Improvement and experimental verification of T‐S fuzzy‐based model predictive control for a zero‐to‐low speed fin stabilizer

Abstract Roll stabilization is an important topic for improving the sea‐keeping capabilities of ships. As the most effect anti‐rolling equipment, fin stabilizers are broadly used to reduce ship rolling. Nevertheless, the time‐varying uncertainties and nonlinearities in the rolling model often cause the anti‐rolling effectiveness of actual ships to be lower than that predicted by theoretical analyses or simulations. To address the limitations of our previous control strategy, the nonlinear rolling dynamics are re‐expressed as a Takagi‐Sugeno fuzzy model with associated uncertainties. Moreover, the model predictive control (MPC) method is employed to derive the optimal control series. The feasibility and stability of the MPC are analyzed, and a nonlinear extended state observer is utilized to estimate the roll parameters. Lastly, an environmental survey vessel is employed to verify the improved control strategy through both simulations and experiments.