Studying of an Adaptive Unscented Kalman Filter as a Target Tracking Algorithm with Matlab Simulation

Abstract

The radar is considered as one of tracking application which we need to tracking unknown targets. This application is using a kalman filter. Also the extended and unscented kalman filter has been created to deal with nonlinear systems such as multidimensional radar tracking. If wideband radar is used, multiple return signals may be obtained by scattering on a single target in a single pulse. It is important to monitor multiple dispersions on an undefined number of artifacts in the specific problem considered, with a standard wideband radar model, whose motion involves a nonlinear rotational velocity component. The measurements used are scattered range, range-rate (in analogy to Doppler), nonlinear functions of the scattered position and velocity, with parameters in the state-space model. In order to evaluate the scene and observe its progression over time, the tracker design uses a Kalman Filter type to model the nonlinear dynamics and measurement items and a multi-pronged association scheme.