With the increasing development of high-resolution sensor techniques, extended target tracking (ETT) attracts more and more attention in civilian and military fields. In order to track an irregular-shaped extended target with a large number of measurements per scan in heavy clutter, a cost-effective Gaussian processes (GPs) based probabilistic data association (CGP-PDA) algorithm is proposed in this article. It consists of the following enhancements. First, a generalized measurement-to-contour association algorithm is proposed to preselect the validated measurements that are associated with the target contour. This algorithm mitigates the influence of clutter within the validation gate on the accuracy of target states estimation. Second, based on the above association algorithm and the Fisher information matrix (FIM)-based submodular function, a measurement selection method is proposed to choose the informative measurements. These measurements are defined to provide the primary information for target states estimation. This selection method greatly limits the scale of data association events and, hence, increases the real-time performance of the filter. Finally, the computational complexity of the CGP-PDA together with that of GP-PDA is investigated in detail. The floating-point operations (FLOPs) are invoked to analyze the computational complexity of the above algorithms, and the lower/upper bound on the relative difference of FLOPs between CGP-PDA and GP-PDA is analyzed. The effectiveness of the proposed algorithm is evaluated in simulation.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Cost-Effective Gaussian Processes Based Extended Target Tracking


    Contributors:


    Publication date :

    2023-12-01


    Size :

    2748623 byte




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English



    Extended Target Tracking using a Gaussian-Mixture PHD Filter

    Granstrom, K. / Lundquist, C. / Orguner, O. | IEEE | 2012


    Learning Target Dynamics While Tracking Using Gaussian Processes

    Veiback, Clas / Olofsson, Jonatan / Lauknes, Tom Rune et al. | IEEE | 2020



    Corrections on: “Extended Target Tracking Using a Gaussian-Mixture PHD Filter”

    Granstrom, Karl / Orguner, Umut / Mahler, Ronald et al. | IEEE | 2017

    Free access