State Estimation based Adaptive Extended Kalman Filter for Moving Object Tracking with Angle-only Measurements

This article deals with an application of adaptive extended Kalman filter for state estimation in moving object tracking problem with angle-only measurements. The nonlinear two-dimensional model of observer-moving object engagement in modified polar coordinates is described with taking maneuver of moving object into consideration. The next section presents the mathematical approach to synthesize algorithm of both extended Kalman and adaptive extended Kalman filters for nonlinear state estimation. In this work, designing and testing of state filters are carried out in MATLAB/Simulink environment. The simulation results indicate that the proposed adaptive extended Kalman filter improves accuracy and stability of estimated parameters of moving object tracking in compare with extended Kalman filter.