Attitude estimation of MEMS-IMU based on fuzzy robust adaptive CKF algorithm

Aiming at the problem that the traditional filtering algorithm cannot accurately estimate the system attitude when the MEMS inertial measurement unit (MEMS-IMU) is disturbed by state mutation and unknown measurement noise, an attitude estimation algorithm based on fuzzy robust adaptive cubature Kalman filter (FRA-CKF) is proposed. By analyzing the statistical characteristics of the filtering innovation sequence, the correction threshold and the correction boundary are set according to the test principle, the membership functions of CKF, robust correction and adaptive correction are constructed, and the corresponding fuzzy correction criterion is formulated to make the algorithm take into account the self-adaptation and robustness. The simulation experiment and static and dynamic experiments verify the effectiveness of the proposed algorithm. The static experiment results show that the root mean square error of the heading angle estimation results of the proposed filtering algorithm reduces by 80% compared with the CKF algorithm, which improves the accuracy and stability of the filtering. © 2022, Editorial Department of Journal of Chinese Inertial Technology. All right reserved.