Ship Dynamic Positioning Control Based on Inverse Disturbance Observer
ZHANG Hanlin
DENG Fang
DING Qiang
YANG Hualin
Abstract：In this paper,a disturbance observer based on inverse recurrent neural network is proposed to estimate the distur-bance of wind,wave and current environment during the ship sailing process.The disturbance estimates are then applied to the dy-namic positioning(DP)control system to eliminate the effect of disturbance.Based on the principle of data-driven and inverse sys-tem,the inverse recurrent neural network model of the DP system is established by using the"thrust input-position and heading output"data of the ship,and then the inverse disturbance observer is established.The"input-output"data set of ship motion are ob-tained via maneuvering simulations.Through training and testing,the established inverse recurrent neural network based distur-bance observer can accurately estimate the random environmental disturbances.Taking the disturbance estimates as the system's feedforward compensation.The DP system can perform well in trajectory tracking control.The proposed method provides a feasible strategy for the robust anti-disturbance control of motion control systems.
Keywords：ship dynamic positioninginverse systemrecurrent neural networkinverse disturbance observer
Publication Date：2025-09-20
Online Publishing Date：2025-12-23(First online date of this platform, not the publication date of the document)
Pages：6( 2422-2427 )