[智能感知]基于四线激光雷达的无人车障碍物检测算法

中国机械工程

[智能感知]基于四线激光雷达的无人车障碍物检测算法

王海1;郑正扬1;蔡英凤2;陈龙2

1.江苏大学汽车与交通工程学院，镇江，212013

2.江苏大学汽车工程研究院，镇江，212013

出版日期:2018-08-10 发布日期:2018-08-06
基金资助:
国家重点研发计划资助项目（2018YFB0105003）；
国家自然科学基金汽车联合基金资助重点项目（U1564201，U1664258，U1764257，U1762264）；
国家自然科学基金资助项目（61601203，61773184）；
江苏省重点研发计划（产业前瞻与共性关键技术）资助项目（BE2016149）

Obstacle Detection Algorithm for Driverless Vehicles Based on Four-layer Lidars

WANG Hai1;ZHENG Zhengyang1;CAI Yingfeng2;CHEN Long2

1.School of Automotive and Traffic Engineering，Jiangsu University，Zhenjiang，Jiangsu，212013

2.Automotive Engineering Research Institute，Jiangsu University，Zhenjiang，Jiangsu，212013

Online:2018-08-10 Published:2018-08-06

摘要/Abstract

摘要：

针对雷达扫描角分辨能力不高，远处点云较稀疏，导致对目标的分割聚类不准确的问题，提出了一种基于四线激光雷达的障碍物检测算法。在对原始雷达数据进行栅格化及滤除之后，对滤波结果进行了聚类分割。依据栅格深度值确定聚类距离阈值，结合连通区域标记算法得到准确的聚类范围；通过匹配相邻障碍物运动状态信息，进一步提高聚类的准确率。实车实验表明，所提方法较已有的分割聚类算法，可以准确地得到路面障碍物信息，并能有效避免过分割现象。

关键词: 激光雷达, 障碍物检测, 栅格地图, 聚类

Abstract: Aiming at the problems that the resolution capability of lidar scanning angle was not high enough， the remote point cloud was sparse and the target segmentation and clustering were not accurate， an obstacle detection algorithm was proposed based on four-layer lidars. After rasterization and filtering of raw radar data， clustering and segmentation of filtering results were performed. The clustering distance threshold was determined by the grid depth values， and the accurate clustering ranges were obtained by connecting the connected region marking algorithm. The matching accuracy of clustering was further improved by matching the movement status informations of adjacent obstacles. The experimental results show that the proposed algorithm， compared with existing segmentation clustering algorithms， may obtain the information of obstacles on road accurately and prevent the over-segmentation effectively.

Key words: lidar, obstacle detection, grid map, clustering

中图分类号:

TP39

王海, 郑正扬, 蔡英凤, 陈龙. [智能感知]基于四线激光雷达的无人车障碍物检测算法[J]. 中国机械工程.

WANG Hai, ZHENG Zhengyang, CAI Yingfeng, CHEN Long. Obstacle Detection Algorithm for Driverless Vehicles Based on Four-layer Lidars[J]. China Mechanical Engineering.

参考文献

[1]QI C R， SU H， MO K， et al. PointNet： Deep Learning on Point Sets for 3D Classification and Segmentation[C］//IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas， 2016： 77-85.
[2]MANDICI S， NEDEVSCHI S. Aggregate Road Surface Based Environment Representation Using Digital Elevation Maps[C］//IEEE International Conference on Intelligent Computer Communication and Processing. Beijing， 2014：149-156.
[3]THRUN S，MONTEMERLO M，DAHLKAMP H， et al. Stanley： the Robot that Won the DARPA Grand Challenge[J］. Journal of Field Robotics， 2006， 23（9）： 661-692.
[4]FERGUSON D， DARMS M， URMSON C， et al. Detection， Prediction， and Avoidance of Dynamic Obstacles in Urban Environments[C］//IEEE Intelligent Vehicles Symposium. Piscataway： IEEE， 2008： 1149-1154.
[5]PFAFF P，TRIEBEL R，BURGARD W.An Efficient Extension to Elevation Maps for Outdoor Terrain Mapping and Loop Closing[J］. The International Journal of Robotics Research，2007，26（2）：217-230.
[6]DOUILLARD B， UNDERWOOD J， MELKUMYAN N， et al. Hybird Elevation Maps：3D Surface Models for Segmentation[C］//IEEE/RSJ International Conference on Intelligent Robots and Systems. St. Louis： IEEE， 2010：1532-1538.
[7]KONRAD M， SZCZOT M， SCHüLE F， et al. Generic Grid Mapping for Road Course Estimation[C］// Intelligent Vehicles Symposium（IV）. Baden-baden， 2011： 851-856.
[8]PETROVSKAYA A， THRUN S. Model Based Vehicle Detection and Tracking for Autonomous Urban Driving[J］. Autonomous Robots， 2009，26（2）：123-139.
[9]程健. 基于三维激光雷达的实时目标检测[D］.杭州：浙江大学， 2014.
CHENG Jian. 3D Lidar Based Real-time Objects Detection[D］. Hangzhou： Zhejiang University， 2014.
[10]YIAKOPOULOS C T， GRYLLIAS K C， ANTONIADIS I A.Rolling Element Bearing Fault Detection in Industrial Environments Based on a K-means Clustering Approach[J］. Expert Systems with Applications， 2011，38（3）： 2888-2911.
[11]NG R T，HAN J.Efficient and Effective Clustering Methods for Spatial Data Mining[C］ //Proc. of the 20th VLDB Conference. Santiago，2010：144-155.
[12]甘志梅，王春香，杨明. 基于激光雷达的车辆跟踪与识别算法[J］. 上海交通大学学报，2009， 43（6）：923-926.

GAN Zhimei， WANG Chunxiang， YANG Ming. A Method for Vehicle Tracking and Recognition Based on Scanning Laser Radar[J］.Journal of Shanghai Jiaotong University，2009，43（6）：923-926.
[13]BORGES G， ALDON M. Line Extraction in 2D Range Images for Mobile Robotics[J］. Journal of Intelligent and Robotic Systems， 2004， 40： 267-297.
[14]段建民，李龙杰，郑凯华. 基于车载4线激光雷达的前方道路可行驶区域检测[J］. 汽车电子， 2016， 28（2）：55-61.

DUAN Jianmin， LI Longjie， ZHENG Kaihua. Preceding Drivable Area Detection Based on Four-layer Laser Radar[J］. Automobile Technology，2016， 28（2）：55-61.
[15]FORSYTH D A， PONCE J. Computer Vision： a Modern Approach[M］. Englewood: Prentice Hall Professional Technical Reference， 2002.
[16]SKRZYPCZYNSKI P. Building Geometrical map of environment using IR range finder data[C］//Intelligent Autonomous Systems. Karlsruhe， 1995： 408-412.
[17]周俊静.基于激光雷达的智能车辆目标识别与跟踪[D］.北京：北京工业大学，2014.

ZHOU Junjing. Research on Key Technology of Lidar Based Object Detection and Tracking for Intelligent Vehicles[D］. Beijing： Beijing University of Technology， 2014.