Design of Ship Controller and Ship Model Based on Neural Network Identification Structures

This paper proposes computationally efficient artificial neural network models for identification of both fuzzy logic ship controller and nonlinear ship model. The first objective demonstrates how to use a nonlinear network to identify the fuzzy controller and compare control surfaces of these two controllers as well as performance indices. The second objective is to use the nonlinear network to identify nonlinear plant in recursive on-line mode and the third one is to integrate designed two neural networks in one control scheme to test resulting system response in the closed loop system.