Adaptive Cubature Kalman Filter Algorithm Based on Quaternion Error Modeling

Abstract

Initial alignment is the key link before initial device enters navigation function. In the actual use of the marine inertial navigation system, the external environment is complex, the error modeling cannot be simply approached and processed by linear filtering. To solve this problem, the adaptive CKF algorithm based on quaternion error modeling is proposed. The proposed method uses multiple fading factors to redistribute the weight of measurement information, so as to reduce the algorithm error caused by inaccurate noise parameters in the complex environment. Simulation results show that the adaptive CKF algorithm based on quaternion error modeling proposed in this paper can solve the large misalignment angle transfer alignment. Compared with UKF and CKF algorithm, when the system noise covariance changes, the proposed algorithm can effectively improve alignment precision and accuracy.