Model-based controller for ship track-keeping using neural network

In the paper the Internal Model Control approach for ship autopilot system is presented. The proposed course controller employs the structure of the cascade system. The internal model of the plant and its inverse are estimated by neural network what made it possible to reduce the uncertainty of the control process. The ship model contains the saturation of the rudder angle and the rudder rate. Therefore to the controller design the structure with feedback connection is used. Computer simulation results are included in the paper to demonstrate the effectiveness of the proposed method.