Design of a hybrid controller for heading control of an Autonomous Underwater Vehicle

The unstructured sea environment and the uncertainty of AUV (Autonomous Underwater Vehicle) model make a hard difficulty to AUV heading control. PID controller has good robustness and high reliability characteristics for that can get accurate model of definite system in math. But PID controller is not satisfactory for the system with inaccurate and nonlinear model in unstructured sea environment. To solve such problems, a hybrid controller is proposed to deal with AUV heading control in this paper. This controller combines the good qualities of both neural network controller and PID controller in order to make maximum profit from both techniques. The adaptability and robustness of this control system is investigated by theory analysis and computer simulation. The results show the better performance of using hybrid controller to control the vehicle heading compared with PID controller, under the conditions with disturbance and unpredictable changes of the vehicle and its environment.