Research On Motion Control Of Ship’s Safe Navigation In The Ocean

As a very important platform for ocean-going transportation and operation,it is necessary to guarantee safe sailing of large marine crafts in complex ocean conditions.So it has remarkable practical significance for active attitudes(roll/heading)control of safe sailing.The robustness and adaptability for ship safe sailing control system are of crucial importance for large container ships with high centers of gravity.Aiming at the research frontiers of the state-of-the-art control techniques,the motion control schemes and strategies are investigated and developed in this dissertation to further improve the sailing capability and safety for large ships.In our investigations,we mainly focus on large-amplitude roll,heading,and roll-heading ship motion control,which is inherently nonlinear.Firstly,the recent development of ship motion attitude control,and the research progress of the relevant control theory are given.The nonlinear four degrees of freedom(4 DOF)coupling motion model is constructed by considering the superposition of control force,hydrodynamic force and environmental force,etc.Moreover,the boundary layer effect from the hull is taken into account to estimate the area of fin stabilizers,and the uncertainty of resistance is evaluated.Then simulations of the irregular wave disturbance forces/moments are given by means of the strip theory calculation and the weighted average of power spectrum of the first-order wave disturbance force/moment.Furthermore,through rational assumptions,an affine nonlinear ship lateral motion control model with a constant speed is established,whose open-loop stability is also analyzed.Secondly,a nonlinear disturbance observer based command-filtered backstepping control strategy and an improved method using integral sliding mode control with prescribed performance are proposed with respect to the roll stabilization problem with the large-amplitude roll motion.A nonlinear disturbance observer is introduced to observe the wave disturbance.And then the backstepping controller is designed based on the nonlinear roll model with nonlinear disturbance observer(NDO),being connected with the amplitude limiting command filter in series,which avoids an inherent disadvantage of differential calculation expansion in conventional backstepping control approaches.The nonlinear disturbance observer based command-filtered backstepping control(NDOCBC)strategy is thus proposed.Furthmore,prescribed performance control(PPC)is introduced on the basis of NDOCBC,which limits the virtual control variable to the preset boundary of the prescribed performance function.The improved approach using integral sliding mode control with prescribed performance is designed,integrating sliding mode control and adaptive technique,which further eliminates the diturbance estimation error and enhances the robustness.Simulations reveal that the improved approach avoids the effect deterioration of roll stabilization with the saturation limit of fin angles in a severe sudden-change sea situation.The controller renders smooth control performance when stabilizing the roll,which takes into account both the control accuracy and reliability.Thirdly,a radial basis function(RBF)neural network based discrete command-filtered backstepping control method is proposed with respect to the heading control problem of the safe sailing in the ocean.The complex nonlinear terms with respect to ship heading control,in the multiple DOFs coupling the nonlinear motion equation,are represented by the RBF neural network which is used to provide global approximation.Meanwhile,the state observer is designed through state reconstruction to realize adaptive estimation of the RBF weights.Finally,the discrete command-filtered backstepping controller is constructed based on the estimation information of RBF model parameters,and a compensator is designed to compensate the filter error accurately and timely,which avoids the control accuracy deterioration caused by discrete command-filter errors.This approach has considerable practical value,which satisfies the real-time control requirements and may significantly improve the accuracy,robustness and adaptability of ship heading control.And then,a nonlinear adaptive proportional-integral-derivative(PID)control strategy of multiple-input multiple-output(MIMO)extended state observer(ESO)is proposed with respect to ship roll-heading control by the rudder-fin system.The dynamics of ship roll-heading control system is analyzed,based on which the control model is obtained by transforming from three DOFs of sway/roll/yaw(heading)to two DOFs of roll/yaw(heading).For the simplified model,the solvable proof of feedback linearization is achieved,and the decoupling pseudo-linearization control system is achieved.Then an adaptive PID pseudo controller based on ESO is designed to deal with the external wave disturbance and the uncertainty of the model.In the proposed control method,the nonlinear ESO is used to estimate the states of the decoupling linearization system and the lumped disturbance,meanwhile reduce the switching function gain of the adaptive PID pseudo controller,which can effectively suppress the control input chattering.As a result,the approach can realize nonlinear decoupling control of ship roll-heading motions via the rudder-fin joint system,and guarantee its robustness and the anti-interference ability.Finally,a robust H_∞-type input constrainted model predictive control(H_∞-ICMPC)is proposed with respect to the performance optimization and the actuator action reliability for the ship safe sailing.Taking into account that there is a delay constraint for the pratical fin stabilizer system,the roll angles are predicted by fading memory recursive least squares(FMRLS)based on an Auto-Regression(AR)model,which is rendered to the controller design.Owing to the capabilities of handling multivariable and dealing with constraints,model predictive control(MPC)is considered in our study.Specifically,the H_∞-type MPC is introduced,which achieves performance optimization of the system with disturbances.The closed-loop control system is guaranteed to be input-to-state stable.Furthermore,the effect caused by disturbances is attenuated.The state feedback control gain is obtained for rudder-fin joint roll-heading control system,by resorting to the constrained optimization problem into matrix inequalities.The input constraints mainly take into account the actuators,i.e.,the rudder and fin stabilizers.The resultant control actions are smooth,which further reduces the energy consumption of the actuators.Simulations reveal that the proposed H_∞-ICMPC approach has outstanding control performance.In addition,it is found that the approach also has significant robustness for the rudder-roll damping(RRD)control system with parameter uncertainties.