Multiple fading Kalman filter based on hypothesis testing

A multiple fading Kalman filtering algorithm was proposed based on hypothesis testing. The algorithm on-line modifies the noise covariance of the Kalman filter by the fading matrix using chi-square testing method so as to improve the accuracy and convergency of the Kalman filter when there are errors or the plant is affected by unmeasurable external disturbance. The proposed algorithm was applied to a vehicle global positioning system/inertial navigation system/integrated navigation system and behaved satisfactorily.