Variable Structure Control of Ship Course Nonlinear System Based on Radial Basis Function Network Adjustment

摘要

A sliding mode variable structure controller basing on radial basis function network adjustment is proposed for nonlinear ship course control system. The cascaded system consists of nonlinear ship response model and rudder actuator dynamics. The response model takes many uncertainties into account, including modeling errors and external disturbances and so on. A ship course sliding mode controller is designed, and the radial basis function neural network, which has strong chattering, is introduced to adjust the controller switching gain. The switching function is regarded as the input of radial basis function neural network, while the output is the absolute value of the switching gain. The simulation results demonstrate the validity, satisfactory adaptation and robustness of the proposed controller.