Neurofuzzy control of a nonlinear multivariable system

This paper discusses the development of a new multivariable autopilot for simultaneous control of the sway and yaw modes of an autonomous underwater vehicle (AUV) and highlights the requirement for the minimization of cross-coupling effects between these modes. An adaptive network structure has been developed and applied to the task of tuning a novel multivariable fuzzy autopilot. This paper discusses the design and tuning of novel multivariable Sugeno fuzzy autopilot for an AUV. Although the proposed tuning method employs a neural network architecture to tune the parameters of the autopilot, the resulting autopilot is purely fuzzy.