With the development of autonomous driving and intelligent transportation system, the outline size detection of a vehicle has been very critical in the field of road traffic. In our paper, we propose a vehicle width detection based on millimeter-wave frequency-modulated continuous wave radar for autonomous and assisted driving. First, the three-dimensional fast Fourier transform is performed on radar echo signals to get the range-velocity-angle (RVA) spectrum. Then, Order Statistic constant false alarm rate (OS-CFAR) is employed to obtain the information of vehicle scatters. To improve the angular resolution, the root multiple signal classification method is used. Finally, to improve measurement accuracy and reduce the complexity in training effectively, we propose a random forest method based on important features selection so that the width of vehicles can be calculated more precisely. Numerical results derived from real-world datasets show that the proposed calculation scheme of vehicle width achieves good accuracy.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Vehicle Width Detection Based on Millimeter-Wave LFMCW Radar for Autonomous Driving


    Contributors:
    Wang, Qiang (author) / Xu, Shaoyi (author)


    Publication date :

    2022-06-01


    Size :

    1769241 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Back-Scattering of a Helicopter with a Millimeter-Wave LFMCW Radar

    Garcia-Fernandez, Angel F. / Grajal, Jesus / Yeste-Ojeda, Omar A. | IEEE | 2013



    A Research on Millimeter Wave LFMCW Radar for Airfield Object Imaging

    Yang, X. / Yang, J. | British Library Online Contents | 2001


    Detection and Estimation of LFMCW Radar Signals

    Geroleo, F. G. / Brandt-Pearce, M. | IEEE | 2012


    Models and Signal Processing for Millimeter-Wave LFMCW SAR Imaging

    Wang, Wen-Qin / Cai, Jingye / Peng, Qicong | IEEE | 2008