Research On Adaptive Backstepping Control Of Flexible Joint Manipulator System

Flexible joint robots can usually complete many tasks that rigid robots can't accomplish,and the research on relevant control strategies for flexible joint robots has attracted more and more researchers at home and abroad.Among them,how to get accurate position tracking control effect is an important research topic in this field.This paper studies the dynamic model of flexible joint mechanical arm and designs several kinds of adaptive command filtering based on neural network control strategies considering the uncertainties in the dynamic model.The main research work includes the following points:1.For the flexible joint manipulator nonlinear system,the adaptive command filtering backstepping control strategy based on neural network is studied.The adaptive control is integrated into the traditional backstepping method,and the error compensation mechanism is established to eliminate the error caused by the applied filter.An adaptive technique based on neural network is used to approximate the uncertainties in the system model of flexible joint manipulator.Finally,the stability of the closed-loop system is verified according to Lyapunov theorem,the desired joint position tracking is achieved,and the Matlab/Simulink simulation experiment is used to verify the effectiveness of the control strategy through numerical simulation.2.Aiming at the flexible joint manipulator nonlinear system,the neural network adaptive command filtering backstepping control strategy is studied by using finite-time control.The finite-time command filter is used to solve the computational complexity problem in calculating the derivative of virtual signal.On this basis,aiming at the filtering error caused by using finite-time command filter,a finite-time command filtering error compensation mechanism is designed to compensate the error.In addition,considering the uncertainty problem of the system model,the control strategy approximates the uncertain dynamics of the model through the neural network,and proves that the proposed control strategy can ensure that the joint position tracking error converges to the desired neighborhood in finite time.Matlab/Simulink simulation experiments are used to verify the effectiveness of the control strategy.3.For multiple flexible joint manipulators,the distributed adaptive finite-time command filtering backstepping control strategy based on directed graph is studied,and the distributed virtual control signal and error compensation mechanism are designed.Based on the finite-time Lyapunov stability theory,it is proved that the joint position of all the following manipulators can track the joint position of the leading manipulator in finite time,and the tracking error converges to the expected neighborhood in finite time,and all the control signals in the closed-loop system are bounded in finite time.Matlab / Simulink numerical simulation are used to verify the effectiveness of the method.