Heave Motion Estimation Based on Cubature Kalman Filter

Abstract

To solve the high-dimensional nonlinear problem of the ship heave motion model, a cubature Kalman filter (CKF) is used to improve the estimation accuracy of the nonlinear filter. The mathematic model of ship heave motion is established based on the Longuet Higgins wave model and the accelerometer measurement model. The fast fourier transform (FFT) is used to analyze the acceleration information. Because of the non-linearity of the heave motion model and the measurement noise and zero bias existing in the inertial measurement unit (IMU), CKF is used to estimate the heave motion. The proposed method is evaluated with simulation and measurement results from an experimental setup. A six-degree-of-freedom motion platform is used for experimental verification. The experimental results show that the heave motion estimation based on CKF has a faster convergence speed and a more accurate estimation accuracy than the unscented Kalman filter algorithm (UKF). The mean square error of the heave motion estimation reaches 0.008m, it can obtain accurate and no-delay heave motion information.