Multiple moving target tracking with hypothesis trajectory model for autonomous vehicles

Detecting and tracking moving objects is a key technology for autonomous driving vehicle in dynamic urban environment. In this paper, we present a hybrid system of multiple moving target tracking (MMTT) for autonomous driving vehicles using 3D-Lidar sensor. To detect targets, geometric model-based method is adopted in the hybrid system. In further, we propose a novel hypothesis trajectory model to detect moving targets. The targets detected by the two methods are fused under Extend Kalman Filter. Besides, a bipartite graph model-based method with the Hungary algorithm is presented to optimize data association matrix. Our tracking algorithm is tested on school road and Beijing's 3rd ring road using an experimental vehicle with velodyne-32-E. And, the experiment results illustrate the hybrid method performs well in real time.

[1]  Yaakov Bar-Shalom,et al.  Sonar tracking of multiple targets using joint probabilistic data association , 1983 .

[2]  Ramakant Nevatia,et al.  Robust Object Tracking by Hierarchical Association of Detection Responses , 2008, ECCV.

[3]  Zhu Zhu Gong Xiao-jin Liu Ji-lin Yang Fei.  Real-time dynamic obstacle detection and tracking using 3D Lidar , 2012 .

[4]  Olivier Aycard,et al.  Detection, classification and tracking of moving objects in a 3D environment , 2012, 2012 IEEE Intelligent Vehicles Symposium.

[5]  S.S. Blackman,et al.  Multiple hypothesis tracking for multiple target tracking , 2004, IEEE Aerospace and Electronic Systems Magazine.

[6]  Raja Sengupta,et al.  Kalman Filter-Based Integration of DGPS and Vehicle Sensors for Localization , 2005, IEEE Transactions on Control Systems Technology.

[7]  William J. Christmas,et al.  Ieee Transactions on Pattern Analysis and Machine Intelligence 1 Layered Data Association Using Graph-theoretic Formulation with Application to Tennis Ball Tracking in Monocular Sequences , 2022 .

[8]  Sebastian Thrun,et al.  Model based vehicle detection and tracking for autonomous urban driving , 2009, Auton. Robots.

[9]  Paulo Peixoto,et al.  3D object tracking in driving environment: A short review and a benchmark dataset , 2016, 2016 IEEE 19th International Conference on Intelligent Transportation Systems (ITSC).

[10]  Markus Maurer,et al.  Multi-Target Tracking using a 3D-Lidar sensor for autonomous vehicles , 2013, 16th International IEEE Conference on Intelligent Transportation Systems (ITSC 2013).

[11]  Simone Calderara,et al.  Visual Tracking: An Experimental Survey , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[12]  Tzvetan Semerdjiev,et al.  A study of a target tracking algorithm using global nearest neighbor approach , 2003, CompSysTech '03.

[13]  Ingemar J. Cox,et al.  On Finding Ranked Assignments With Application to Multi-Target Tracking and Motion Correspondence , 1995 .