Attitude Angle Estimation Based on Iterative LM-CDKF Algorithm

In order to estimate the attitude angle accurately, improve the accuracy of attitude and heading reference system(AHRS), aiming at larger errors of extended Kalman filter(EKF) in estimating attitude angle, a method based on iterative LM-CDKF algorithm has been proposed to estimate the attitude angle. In the paper, iterative filtering theory and Levenberg-Marquardt method is introduced into the central differential Kalman filter algorithm, which apply the method of quaternion to describe the attitude angle of carrier. The mathematical model of the attitude angle is established. The MATLAB Simulink tools are used to simulate the algorithm. The results of iterative LM-CDKF algorithm are compared to the those of the central differential Kalman Filter algorithm, and found to have excellent convergence and the better accuracy of the attitude angle.