Abstract:  The centered error entropy Kalman filter algorithm has strong robustness under nonGaussian noise，but it still faces the challenge of how to choose the weight coefficient.To solve this issue，an adaptive centered error entropy Kalman filter algorithm with a variable weight coefficient was proposed.The weight coefficient is adaptively adjusted according to the error vector，which increases the sensitivity of the cost function to the error and improves the filtering accuracy.By applying to relative navigation of formation satellites，the simulation results demonstrate that the proposed algorithm outperforms the Kalman filter and the centered error entropy Kalman filter algorithms when dealing with the state estimation problem in the linear non-Gaussian system.A higher filtering accuracy and a stronger ability to suppress non-Gaussian noise are presented.

Key words:  , Kalman filter；adaptive centered error entropy；non-Gaussian noises；relative navigation；state estimation