Robust predictive control of ship steering system considering time delay and property variation

It is very important to consider course control of ship steering system in order to construct the autopilot with high precision and safety. In ship steering system, it is difficult to control well for large time delay, limits of rudder angle and rudder angle velocity, variation of ship motion property depending on many changes such as ship velocity or loading condition and disturbance of the wind and the waves. Therefore, it is necessary to design the control system considering existence of dead time, model error and saturation and so on. In this paper, a control system is designed paying attention to time delay namely dead time and ship motion property variation namely model error among these problems. For the dead time and the model error, H-infinity predictive control to guarantee robust stability is applied especially. And, for the prediction of plant output, a predictor of Smith type guaranteeing robust predictive performance is used. It is shown that the designed control system is effective by execution of numerical simulation.