Research On Path Following Control Of USV In Complex Sea Conditions

Unmanned surface vehicle(USV)plays an increasingly important role in marine resources development and marine defense.Path following control is one of the key technologies for USV to achieve precise navigation and perform some complex line following tasks.Therefore,researching on USV path following control has great significance to enhance the automation and intelligence level of USV.When USV moves in complex sea conditions,it will be disturbed by environmental forces such as wind,waves and ocean currents,and the model parameters of USV will have uncertainty.Consequently,robust control method is needed to ensure the stability of USV control system and the accuracy of USV motion.Under this background,this paper analyses the difficulties in USV path following control,and focuses on the study of path following control in complex sea conditions.The main research contents are as followsBy analyzing the kinematics,dynamics of USV and the environmental disturbance characteristic,the three-degree-of-freedom horizontal motion model of USV in complex sea conditions is built based on the six-degree-of-freedom motion mathematical model of USV Meanwhile,motion control characteristics of USV are sorted out,including underactuation,model uncertainty,environmental disturbance and actuator constraints,which lay the theoretic foundation for the development of path following control algorithm.The way-point based path following control problem for USV is investigated.Considering that the USV path following for engineering applications is usually a straight line path following based on predefined way points,switching Line-of-Sight(LOS)guidance law and S plane heading controller with transition objective function are designed.The switching LOS guidance law switches different LOS radius according to the change of path following error to guide the USV to converge to the desired path quickly.The transition objective function S plane heading controller arranges a transition process for the desired heading angle to deal with the sudden change of the target heading angle.An experimental USV platform named "KV-1" is developed independently,and field experiments on path following control are carried out.The path following control experiments include broken line path following with three way points and comb line path following with eight way points.The curve path following control problem for USV in sea conditions with ocean current disturbance is investigated.Considering the influence of ocean current on USV kinematics,a new USV path tracking error kinematics model in the Serret-Frenet frame is established by using the relative velocity information of USV.In order to compensate for the current drift force for USV.a current state observer is designed to estimate the current velocity,and then an adaptive LOS guidance law is proposed.The path following dynamic control law is designed based on backstepping method,and the relative velocity drift angle acceleration is solved algebraically.The proposed path following control method can guarantee all the error signals in the closed-loop system to be global K-exponential convergence.Simulation studies on straight line path following control with constant current disturbance and curve path following control with change current disturbance are presented to demonstrate the effectiveness of the designed control systemThe curve path following control problem for USV in complex sea conditions with mixed disturbance containing wind,waves and ocean currents is investigated.The guidance law is designed by employing virtual ship principle.The order of virtual control variable is reduced by using dynamic surface control(DSC)technique,and then the kinematics controller is obtained.Considering the model uncertainties of USV in complex sea conditions,a concise adaptive composite radial basis function neural network(RBFNN)path following control law is proposed by using the nonlinear function approximation function of the neural network.The designed composite RBFNN uses the state tracking error and state prediction error of USV as the input information of the neural network,and compresses the online learning parameters of the neural network based on minimum parameter learning method,so as to optimize the structure of the network.The proposed path following control law can guarantee the closed-loop control system uniform ultimate boundedness.Simulation experiments are conducted to validate the effectiveness of the proposed control method.The curve path following control problem for USV in complex sea conditions with mixed disturbance containing wind,waves and ocean currents and actuator constraint is investigated.Considering the saturation nonlinearity of USV actuator,a saturation function approximation model based on parametric hyperbolic tangent function is designed.Based on the designed saturation function approximation model and using the mean value theorem,the USV dynamic equation is transformed,so that the control signal is expressed explicitly to facilitate the design of the dynamic controller.The path following kinematic controller is designed using the DSC technique.Considering the model uncertainty and time-varying unknown disturbance of USV in complex sea conditions,a neural network disturbance observer based on RBFNN is designed to estimate and compensate the compound time-varying disturbance consisting of the saturation function approximation error and the environmental disturbance,then a concise adaptive neural network path following control law is proposed.The proposed path following control law can guarantee that the tracking errors of the closed-loop system are uniformly ultimately bounded.The effectiveness of the proposed control method is verified by simulation experiments.The algorithms developed in this paper are implemented by Visual C++/Matlab hybrid programming.The experimental results show that the proposed path following control methods are effective and have good tracking control performance.