| The manipulator system is difficult to carry out real-time robust control because of its high complexity,strong coupling,uncertainty,and nonlinearity.Sliding mode variable structure control has the advantages of strong robustness,no need for an accurate mathematical model,insensitive to parameter perturbation,and fast response speed.It has become an effective control strategy for the manipulator system.However,the existing manipulator trajectory tracking sliding mode control methods still have some shortcomings,such as a simple controlled object,unreasonable expression of system uncertainty,serious control input chattering,the state can not converge within finite time,and the controller gain design depends on the prior knowledge of the upper bound of system uncertainty.To solve the above problems,this paper studies three aspects:(1)Aiming at the problem that the controlled object is limited to the two degrees of freedom manipulator and does not consider that the manipulator should be multi-degree of freedom in practical engineering,the dynamic model of the classical three degrees of freedom coordinate robot is established and the controller is designed based on it.In addition,the modeling error separation method rather than the Gaussian basis function is used to express the irregular uncertainty of the system,which is more in line with the characteristic of actual disturbance changes.(2)Aiming at the problems that the control input chattering is serious,the system state can not converge within finite time and depends on the prior information of uncertain upper bound,a non-singular fast terminal sliding mode control method based on disturbance observer is proposed.Firstly,a non-singular fast terminal sliding mode surface that can improve the design efficiency of the controller is proposed,and the finite-time convergence characteristic of the surface is proved.Secondly,the control law is designed and the stability is proved based on this surface.Finally,the method of combining disturbance observer and boundary layer technology is used to effectively weaken the control chattering.The simulation results in MATLAB/Simulink module show that the manipulator system realizes high-precision and fast-tracking of target trajectory with low control input chattering when the upper bound of uncertainty is unknown.(3)Aiming at the problems that the control input chattering is serious,the system state can not converge within finite time and depends on the prior information of uncertain upper bound,a gain adaptive non-singular fast terminal sliding mode control method is proposed.Firstly,the adaptive technique is used to approximate the upper bound of uncertainty.Secondly,the boundary layer technology is used to weaken the chattering phenomenon of control input.In addition,this method not only proves the existence of the upper bound of the uncertainty of the system including friction which extends the existing adaptive terminal sliding mode control technology but also proposes a control strategy to delay the occurrence of chattering by adjusting the thickness of the boundary layer according to the boundary layer theory.The simulation results show that the manipulator system can realize high-precision and fast-tracking of target trajectory under unknown upper bound and low control input chattering. |