General fuzzified CMAC-based model reference adaptive control for ship steering using float-encoding genetic algorithm

Abstract

The application of CMAC control techniques for improving the performance of ship steering is discussed. Aiming at a conventional Cerebellar Model Articulation Controller (CMAC), combining CMAC-addressing schemes with a fuzzy logic idea, a general fuzzified CMAC (GFCMAC) is proposed, in which the fuzzy membership functions are utilized as the receptive field functions. Incorporating GFCMAC into a control system, a GFCMAC-based model reference adaptive controller is described, whose required teacher signals are explicitly constructed by a reference model. In order to improve the control performance, using a float-encoding genetic algorithm (FGA) to optimize parameters, an FGA-GFCMAC controller is discussed. Applying the FGA-GFCMAC controller to a ship steering control system, the simulation results show that the ship’s course can be properly controlled when changing wind and waves exist.