Robust Control Of Unmanned Surface Vessels

Unmanned surface vessel(USV)can perform tasks in complex marine environments,it has obtained more and more attention all over the world for its outstanding performance in autonomous reconnaissance,surveillance,minehunting,rescue,environmental monitoring and so on.However,the continuous change of the speed and the weight of USV cause the perturbation of model parameters,sea winds,sea waves,ocean currents and other external environmental factors bring external disturbance to the model.The perturbation of model parameters,external disturbance and modeling errors cause the uncertainty of USV model,which seriously affects the robustness of the control system.Therefore,it is of great theoretical and practical significance to research the robust control of USV.The main contributions and innovative achievements are stated as follows:In order to solve the problem of robust control for USV with no system constraints,a disturbance rejection control using composite-errors-based(tracking error and estimation error)extended state observer is proposed.The composite error can improve the smoothness of control response,and it also can achieve small disturbance estimation error and small trajectory tracking error without exciting high-frequency unmolded dynamics.The extended state observer utilizes both estimation error and tracking error to deal with disturbance,so the proposed control is a feedforward feedback composite control,and makes full use of feedforward control and feedback control to suppress the system disturbance.Theoretical analysis and simulation results illustrate the robustness of the proposed control.According to the USV model with position and velocity constraints,a trajectory-linearization based tube model predictive control(MPC)is proposed.The tube MPC includes a linear feedback control and a nominal model based MPC.The nominal model based MPC makes the nominal trajectory converges to the desired trajectory,and the linear feedback control guarantees the USV's actual trajectory being contained in a small area centered at the nominal trajectory.Using trajectory linearization gets the linear time-varying prediction model,which can reduce the computational complexity of nonlinear MPC.The feasibility and robustness of the proposed control is demonstrated by Lyapunov theory and simulation results.According to the USV model with control constraints,a nonlinear continuous time model based tube MPC is proposed.The tube MPC is composed of a nonlinear feedback control and a continuous-time model based dual-mode MPC,the continuous-time model based dual-mode MPC guarantees asymptotical stability of the nominal system,and the nonlinear feedback control can ensure the actual trajectory is contained in a tube with the nominal trajectory as the center.Theoretical analysis and simulation results show that the proposed control has better robust performance when the constraint is guaranteed.