The use of artificial neural networks in the design of fuzzy autopilots for ships

The problem addressed is that of the design of a fuzzy autopilot for ship yaw control using artificial neural networks. The algorithms of temporal backpropagation, Alopex and Chemotaxis, are utilised as methods of fusing the techniques of fuzzy logic and artificial neural networks into intelligent controllers. The results presented demonstrate the viability of the approach and that effective designs with improved performance characteristics can be produced.