基于激光雷达的无人驾驶障碍物检测和跟踪

王涛; 曾文浩; 于琪

doi:10.3969/j.issn.0258-2724.20200240

基于激光雷达的无人驾驶障碍物检测和跟踪

doi: 10.3969/j.issn.0258-2724.20200240

西南交通大学电气工程学院，四川成都 610031

基金项目: 国家自然科学基金（51477146）

详细信息

作者简介:
王涛（1972—），男，教授，博士，研究方向为计算机控制技术，E-mail：wangtao618@126.com

中图分类号: TP391
计量
- 文章访问数: 1037
- HTML全文浏览量: 1130
- PDF下载量: 128
- 被引次数: 11
出版历程
- 收稿日期: 2020-04-26
- 修回日期: 2020-11-15
- 网络出版日期: 2021-01-05
- 刊出日期: 2021-01-05

Obstacle Detection and Tracking for Driverless Cars Based on Lidar

School of Electrical Engineering，Southwest Jiaotong University，Chengdu 610031，China

摘要

摘要:
针对激光雷达动态障碍物检测与跟踪过程中聚类适应性差、实时性低和跟踪准确度不高等问题，提出一种自适应的密度聚类算法和多特征数据关联方法，分别用于检测和跟踪. 首先，对激光雷达采集的点云进行路沿检测、感兴趣区域提取和地面分割等预处理，去除无关点云；然后，基于自适应的密度聚类算法对非地面的点云进行聚类，完成障碍物点云检测；最后，利用加权多特征数据关联算法结合卡尔曼滤波器实现对动态障碍物跟踪. 通过实验表明：本算法能够根据10 Hz的激光雷达数据实现对障碍物准确、稳定的检测和跟踪，且聚类时间缩短32%.
- 激光雷达 /
- 聚类 /
- 数据关联 /
- 卡尔曼滤波 /
- 目标跟踪
Abstract:
To improve clustering adaptability, real-time performance, and tracking accuracy in the process of dynamic obstacle detection and tracking with lidar, an adaptive density clustering algorithm and a multi-feature data association method are developed for detection and tracking respectively. Firstly, the point cloud collected by the lidar is pre-processed such as curb detection, area-of-interest extraction and ground segmentation to remove irrelevant point clouds. Then the non-ground point clouds are clustered according to the adaptive clustering algorithm to complete the obstacle point cloud detection. Finally, a weighted multi-feature data association algorithm combined with Kalman filter is used to track dynamic obstacles. The experiments show that the entire detection process with the proposed methods can accurately and stably detect and track the 10 Hz lidar data, and shorten the clustering time by 32%.
- lidar /
- cluster /
- data association /
- Kalman filter /
- target tracking

HTML全文

图 1 “智轨”列车平台以及株洲大道

Figure 1. ART Train Platform and Zhuzhou Avenue

下载: 全尺寸图片幻灯片

图 2 路沿点提取效果

Figure 2. Effect of curb point extraction

下载: 全尺寸图片幻灯片

图 3 地面分割效果

Figure 3. Effect of ground segmentation

下载: 全尺寸图片幻灯片

图 4 球型搜索和椭球型搜索示意

Figure 4. Illustration of spherical search and ellipsoid search

下载: 全尺寸图片幻灯片

图 5 物体发生倾斜时的示意

Figure 5. Schematic diagram in the case of tilted object

下载: 全尺寸图片幻灯片

图 6 椭球域及其代表对象的三维邻域

Figure 6. Three-dimensional neighborhood ofellipsoidal domain and its representative objects

下载: 全尺寸图片幻灯片

图 7 障碍物跟踪算法流程

Figure 7. Flowchart of obstacle tracking algorithm

下载: 全尺寸图片幻灯片

图 8 障碍物与激光雷达的相对位置

Figure 8. Relative position of obstacles and LiDAR

下载: 全尺寸图片幻灯片

图 9 第260帧的聚类结果

Figure 9. Clustering results at the frame 260th

下载: 全尺寸图片幻灯片

图 10 第280帧的聚类结果

Figure 10. Clustering results at the frame 280th

下载: 全尺寸图片幻灯片

图 11 聚类所消耗的时间对比

Figure 11. Comparison of time consumed by clustering

下载: 全尺寸图片幻灯片

图 12 动态障碍物跟踪效果

Figure 12. Effect of dynamic obstacle tracking

下载: 全尺寸图片幻灯片

参考文献(14)

[1]	谢德胜,徐友春,王任栋,等. 基于三维激光雷达的无人车障碍物检测与跟踪[J]. 汽车工程,2018,40(8): 952-959. XIE Desheng, XU Youchun, WANG Rendong, et al. Obstacle detection and tracking for unmanned vehicles based on 3D laser radar[J]. Automotive Engineering, 2018, 40(8): 952-959.
[2]	AZIM A, AYCARD O. Layer-based supervised classification of moving objects in outdoor dynamic environment using 3D laser scanner[C]//2014 IEEE Intelligent Vehicles Symposium Proceedings. Dearborn: IEEE, 2014: 1408-1414.
[3]	黄钢,吴超仲,吕能超. 基于改进DBSCAN算法的激光雷达目标物检测方法[J]. 交通信息与安全,2015,33(3): 23-28. doi: 10.3963/j.issn.1674-4861.2015.03.004 HUANG Gang, WU Chaozhong, LYU Nengchao. A study of laser radar object detection based on improved DBSCAN algorithm[J]. Journal of Transport Information and Safety, 2015, 33(3): 23-28. doi: 10.3963/j.issn.1674-4861.2015.03.004
[4]	包瑞胜,马新,崔熠明. 基于优化DBSCAN算法的智能车载激光雷达数据处理技术研究[J]. 中国交通信息化,2017(3): 135-140.
[5]	ZHANG J S, KEREKES J. An adaptive density-based model for extracting surface returns from photon-counting laser altimeter data[J]. IEEE Geoscience and Remote Sensing Letters, 2015, 12(4): 726-730. doi: 10.1109/LGRS.2014.2360367
[6]	CHEN J R, XU H, WU J Q, et al. Deer crossing road detection with roadside LiDAR sensor[J]. IEEE Access, 2019, 7: 65944-65954. doi: 10.1109/ACCESS.2019.2916718
[7]	LEONARD J, HOW J, TELLER S, et al. A perception-driven autonomous urban vehicle[J]. Journal of Field Robotics, 2008, 25(10): 727-774. doi: 10.1002/rob.20262
[8]	NAVARRO-SERMENT L E, MERTZ C, HEBERT M. Pedestrian detection and tracking using three-dimensional LiDAR data[J]. The International Journal of Robotics Research, 2010, 29(12): 1516-1528. doi: 10.1177/0278364910370216
[9]	汪世财,谈东奎,谢有浩,等. 基于激光雷达点云密度特征的智能车障碍物检测与跟踪[J]. 合肥工业大学学报(自然科学版),2019,42(10): 1311-1317. doi: 10.3969/j.issn.1003-5060.2019.10.003 WANG Shicai, TAN Dongkui, XIE Youhao, et al. Obstacle detection and tracking for intelligent vehicle based on density characteristics of point cloud using 3D lidar[J]. Journal of Hefei University of Technology(Natural Science), 2019, 42(10): 1311-1317. doi: 10.3969/j.issn.1003-5060.2019.10.003
[10]	HAN J, KIM D, LEE M, et al. Enhanced road boundary and obstacle detection using a downward-looking LIDAR sensor[J]. IEEE Transactions on Vehicular Technology, 2012, 61(3): 971-985. doi: 10.1109/TVT.2012.2182785
[11]	ZHOU Y, WANG D, XIE X, et al. A fast and accurate segmentation method for ordered LiDAR point cloud of large-scale scenes[J]. IEEE Geoscience and Remote Sensing Letters, 2014, 11(11): 1981-1985. doi: 10.1109/LGRS.2014.2316009
[12]	XU S, WANG R S, WANG H, et al. An optimal hierarchical clustering approach to mobile LiDAR point clouds[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 21(7): 2765-2776. doi: 10.1109/TITS.2019.2912455
[13]	刘健. 基于三维激光雷达的无人驾驶车辆环境建模关键技术研究[D]. 合肥: 中国科学技术大学, 2016.
[14]	ZHANG Y S, SUN X, XU H, et al. Tracking multi-vehicles with reference points switches at the intersection using a roadside LiDAR sensor[J]. IEEE Access, 2019, 7: 174072-174082. doi: 10.1109/ACCESS.2019.2953747

施引文献

期刊类型引用(7)

1.	尹向雷，苏妮，解永芳，屈少鹏. 基于坐标注意力和软化非极大值抑制的密集安全帽检测. 现代电子技术. 2025(02): 153-161 . 百度学术
2.	康文，魏文博，王涛，薛洲喜，高召涛. 变电站巡检目标自动识别技术的研究. 中国新技术新产品. 2025(02): 20-22 . 百度学术
3.	王嘉璇，王天宁，刘兵，王祝. 基于改进AdvancedEAST的变电站仪表数显区域检测. 电子测量技术. 2024(05): 45-53 . 百度学术
4.	唐俊，赵锦科，赵鑫洋，赵博洋，袁妮，李晨洋. 煤矿巡检机器人仪表识别技术的研究. 工业控制计算机. 2024(06): 10-12 . 百度学术
5.	何敏，秦亮，刘开培，邓欣兰，李博强，李强，徐兴华. 变电站电力仪表智能检测算法研究. 高电压技术. 2024(07): 2942-2954 . 百度学术
6.	迟钰坤，纪洪伟，焦之明，巩方波，刘兆霞，王倩倩. 基于场景变化检测与目标检测的变电站隐患识别算法的对比研究. 电力大数据. 2024(09): 40-50 . 百度学术
7.	陈学台，欧郁强，李敏，许镇宇，黄观荣. 基于YOLOv5及深度神经网络的电力智慧安监算法. 机械与电子. 2024(12): 37-42+48 . 百度学术