ANFIS-based course-keeping control for ships using nonlinear feedback technique

Course keeping for ships is the core of automatic navigation in marine technology. A nonlinear Nomoto model and a maneuvering model group (MMG) model of Yupeng ship are established and verified by the turning trial at sea, then an adaptive neuro-fuzzy inference system (ANFIS) controller is trained by learning the actual ship trial data. There is a limit to the achievable performance of ANFIS controller as the structure is fixed in the training process, many researchers pursue advanced control strategies to improve performance. In this research, the performance is improved in another way, it modulates control error using proposed nonlinear feedback scheme. The simulation result shows that the settling time of nonlinear controllers decreases considerably, dropping by 62.5% of arc tangent function, dropping by 29.2% of bipolar sigmoid function and dropping by 37.5% of sine function based on nonlinear Nomoto model, and the settling time of nonlinear sine controller decreases by 13.3% based on MMG model. It is a useful research that the control performance is improved by nonlinear feedback technique for project application in marine practice.