Research On Distributed Model Predictive Control Strategy Of Surface Vehicle Based On Multi-agent

With the increasingly improved technology of Unmanned Surface Vehicle(USV),USV has gradually become a new force in the civilian and military fields.Considering the weak antiinterference ability of a single USV,the vigorous development of multi-agent technology in recent years has given important research significance and application value to multiple USV formation control systems.Multi-USV formation systems exhibit exceptional capabilities in ocean search and rescue,prey capture,and marine environmental cleaning.Multi-USV formation is characterized by flexibility,autonomy,and robustness.Therefore,this article takes USV as the research object and conducts research on the cooperative control strategy of USV.The main contents are as follows:The mathematical model of USV’s six degrees of freedom motion is introduced.During the modeling process,the effects of sea wind,waves,and currents on the ship are comprehensively considered,and the interference forces and moments generated by the above interference on the ship’s hull are analyzed in detail.Compared with manned ships,USV can ignore the consideration of comfort,so when studying the formation problem of USV,it is only necessary to consider three degrees of freedom mathematical models of yaw,sway,and surge.Fully considering the communication between USVs,information sharing(state information and speed information)of USVs is realized,and a simulation model for formation cooperative control that integrates kinematic and dynamic models is established.2.Aiming at the trajectory tracking problem of USV,a trajectory tracking control algorithm for surface vehicles based on Model Predictive Control(MPC)is designed.Based on the mathematical model of the three degree of freedom surface vehicle motion,the trajectory tracking state equation of the surface vehicle is established.Using MPC algorithm,the optimal control of the surface vehicle is achieved through rolling optimization and feedback correction.In order to verify the effectiveness of the kinematic model and the algorithm,and the linear trajectory and the sinusoidal trajectory were simulated and analyzed,respectively.3.Aiming at the formation and maintenance of USV based kinematic model formation,a distributed model predictive control(DMPC)based formation collaborative control strategy was designed.A DMPC controller is designed based on the requirements for the speed of each USV and the relative position relationship between USVs in the formation,and the detailed design process is given.To verify the effectiveness of the control strategy,simulation analysis was conducted for the formation coordination problem under two scenarios.4.Aiming at the formation and maintenance of USV formation,a collaborative control strategy for USV formation based on post check DMPC is designed.A hierarchical control structure is considered in the controller design process,where the upper layer determines the reference signal for the USV speed,while the lower layer optimizes the control input for each USV to track the reference signal of the upper layer.In order to solve the failure state caused by local errors or network attacks,a post verification program was introduced.Each USV compares the actual state with the predicted state obtained at the previous moment to reduce formation errors to allow for local disturbances.In addition,if the prediction error value meets preset conditions,the USV is considered to be in a fault state,forcing a change in the communication topology and using a standby topology to improve formation stability.To verify the effectiveness of the proposed control strategy,a simulation analysis of the USV formation coordination problem under external interference was conducted.