An Application of the Finite Set Statistics to Space-based Multi-object Tracking

Abstract

This thesis proposes a multi-target tracking algorithm based on the Gaussian mixture probability hypothesis density (GM-PHD) filter and the unscented Kalman filter (UKF). Unlike existing GM-PHD filters that assume the presence of birth, death, or spawning events, this algorithm does not require such events to occur. It is, however, compatible with their addiction. The algorithm utilizes a set of time-varying weights to adaptively update the predicted target states and their corresponding uncertainties. The effectiveness of the algorithm was tested through simulation experiments on synthetic data. The results show that the proposed algorithm achieves appreciable tracking performance while requiring fewer assumptions and parameters. Extensions of this algorithm have potential applications in various fields, such as surveillance, robotics, and autonomous vehicles.