Ship Path Tracking Control Based On Model Predictive Control Method

In order to address the external environmental disturbances and rudder constraints,the model predictive control(MPC)is combined with the radial basis function(RBF)neural network for the path following of under-actuated ship.Then it may achieve that,not only ship can follow the reference path accurately in case of the external disturbances,but also the designed control rudder angle is within the constraints.The main contributions in this paper are as follows:1.Combining with the advantages of MPC,namely obtaining the suboptimal solution under the control input constraints,the rudder angle saturation will be handled.2.The ship mathematical model group(MMG)model with the coupling and nonlinear is used as the MPC internal model,which could promote the accurate of the internal model.And the Euler iteration method is introduced to discrete and predict the ship states,simplifying the prediction of the MMG model and MPC computation.3.The radial basis function(RBF)neural network is exploited to approximate the external disturbances by using the ship historical information,which can compensate for the MPC controller and improve the robustness to the external disturbances such as wind and current.The final simulation experiment results show that,the designed MPC controller can force ship follow the reference path accurately under the environmental disturbances and rudder angle constraints,and the designed control rudder angle is always within the constrained value,so the effectiveness of the proposed algorithm is illustrated.