Autopilots are important steering equipment for ships. Track keeping is essential for the safe and economical navigation. Because of the complexity of the ship's movement during navigation, it is very difficult for a normal control method to achieve good effect. this dissertation focuses on the methods of fuzzy logic and NN, and their application in shipping track keeping.The origin, history and latest research status in the corresponding field both in China and the world are reviewed and summerized. The mathematics models of shipping control and steering are demonstrated, and the controlling mechanism and steerability of the models are explored. In this dissertation FNN indirect track contoller is proposed, and parameters of FNN track controller are optimized through GA, which reduces the effect of the change of the ship's parameters and the enviorment to controlling performance. Finally, author designs a simulation program by MATLAB/SIMULINK. In the simulation process, author makes comparisons between PID track controller and FNN indirect track controller. In this stage, author gives a signal to the ship model, and the signal simulates the wave and current disturbance. Comparing the response curve proved that FNN track controller has advantages of both fuzzy logic and NN, which improve the features of shipping track controller greatly, thus the track will be quite precise and robust. |