|
|
Real-time vehicle detection using 3D lidar under complex urban environment |
CHENG Jian1,2, XIANG Zhi-yu1,2,YU Hai-bin3, LIU Ji-lin1,2 |
1.Department of Information Science and Electronic Engineering, Zhejiang University, Hangzhou 310027, China; 2. Zhejiang Provincial Key Laboratory of Information Network Technology, Hangzhou 310027, China; 3. School of Electronic and Information, Hangzhou Dianzi University, Hangzhou 310018, China |
|
|
Abstract In order to improve the vehicle detection performance, a real-time algorithm to accurately detect static or dynamic vehicle in a single frame under complex urban environment was presented. The system combines 3D point clouds and 2D grid map, i.e., obstacle segmentation with contour extracting in 2D grid map and obstacle and vehicle detection in 3D point clouds. The detecting performance mainly depends on the designed features which should have obvious distinguishing characteristics. To solve the difficulties in distinguishing geometry similar objects and handle occlusion issues, three novel features, i.e., intensity reflection probability distribution, lengthways contour in height, and position-related features were proposed. Finally a support vector machine (SVM) classifier was used to realize real-time detection. The comparison of detecting performance from different features and combined features was analyzed as well. Experiments under real urban environment show that the algorithm can stably detect vehicles within 200 ms per frame, verified the robustness of the proposed method.
|
Published: 01 December 2014
|
|
城市复杂环境下基于三维激光雷达实时车辆检测
针对目前车辆检测准确性不高的问题,提出一种在城市复杂环境下单帧准确实时检测静止或运动车辆的算法.车辆检测系统采用三维激光雷达点云和二维栅格相结合,其中障碍聚类和障碍块轮廓在二维栅格地图上提取,而障碍检测和车辆检测在三维点云数据上处理.检测准确性的高低主要取决于所选的特征是否具有很好的区分性.针对几何形状相似难以区分和物体遮挡等问题,采用3种特征描述:反射强度概率分布、纵向高度轮廓分布和位置姿态相关特征,并利用支持向量机(SVM)训练分类器实现车辆的实时检测.实验中比较和分析了不同单特征和综合特征的检测性能,结果表明:自主车在城市环境下能以每帧200 ms实时准确地检测环境中的车辆.
|
|
[1] LUETTEL T, HIMMELSBACH M, WUENSCHE H J. Autonomous ground vehicles—concepts and a path to the future [J]. Proceedings of the IEEE, 2012, 100(13): 1831-1839.
[2] MONTEMERLO M, BECKER J, BHAT S, et al. Junior: The stanford entry in the urban challenge [J]. Journal of Field Robotics, 2008, 25(9): 569-597.
[3] ZHAO H, ZHANG Q, CHIBA M, et al. Moving object classification using horizontal laser scan data [C]∥ Robotics and Automation, 2009. ICRA′09. IEEE International Conference on. Kobe, Japan: IEEE, 2009: 2424-2430.
[4] NAVARRO-SERMENT L E, MERTZ C, HEBERT M. Pedestrian detection and tracking using three-dimensional ladar data [J]. The International Journal of Robotics Research, 2010, 29(12): 1516-1528.
[5] WOJKE N, HASELICH M. Moving vehicle detection and tracking in unstructured environments [C]∥ Robotics and Automation(ICRA),2012 IEEE International Conference on. Anchorage, Alask:IEEE, 2012: 3082-3087.
[6] PETROVSKAYA A, THRUN S. Model based vehicle detection and tracking for autonomous urban driving [J]. Autonomous Robots, 2009, 26(2/3): 123-139.
[7] HIMMELSBACH M, LUETTEL T, WUENSCHE H J. Real-time object classification in 3D point clouds using point feature histograms [C]∥ Intelligent Robots and Systems, 2009. IROS 2009. IEEE/RSJ International Conference on. St. louis, Missouri, USA: IEEE, 2009: 994-1000.
[8] AZIM A, AYCARD O. Detection, classification and tracking of moving objects in a 3D environment [C]∥Intelligent VehiclesSymposium(IV).Vilamoura, Portugal: IEEE,2012: 802-807.
[9] HIMMELSBACH M, HUNDELSHAUSEN F, WUENSCHE H. Fast segmentation of 3D point clouds for ground vehicles [C]∥ Intelligent Vehicles Symposium (IV), 2010 IEEE. Taipei, Taiwan :IEEE, 2010: 560-565.
[10] DARMS M, RYBSKI P E, URMSON C. An adaptive model switching approach for a multisensor tracking system used for autonomous driving in an urban environment [C]∥ AUTOREG 2008. Steuerung und Regelung von Fahrzeugen und Motoren: [s.n.], 2008:521-530.
[11] MORI T, SATO T, NOGUCHI H, et al. Moving objects detection and classification based on trajectories of LRF scan data on a grid map [C]∥ Intelligent Robots and Systems (IROS), 2010 IEEE/RSJ International Conference on. Taipei, Taiwan: IEEE, 2010: 2606-2611.
[12] SPINELLO L, ARRAS K O, TRIEBEL R, et al. A layered approach to people detection in 3D range data [C]∥ American Association for Artificial Intelligence(AAAI). Atlanta, Georgia:[s.n.], 2010.
[13] KIDONO K, MIYASAKA T, WATANABE A, et al. Pedestrian recognition using high-definition LIDAR [C]∥ Intelligent Vehicles Symposium (IV), 2011 IEEE. San Francisco, CA: IEEE, 2011: 405-410.
[14] LALONDE J F, VANDAPEL N, HUBER D F, et al. Natural terrain classification using three‐dimensional ladar data for ground robot mobility [J]. Journal of Field Robotics, 2006, 23(10): 839-861.
[15] WILLIAMS D. Methods of experimental physics [M]. Waltham, Massachusetts: Academic Press, 1976.
[16] CHANG C C, LIN C J. LIBSVM: a library for support vector machines [J]. ACM Transactions on Intelligent Systems and Technology (TIST), 2011, 2(3): 27. |
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
|
Shared |
|
|
|
|
|
Discussed |
|
|
|
|