Research On Sliding Mode Control Of Path Tracking Of Unmanned Vessel

With more and more attention paid to the development of unmanned vessel and related technologies,this paper focuses on the path tracking of unmanned vessel.In order to solve the problem of path tracking controller design,LOS guidance law is used to guide the path of the unmanned vessel,and the guidance law is improved according to the:specific situation of the unmanned vessel.At the same time,on the basis of backstepping,sliding mode control theory,neural network and other methods,a path tracking control law with good tracking accuracy,stable performance and strong robustness is designed to solve the problem of model uncertainty,input saturation,external environment disturbance,current interference and other problems that affect the accuracy of path tracking control in the path tracking research of underactuated unmanned vessel.The main research contents and achievements of this paper are as follows:(1)Firstly,considering the influence of constant current disturbance on the path tracking of unmanned vessel,an improved LOS guidance law(ILOS)is proposed to reduce the negative impact of constant current disturbance on the path tracking accuracy,which can make the position error of unmanned vessel converge to zero.At the same time,considering the uncertainty of the model parameters of the unmanned vessel,the paper introduces the nonsingular terminal sliding mode control theory to design the path tracking control law,which effectively solves the problem of the uncertainty of the model parameters and avoids the trouble of high-order derivation of the virtual control law.Then,the convergence of the error in the path tracking control mechanism is proved by the Lyapunov stability theory.Based on the MATLAB simulation platform,two different unmanned vessels are simulated to verify the effectiveness of the designed path tracking control guidance law and control law.(2)Considering the influence of time-varying unknown current,the traditional guidance law and ILOS guidance law can’t make the UAV track the desired path exactly.Therefore,the IAILOS guidance law is proposed,which can estimate the time-varying unknown current disturbance in real time and pass the designed integral term compensation for current disturbance.Then,considering the input saturation limit caused by the actuator(rudder,propeller)and the multiple effects of time-varying external disturbance,the backstepping adaptive sliding mode path tracking control law is designed by combining backstepping,adaptive method and sliding mode control method:At the same time,the compensation system is designed to solve the problem of input saturation.An adaptive law is designed to estimate the unknown disturbance force.Filter is used to solve the problem of differential explosion caused by high order derivative of virtual control law in backstepping,and the design process of controller is simplified.Then the stability of the proposed path tracking control mechanism is proved.Finally,the MATLAB simulation verifies the obvious advantages of the designed IAILOS guidance law in the unknown time-varying current path tracking guidance and the effectiveness of the backstepping sliding mode path tracking control law;the method realizes the effective tracking of the expected path and the expected speed,and meets the maximum control input amplitude limit of the system.(3)Based on the consideration of time-varying current disturbance and input saturation,this paper aims at the problem of unknown system dynamics caused by unknown model parameters of unmanned vessel.The neural network technology with minimum learning parameters is used to obtain the unknown dynamic information effectively.This method has the advantage of not relying on any model parameter information and only training two neural network weight parameters.Then,the neural network first-order sliding mode control.method and neural network PI sliding mode control method are proposed,and the input saturation compensation mechanism is designed by combining neural network and sliding mode.The stability of the two designed path tracking control systems is proved.The effectiveness and advantages and disadvantages of the two proposed path tracking mechanisms are analyzed by simulation.It is proved that the two controllers can effec tively track the desired path and desired speed and meet the limit of system input saturation.The neural network with the minimum learning parameters can effectively estimate the unknown dynamics of the unmanned vessel.