Command Filtering Adaptive Terminal Sliding Mode Control For Robotic Manipulator

When designing a control scheme for a nonlinear system,the unknown state of the system,backlash,friction and other nonlinear links will interfere with the tracking effect of the closed-loop signal output of the whole system,affect the stability of the system,and even lead to the system collapse,including the torque delay of backlash in a certain period of time,and the complex friction makes it difficult to establish the system model,It further leads to the tracking error of the system to the reference signal and some unknown disturbance signals when the manipulator joint is reversing.Based on the above analysis,how to improve the transient and steady-state performance of the whole system by using excellent control schemes such as adaptive control,sliding mode control,and learning control to estimate,compensate,and suppress nonlinear links has been a hot topic of research by contemporary scientific research teams.In this paper,the backlash and friction nonlinearity are sorted out respectively,and introduced into the dynamic model of the manipulator system.How to realize the accurate and efficient tracking of the reference signal of the manipulator system with backlash and friction is analyzed.The extended state observer in auto-disturbance rejection control is used to observe and feedback the "total disturbance" sorted out by the system in real time.According to the structural characteristics of the manipulator system,the corresponding extended state observer is applied to improve,or the neural network is used to real-time fit the system state and unknown disturbance signals,and the friction of complex structure is accurately approximated by the neural network based on stored knowledge,According to the required performance index,the adaptive law is designed to modify the parameters used in the controller design.Finally,the adaptive sliding mode controller is designed to ensure the finite-time convergence of the system error.The stability and effectiveness of the control scheme are proved by theoretical analysis and simulation.1.Aiming at the single-arm system with nonlinear effects such as friction,a command filtering adaptive sliding mode control scheme with predictive performance is proposed.The disturbance characteristics of backlash and friction nonlinear links are analyzed,the series integral system structure is established,and the backlash and friction are converted into the total disturbance of the system.A novel non-singular sliding surface is established,and the predictive performance function is added to make the system state converge faster and reduce chattering.In order to obtain a stable control signal in the feedforward link,an adaptive regulation law is designed to modify the controller parameters online,and the unknown system information is estimated through the adaptive law.The simulation results show that the control scheme has a good tracking effect on the backlash and friction nonlinearity of the manipulator system.2.Due to the unknown nonlinear friction,there was a barricade preventing the precision of manipulators from further improvement.To overcome this challenge,the dynamic model of 2-degrees of freedom robot manipulator based on LuGre friction is established in this paper.The adaptive sliding mode observer is used to estimate the immeasurable states,meanwhile the neural network to approximate the friction.On this foundation,a neural-learning-based finite-time trajectory tracking control is designed to improve robustness.In particular,the closed-loop system stability is investigated by the Lyapunov theorem and computed the finite convergence rate thereafter.Finally,simulation results show that the control scheme has a better control effect of the manipulators.