Neural Adaptive Output Feedback Control For Multi-link Robot Manipulators

The multi-joint robot system involves the coordinated control among the joint actuators,which has the dynamic characteristics of coupling,time-varying and non-linearity,so it is difficult to control the multi-joint robot.In addition,if uncertainties such as unmodeled dynamic errors,actuator faults,friction between robotic joints,external disturbances,unknown delay,non-smooth dead-zone characteristics and signal measurement errors are taken into account,the dynamics model of multi-joint robotic systems will become more complex.However,these factors always exist and are inevitable in the actual engineering environment.At the same time,the unmeasurable states of the system will also have a serious impact on the stability of the system,which will bring some difficulties to the design of the controller.In this paper,the tracking control of multi-joint robot cooperative system with external disturbance,friction and other uncertain factors under the action of output feedback controller.is studied.Firstly,for the multi-joint robot system with uncertain factors such as joint friction and external disturbance,a relatively accurate mathematical model is established by using the Lagrange method.Based on this model,a multi-arm cooperative grasping payload model is analyzed and a dynamic model is established.Secondly,a neuro-adaptive output feedback controller is designed under full-state constraints for the cooperative system of multi-joint robots with external disturbances,friction between joints,and part of the state is not measurable.The approximation performance of RBF neural network function is used to deal with uncertainties such as external disturbances and joint friction in multi-joint robot system;a linear state observer is designed to estimate unmeasurable state variables and an auxiliary system is used to eliminate the influence of input saturation;on this basis,the integrated obstacle Lyapunov function(iBLFs)and Backstepping design method are combined to deduce the adaptive rate and controller.The deduction results show that the designed control strategy can ensure that the state of the system satisfies the constraints,ensure the stability of the closed-loop system,and all the signals in the closed-loop system are ultimately uniformly bounded.The validity of the proposed controller is further verified by the simulation of Matlab software.Finally,a neuro-adaptive output feedback controller is designed under unknown desired trajectories for the cooperative system of multi-joint robots with external disturbances and friction between joints and some states are not measurable.The approximation performance of RBF neural network function is used to deal with uncertainties such as external disturbances and joint friction in multi-joint robot system;the neural network function with time-varying weights is used to reconstruct the unknown desired trajectory;the high-gain K filter observer is used to estimate the unmeasurable state variables of the system;and the Backstepping design method and Lyapunov function were introduced to design the controller,and the results show that the control strategy can ensure that the system can track the unknown desired trajectory and ensure that all signals of the closed-loop system are ultimately uniformly bounded.The numerical simulation results further confirm the feasibility of the designed controller.