Application of optimized RBF neural network in ship's autopilot design

Intelligent autopilot is designed for ship on the basis of Radial Basis Function neural network optimized by Genetic Algorithms. In consideration of the nonlinearity and uncertainty of procedure for Ship course control design, RBF network is used to approach the ship's inside uncertainties and external disturbances, and subsequently the ship course control law is designed by Lyapunov theory. Genetic Algorithms is used to optimize the Radial Basis Function neural network output weights, width and center of hidden units so as to improve the performance of network approaching and suppress input chattering of system. Simulation results show that the designed controller is faster 40%, lower overshoot is reduced 100%, and the output is not sensitive to internal and external interference, comparing with adaptive control algorithm and fuzzy PID control algorithm, under the same conditions.