Random Matrix-Based Group Target Tracking Using Nonlinear Measurement

Random matrix (RM) can estimate the dynamic and extension state jointly for group target (GT) tracking. In this paper, to improve the tracking performance for the rotation, the rotation matrix is added to the original random matrix in group target tracking algorithm based on variational Bayesian (VB) approach for nonlinear measurement. Different from the existing method, the model of nonlinear measurement is established, where the prediction and update process are given. Simulation results show that the proposed method have better performance in the estimation of group state and extension than the original method.