A robust adaptive controller for a two-input-two-output nonlinear system using neural networks

Abstract

This paper presents a robust adaptive nonlinear control scheme for diving behavior of an autonomous underwater vehicle (AUV), whose diving dynamics could be expressed as a two-input-two-output (TITO) second-order nonlinear function. Because of the singularity problem, above dynamics could not be properly solvable using general backstepping method, although it is in a well-known strict-feedback form. Further, the dynamics is in an interacting form, so traditional non-interacting control methods also could not be directly applied. In this paper, we solve above TITO second-order nonlinear dynamics using a backstepping similar method, where the desired value of one of two medium state variables (also called as virtual control in general backstepping scheme) is derived from predefined desired trajectory of the vehicle instead of the stability point of view such that the singularity problem could be avoided. Proposed neural network adaptive control scheme can guarantee all of the signals in the closed-loop system are uniformly ultimately bounded (UUB). ？ 2005 SICE.