Research On Formation Control Of Underactuated Ships With The Consideration Of Navigation Safety

In recent years,with the development of ship intelligent navigation technolgy,the formation control of multi-ship has gradually become one of the hottest issues in the field of ship motion cotnrol.Compared with the single ship operation,multi-ship cooperative operation is with the characteristics of high efficiency,wide coverage and strong system fault tolerance.The formation control strategy of the ship swarm system is the theoretical basis for realizing multi-ship cooperative operation,its design not only needs to consider the complex task background,but also needs to trade off the nonlinear characteristics of the ship itself and various constraints in the formation process.In this thesis,by taking the underactuated ship as the research object,with the considration of unstable factors that affect the navigation safety of ship in the formation control task,such as unknown model dynamics,marine environment disturbance,actuator failure and input limitation,a set of concise formation control algorithm is proposed by combining theoretical tools such as nonlinear control system theory,algebraic topology theory,fixed-time theory,dynamic surface control technology,neural network approximation technology,and event-triggered technology.The main work of this thesis includes the following three aspects:(1)Aiming at the leader-follower formation control problem of underactuated ships,an adaptive velocity regulation law is designed for virtual ships and a fixed-time control law is designed for follower ships respectively.The velocity regulation law enables the virtual ship to track the reference position planned by the leader ship,and provides guidance information of the formation maneuver to the follower ship.For the kinematic system of follower ship,a fixed-time filter is constructed by combining fractional power technology and filtering technology,which avoids the problem of algorithm complexity explosion.Based on the ship kinetic system,the parameter adaptive technology is used to estimate the unknown dynamic nonlinear function and the upper bound of external disturbance in the model.Furthermore,an event-triggered mechanism is introduced to the sensor-controller channel,which significantly reduces the update frequency of the control law and reduces the wear of the actuator.(2)Aiming at the cooperative path following formation control of multiple underactuated ships on parameterized paths,a robust neural network control law is designed for the single-vessel path-following loop and an event-triggered cooperative law is designed for the multi-ship cooperative control loop respectively.In the control law,the dynamic surface control technology is introduced to filter the virtual control law,which avoids the problem of increasing algorithm complexity induced by repeated derivation of the virtual control law in the subsequent design.The neural network is used to approximate the unknown dynamics of the model,meanwhile,the neural network approximation error,external disturbance and additional fault of the actuator are integrated into a lumped disturbance term,then the upper bound of the neural network weight matrix and the upper bound of the lumbed disturbance are estimated by the parameter adaptation technology,and then achieve the purpose of compressing the number of adaptive parameters,reducing the computational load of the algorithm.In the cooperative law,the triggering threshold condition is set based on the path parameter error,so that the cooperative law can only be updated under certain condition,which effectively reduces the communication cost among formation members,so as to achieve the purpose of compressing the number of adaptive parameters and reducing the computational load of the algorithm.(3)A distributed fault-tolerant saturated formation control law is designed for the parallel formation control of mulitple underactuated ships.In the kinematics system,a distributed virtual control law is designed based on the local information of the communication topology,which promotes the formation configuration and formation maintaining motion of the ship swarm,reduces the communication burden between the formation members simultaneously.In the kinetic system,the neural network is employed to approximate the unknown dynamics of the model,and the parameter adaptation technology is used to estimate the upper bound of the weight and the upper bound of the lumped disturbance.In addition,a general saturation adaptive law is proposed for the problem of limited input,which can eliminate the adverse effect of input saturation on the closed-loop system,improve the safety assurance of ship navigation.To sum up,this thesis adopts three formation methods to study the formation control of underactuated ships under different constraints.For the control law,the filter technology and the boundary feedback adaptive law are used to optimize the form of the algorithm and reduce the computational load.The event-triggered mechanism is used to discretize the algorithm aperiodically,which significantly reduces the communication frequency.The algorithm in this thesis is with the advantages of concise form,small computational load,strong robustness,safe and reliable,and plays an important role in promoting the automation and intelligent of the surface vessel in our country.