An Extended Kalman Filter Application on Moving Object Tracking

In this paper, the problem of moving object tracking on 2D plane is addressed by combining uncertain information from measurement of the object to accurately estimate its trajectory. Due to the nonlinear motion model of the tracked moving object, the extended Kalman filter technique (EKF) is applied. In particular, the models of object motion and measurement including noise are established. After substituting those models to the equations of EKF, an optimal estimated trajectory can then be rendered that stays as close to the expected one. An example is given to perform the process of EKF algorithm. Simulation results with Monte Carlo simulation are shown to verify the validity of the EKF in solving the moving object tracking problem.