Robust Steering Autopilot Design for Marine Surface Vessels

In this paper, a novel design based on an extended state observer (ESO) technique for yaw control of marine surface vessels is proposed. A marine vessel by virtue of its operation at sea is inherently under the influence of uncertainties and disturbances. Uncertainties arise because of parametric variation in hydrodynamic coefficients due to trim, loading, ballast, speed, and changes in underwater depth, where external disturbances due to winds and waves are present. The uncertainties and disturbances pose a great challenge in designing an effective autopilot for such vehicles. In this work, a robust design for steering autopilot is proposed wherein the effect of uncertainties and disturbances is estimated using ESO and the estimate is used to augment an input-output linearization (IOL)-based controller designed for nominal system to achieve robustness. The notable feature of the design is that while achieving robustness, it neither requires accurate plant model nor any information about the uncertainties and disturbances. The effectiveness of the ESO in estimation of the total disturbance and states and in meeting the specified performance in the presence of significant uncertainties, disturbances, and measurement noise is illustrated by simulation. Further, to demonstrate its efficacy, performance comparison of the proposed controller with some well-known existing designs in literature is carried out and the results are presented.