Neuro-adaptive trajectory tracking control of underactuated autonomous surface vehicles with high-gain observer

This paper addresses the trajectory tracking control of underactuated autonomous surface vehicles subject to immeasurable velocities and parameter uncertainties. An adaptive control scheme is proposed based on backstepping method, neural network and low-frequency techniques. In addition, the overall signals of closed-loop system are ensured uniformly ultimate boundness based on Lyapunov stability theory. The advantages are highlighted as follows: (i) To avoid explosion of complexity of conventional backstepping method, the derivations of virtual control signals are obtained through second-order tracking differentiator. (ii) The proposed controller does not require any previous knowledge about hydrodynamic damping and external disturbances, which is easily applied into practice. (iii) To solve the problem of immeasurable velocities, the controller does not acquire velocity signals by employing high-gain observer. Finally, simulation results are provided to verify strong robustness and tracking effectiveness of proposed control scheme