Positioning controller of a remotely operated vehicle using RWNN-based adaptive robust controllers

In this paper, a robust adaptive controller is designed for the dynamical positioning of a remotely operated vehicle (ROV) with unknown hydrodynamic coefficients and disturbances. Firstly, the Multiple Input Multiple Output (MIMO) system can be divided into several Single Input Single Output (SISO) systems for convenience of controller design. Then, a recurrent wavelet neural network (RWNN) with feedback loops is employed to approach the unknown part of the controller, and it can greatly reduce the number of neural network´s inputs. Though the upper bounds of the network´s approximation errors and disturbances are often used in the existed work, they are not necessary in this paper by presenting a special robust compensator. The stability analysis is given by Lyapunov theorem, and the tracking errors are proved to convergence to zero. Simulation result are presented for the ROV´s positioning system, which indicate that the proposed controller has a good performance.