Research On Sliding Model Control Of Space Manipulators In Cross-Scale Environment Based On Neural Network

In recent years,space manipulator has played a more and more important role in space activities as a key technology in orbit service.The space manipulator is assembled in the ground gravity environment and is served in the micro-gravity environment of space,thus leading to changes in the motion behavior of space manipulator because of the change of gravity environment,which is a worthy subject of study to solve this kind of difference.In this paper the trajectory tracking control problem of space manipulator in cross-scale environment is studied.The specific contents are as follows:Firstly,the current status of micro-gravity motion behavior and the trajectory tracking control of space manipulator at home and abroad are described,as well as the kinematics and dynamic models of space manipulator.Secondly,a global PID sliding mode control scheme based on the nominal model of space manipulator is proposed to solve the joint trajectory tracking control problem of space manipulator in cross-scale environment.In order to make the initial state of the system is on the sliding surface,a global sliding surface is designed,eliminating the movement stage of the traditional linear sliding mode control.Based on the nominal model of the space manipulator,the equivalent control law is derived,and the switching controller based on the exponential reaching law is used to overcome the influence of the uncertainty in the model.The asymptotic stability of the closed-loop system is proved based on Lyapunov's theory.Then,a global sliding mode control method based on neural network compensation is designed to solve the chattering problem in last chapter.RBF neural network is introduced to approximate the uncertainty in the model,which includes the modeling error and so on,the approximation error is compensated by the switching controller,where introducing the hyperbolic tangent function to replace the sign function,The weight adaptive law of neural network is designed,and the asymptotic stability of the closed loop system is proved based on the Lyapunov theory.The simulation results show that the designed controller can greatly weaken the chattering phenomenon of the system,and it has strong robustness.Finally,the influence of the changing gravitational term on the motion control of the space manipulator in the cross scale environment is analyzed.A new neural network terminal sliding mode control strategy is proposed to solve the problem of gravity variation of space manipulator under cross scale environment.A double exponential terminal sliding surface is designed,RBF neural network is introduced to approximate the gravity term of the mode on-line,f_e(28)k_iòsdt is to overcome the approximation error,and the switching controller is to compensate the uncertainty in the mode.Based on the Lyapunov theory,the stability of the closed loop system is proved.The simulation results show the effectiveness of the control method.