Y C, et al. Roadbev:road surface reconstruction in bird’s eye view[J]. IEEE Transactions on Intelligent Transportation Systems, 2024, 25(11):19088-19099. doi: 10.1109/TITS.2024.3431671 URL
[14]	PARASNIS G, CHOKSHI A, JAIN V, et al. Roadscan:a novel and robust transfer learning framework for autonomous pothole detection in roads[C]// Proceedings of the 2023 IEEE 7th Conference on Information and Communication Technology.Jabalpur,India:IEEE, 2023:1-6.
[15]	SAHA P K, ARYA D, KUMAR A, et al. Road rutting detection using deep learning on images[C]// Proceedings of the 2022 IEEE International Conference on Big Data.Osaka,Japan:IEEE, 2022:1362-1368.
[16]	刘家银, 唐振民, 王安东, 等. 基于多激光雷达与组合特征的非结构化环境负障碍物检测[J]. 机器人, 2017, 39(5):638-651. doi: 10.13973/j.cnki.robot.2017.0638
	LIU J Y, TANG Z M, WANG A D, et al. Negative obstacle detection in unstructured environment based on multiple lidars and compositional features[J]. Robot, 2017, 39(5):638-651. (in Chinese) doi: 10.13973/j.cnki.robot.2017.0638
[17]	LI N, YU X W, LIU X Y, et al. 3D-lidar based negative obstacle detection in unstructured environment[C]// Proceedings of the 2020 4th International Conference on Vision,Image and Signal Processing.New York,NY,US:ACM, 2021:1-6.
[18]	SHANG E K, AN X J, LI J, et al. A novel setup method of 3d lidar for negative obstacle detection in field environment[C]// Proceedings of the 17th International IEEE Conference on Intelligent Transportation Systems.Qingdao,China:IEEE, 2014:1436-1441.
[19]	SHANG E K, AN X J, WU T, et al. Lidar based negative obstacle detection for field autonomous land vehicles[J]. Journal of Field Robotics, 2016, 33(5):591-617. doi: 10.1002/rob.2016.33.issue-5 URL
[20]	CHEN L, YANG J, KONG H, et al. Lidar-histogram for fast road and obstacle detection[C]// Proceedings of 2017 IEEE International Conference on Robotics and Automation,Singapore:IEEE. 2017:1343-1348.
[21]	MORTON R D, OLSON E. Positive and negative obstacle detection using the HLD classifier[C]// Proceedings of the 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems.San Francisco,CA,US:IEEE, 2011:1579-1584.
[22]	XIE P, WANG H C, HUANG Y X, et al. Lidar-based negative obstacle detection for unmanned ground vehicles in orchards[J]. Sensors, 2024, 24 (24):7929. doi: 10.3390/s24247929 URL
[23]	CHEN W, LIU Q J, HU H S, et al. Novel laser-based obstacle detection for autonomous robots on unstructured terrain[J]. Sensors, 2020, 20 (18):5048. doi: 10.3390/s20185048 URL