Cooperative Guidance And Control For Unmanned Surface Vessels

The rapid development of military technology makes the underwater defensive system increasingly effective.Therefore,it becomes more and more difficult for active assault underwater weapons such as torpedo to attack them successfully.Multi-vehicles cooperative attack is one of the most important way to increase the probability of attacking.Moreover,in civilian,since the single-vehicle operating system exists the problems of low detective scope,low precision and inefficiency.It's difficult for single-vehicle to accomplish tasks such as multi-targets detection and cooperative capture in complex marine environments.Comparing with the single-vehicle system,multi-vehicles operating system can complete these tasks more efficiently,and more securely.In such a case,this paper studies cooperative estimation,cooperative guidance and control law of underactuated unmanned surface vessels(USV).The main research work is given as follows:1)The kinematic and dynamic models of the underactuated USV were drived.More-over,some knowledge about the target evaluation performance index that based on Fisher Information Matrix(FIM),multi-vehicles information consistency and prescribed transient performance were introduced.The above-mentioned knowledge are the basic theory for the subsequent researches.2)The cooperative estimation of multi-USVs was studied.To solve the nonlinear prob-lem of sensor observation equations in the marine environmenti,a distributed Kalman filter(CE-DEKF)that consensus on estimates was designed.Under the assumption of bounded states,the stability of the algorithm is proved via Lyapunov stability theory and st,ochastic stability theory,and a simulation is carried out to illustrate the effectiveness of the algorithm.3)The cooperative guidance law for underactuated USV was studied.A proportional guidance that based on Extended Kalman filter was designed with bearing only measure-ments.An USV experimental platform was established,and the experiments in the lake were completed.Moreover,a salvo attack that based on Model predictive control xvas proposed,and the cost function was consisted by t,ime-to-go term and FIM term.The simulation results have shown the effects of the proposed control.4)An adaptive trajectory tracking control for underactuated USV with guaranteed transient performance was studied.To meet the realistic dynamical model of USV.the not diagonal mass and damping matrices were considered.Neural Networks(NNs)were em-ployed to approximate the unknown external disturbances and uncertain hydrodynamics of USV.Moreover,both full state feedback control and output feedback control were presented in this thesis,and the unmeasurable velocities of the output feedback controller were esti-mated via high-gain observer.Unlike the conventional control methods,we employed the error transformation function to guarant,ee the transient tracking performance.Theoretical results were illustrated via the simulations and experiments of USV.