Research On Application Of Nonlinear Filter In Dynamic Positioning

Ship dynamic positioning technology is an essential support system for ships that work in the ocean at the setting point for a long period. It can maintain the position and heading of the ship, have tracking function, and is not limited by depth. Filtering and estimating are the important phase in DP. The normally work of DP depends on the measurement of the ship position. Now most DP ships use GPS to measure their positions. Sometimes, the DP ship at work may lose GPS by some reasons, this may lead to the deviant operation of the DP system and make the ships drift away from the setting point. In order to maintain the ship's position near the operation point in a given period after losing the GPS, and when regain the GPS, it can return to the setting point, this paper presents a solution:Firstly, based on the thinking of separate ship modeling, we get 3DOF mathematical model of the DP ship from its manipulation formula. According to the randomicity of environmental disturbances, this paper presents an environmental disturbances model base on stochastic process. Base on the ship and environmental disturbances model, we get the integrated model of the ship.Then, in order to solve the problem of losing GPS, this paper uses the filter and history data solution, and processing the history to solve some problems of this solution.After comparing several nonlinear filters, this paper choses the new passive nonlinear filter and commonly used centeral differenced Kalman filter (CDKF).At the end of this paper, we make some simulations on the nonlinear filters in Marine System Simulator (MSS). We have normal operation simulation, no filter simulation, with filter but no processed history data simulation, and with filter and processed history data simulation. The simulation results show that, in the last case, there was an obvious improvement in estimating the real time position of the ship.After analyzing the simulation results, we can see that the nonlinear filters designed can maintain the ship's position near the setting point with the bias about 10 meters in half an hour after losing the GPS. When regaining the GPS, the ship can return to the setting point quickly. This means that the design of the nonlinear filter hits the target.