Dynamic inverse control of ship rudder roll/yaw stabilization based on neural network disturbance observer
摘要
This study proposes a dynamic inversion control method for ship rudder roll/yaw stabilization based on a Neural Network Disturbance Observer (NNDO). First, a nonlinear mathematical model considering wind and wave disturbances for a four-degree-of-freedom ship is established. Second, to address model uncertainties and stochastic disturbances, an NNDO combining Radial Basis Function (RBF) neural networks and sliding mode control theory is designed. The NNDO is utilized to estimate the system state functions and external disturbances. Third, accounting for the strongly nonlinear and coupled nature of the ship's rudder roll/yaw stabilization system and the characteristics of the rudder mechanism, an NNDO-based state feedback dynamic inversion controller is developed. Finally, the bounded stability of the closed-loop control system is proven using the Lyapunov stability theorem. Simulation results under severe sea conditions demonstrate the effectiveness of the designed controller in controlling roll and yaw motions.