A High-Gain Observer-Based Approach to Robust Motion Control of Towed Underwater Vehicles

This paper proposes a robust motion control method for towed underwater vehicles (TUVs) with movable wings. TUVs are inherently under influences of nonlinearities and uncertainties due to the flexible towing cable, hydrodynamic forces, parametric variations, and external disturbances. In this paper, the cable is approximated by the lumped-mass method so that the number of the cable segments decides the order of the system. A robust control design for TUVs is presented, wherein the states are estimated by a high-gain observer designed for the lowest order system and the estimate is used by a linear-quadratic-integral-based controller to make use of robustness of the scheme. A direct consideration of the nonlinear dynamics is one of the main features of this paper. Simulations comparing the high-gain observer-based controller and a linear Kalman filter based one from several viewpoints are carried out to evaluate the proposed method and to investigate differences from a linear-based controller. Furthermore, the same controllers are applied to higher order systems and it is also illustrated by simulations that the proposed controller is satisfactorily robust to control the system under uncertainties even for higher order systems.