An Unscented Kalman Filter based wave filtering algorithm for dynamic ship positioning

Abstract

A dynamic positioning (DP) system is used to maintain a ship on a specified position and at a desired heading. A ship is exposed to the wind, currents and waves in the ocean. In order to counteract these disturbances, the dynamic positioning ship uses solely active propulsion. Since the first-order wave force is too large for the ship's thrust system to deal with and it only causes the high-frequency motion, which do not induce the finial position drift. A filtering algorithm is needed to separate the high frequency motions from the low frequency ones. In this paper the Unscented Kalman Filter (UKF) method is used for the wave filtering of dynamic positioning ships. Noisy position and heading measurements are used by the UKF to estimate the low-frequency motion and the high-frequency motion. A nonlinear platform supply vessel (PSV) model is employed for both the model simulation and the estimator design. An approximation method for the first and second order wave force calculation is used to produce the high frequency and low frequency ship motion. With considering of the wind, currents and waves, some illustrative results of the UKF based wave filtering algorithm are presented.