PID – NEURON CONTROLLER FOR SHIPS HEADING WITH NEURON INDENTIFICATION

Abstract

This paper presents new results of PID neural network controller for heading control of ship. The control system uses one neural network to drive parameters of a PID controller. Another neural network is used to predict the ship’s heading so that it gives the predicted heading of k steps in advance. Output of the second neural network is sent to the input layer of the first one, that produces again the PID controller’s parameters. We carry out computer-based simulations to see how the proposed control algorithm performs. It is shown by the simulation results that the controller can perform heading control task well employing the advantages of both neural networks and traditional PID.