Nonlinear control of an autonomous underwater vehicle: A RISE-based approach

Autonomous and remotely operated marine vehicles such as ships and submarines are becoming a key component in several aspects of maritime industry and defense. This paper explores the development of a nonlinear controller for a fully actuated autonomous underwater vehicle (AUV) using a robust integral of the sign of the error (RISE) feedback term with a neural network (NN) based feedforward term to achieve semi-global asymptotic tracking results in the presence of complete model uncertainty and unknown disturbances. A simulation is provided to demonstrate the proposed controller on an experimentally validated AUV model.