|
|
Real-time dynamic obstacle detection and tracking using 3D Lidar |
YANG Fei, ZHU Zhu, GONG Xiao-jin, LIU Ji-lin |
Department of Information Science and Electronic Engineering, Zhejiang University, Hangzhou 310027, China |
|
|
Abstract In order to detect and track obstacles under large amount of data efficiently, an approach for real-time multiple obstacle detection and tracking in dynamic unknown environment was presented. The Velodyne 64E 3D Lidar has the property of large amount of data and high accuracy, which was combined with camera for environment perception. The algorithm firstly coverts the region of interest of the Lidar data into a grid map according to road lane information obtained from image processing, then uses region labeling and template matching to detect box-model obstacles on the grid map, and finally tracks the obstacles. In order to avoid false alarm or miss matching, multiple hypothesis tracking and Kalman filter were used for obstacle tracking. The approach can detect obstacles accurately and track stably within 100 ms per frame on the autonomous vehicle.
|
Published: 01 September 2012
|
|
基于三维激光雷达的动态障碍实时检测与跟踪
为了解决在大数据量的情况下实现高效检测与跟踪的难点,提出一种室外动态未知环境下自主车的多障碍实时检测与跟踪的算法.由于Velodyne 64线三维激光雷达具有数据量大、精度高等特点,采用其与相机结合感知环境.算法结合从图像处理中得到的道边信息将原始激光雷达数据的感兴趣区域转化为栅格地图,在地图上采用区域标记和模板匹配的方法进行聚类和特征提取,检测得到盒子模型的障碍物,并进行障碍物跟踪.为了避免在多障碍物的情况下出现虚警和漏检,基于多假设跟踪数据关联和卡尔曼滤波来跟踪连续多帧的障碍物.本算法在自主车平台上能够以每帧100 ms实现准确、稳定地检测和跟踪.
|
|
[1] VU T D, BURLET J, AYCARD O. Gridbased localization and online mapping with moving objects detection and tracking: new results[C]∥Intelligent Vehicles Symposium. Eindhoven: IEEE, 2008: 684-689.
[2] PETROVSKAYA A, THRUN S. Model based vehicle detection and tracking for autonomous uban diving[J]. Autonomous Robots, 2009, 26:123-129.
[3] DARMS M, RYBSKI P E, BAKER E, et al.. Obstacle detection and tracking for the urban challenge[J]. Intelligent Transportation Systems, 2009, 10(3): 475-485.
[4] DARMS M, RYBSKI P E, URMSON C. Classification and tracking of dynamic objects with multiple sensors for autonomous driving in urban environments[C] ∥Intelligent Vehicles Symposium. Eindhoven: IEEE, 2008: 1197-1202.
[5] BLACKMAN S S. Multiple hypothesis tracking for multiple target tracking[J]. Aerospace and Electronic Systems Magazine, 2004, 19(1): 5-18.
[6] LEONARD J, HOW J, TELLER S, et al. A perceptiondriven autonomous urban vehicle[J]. Springer Tracts in Advanced Robotics, 2009, 56: 163-230.
[7] KONSTANTINOVA P, UDVAREV A, SEMERDJIEV T. A study of a target tracking algorithm using global nearest neighbor approach[C]∥ International Conference on Computer Systems and Technologies. Sofia:Citeseer, 2003: 290-295.
[8] HE Lifeng, CHAO Yuyang, SUZUKI K, et al. Fast connectedcomponent labeling[J]. Pattern Recognition, 2009, 42(9): 1977-1987.
[9] TUBBS J D. A note on binary template matching[J]. Pattern Recognition, 1989, 22(4): 359-365.
[10] FERGUSON D, DARMS M,URMSON C, et al. Detection, prediction, and avoidance of dynamic obstacles in urban environment[C]∥ Intelligent Vehicles Symposium. Eindhoven: IEEE, 2008: 1149-1154.
[11] REID D B. An algorithm for tracking multiple targets[J]. Automatic Control, 1979, 24(6): 843-854.
[12] WELCH G, BISHOP G. An introduction to the Kalman filter[R]. North Carolina, USA: University of North Carolina, 1995. |
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
|
Shared |
|
|
|
|
|
Discussed |
|
|
|
|