An Asymptotically Stable Identifier Design for Unmanned Surface Vehicles Based on Neural Networks and Robust Integral Sign of the Error.

In this paper, a robust identifier is developed for unmanned surface vehicles (USVs) subject to uncertain dynamics. The uncertain dynamics comes from parametric model uncertainty and external ocean disturbance. The identifier for USV is designed based on Robust Integral Sign of the Error (RISE) and neural networks. With the proposed identifier, asymptotic stability of the estimation errors can be proven in the presence of parametric model uncertainties and external ocean disturbances. The proposed method can be used in a variety of practical settings such as trajectory tracking and formation control of marine vehicles for achieving better performance.