Method of improving central difference Kalman filter

In order to improve tracking estimation accuracy of existing Central Difference Kalman Filte(rCDKF),a new itera-tive central difference kalman filter is proposed.In this paper,iterative filtering theory is introduced into the extended kalman filter algorithm,and observation information is reused.Taking the classic non-linear and non-gaussian model for example,sever-al simulation experiments are given by using the algorithm such as extended kalman filter(EKF),Central Difference Kalman Filter,and iterative central difference kalman filter.In comparison with the tracking performance and root mean square error,iterative central difference kalman filter(ICDKF) algorithm not only has no need to calculate Jacobian matrix,but also has a higher estimation accuracy.