Trajectory-linearization Based Robust Model Predictive Control for Unmanned Surface Vessels with System Constraints
In this paper, a trajectory-linearization-based robust model predictive control (MPC) is proposed for unmanned surface vessels with system constraints and disturbances. The trajectory linearization technique is used to translate continuous-time nonlinear model of vessels into a linear time-varying predictive model and decreases complexity of nonlinear MPC. The proposed MPC is composed of a linear feeback control and a MPC. The linear feedback control ensures the real trajectory contained in a tube of trajectory of a nominal system, while the MPC guarantees the asymptotial stability of the nominal system. Theoretical analysis and simulation results illustrate the effectiveness of the proposed control law.