Line of sight rate estimation of strapdown imaging guidance system based on unscented Kalman filter

Models of the strapdown imaging guidance system are derived and corresponding estimation problem is considered in this paper. In comparison with the conventional imaging guidance systems, there are high nonlinearity in both process and measurement models, and measurement is more seriously corrupted by noise since wider instantaneous field of view for the stapdown imaging seeker. Considering these properties, the unscented Kalman filter (UKF) is applied to estimate line of sight (LOS) rate. The UKF propagates statistics of random variable more accurately than the extended Kalman filter (EKF) does for nonlinear system. Furthermore, the UKF avoids calculating Jacobian matrices which are complicated usually, and sometimes singular and thus in feasible for the EKF. At the last, Monte Carlo simulations are performed, and results show that the UKF is superior to the EKF for the strapdown imaging guidance system.