Vehicle tracking is an essential topic in autonomous driving. Currently most systems rely on radars and lidars to perform vehicle tracking. In this paper, we present a novel cross traffic vehicle tracking system with several unique contributions. First of all, it employs a state machine to manage the life cycles of particle filters, resulting in higher tracking robustness. Secondly, the entire software framework is designed to be extensible to support multiple sensors and tracking algorithms. Lastly, we implemented a sensor-vehicle co-simulator to evaluate the tracking performance. We show through experiments that our vehicle tracking system can track multiple vehicles up to 170m away with less than 1m average positional error. We also show that our proposed state machine improves tracking rate under frequent occlusion.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    A State Machine-based Multi-Vehicle Tracking Framework with Dual-Range Radars


    Contributors:
    Huang, Jiawei (author) / Ma, Lichao (author)


    Publication date :

    2018-06-01


    Size :

    2261817 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    A STATE MACHINE-BASED MULTI-VEHICLE TRACKING FRAMEWORK WITH DUAL-RANGE RADARS

    Huang, Jiawei / Ma, Lichao | British Library Conference Proceedings | 2018


    Steady-state tracking with FMCW radars

    McDonough, Michael / Blair, W. Dale | IEEE | 2018



    Range Instrumentation Radars

    Nessmith, Josh T. | IEEE | 1976