Finite-time Neural Network Impedance Control Of Manipulator Considering State Constraint

With the vigorous development of manufacturing industry in China,the development and application prospect of robotic manipulators is becoming more and more extensive,which makes the control research of robotic manipulators get extensive attention.In practical applications,the work that requires physical contact between the robotic manipulators and the environment increases rapidly,and the mathematical model parameters of the robotic manipulators are difficult to obtain accurately.In addition,due to the influence of the working environment and hardware performance,the robotic manipulators are affected by unknown nonlinear disturbance and various constraints during the working process.And for the requirements of safety,efficiency and high precision,the control systems of robotic manipulators are required to have faster transient response speed and high-precision tracking performance in some application fields.In order to solve the above problems,a finite time neural network impedance control method for the robotic manipulators considering state constraints is proposed in this dissertation by combining command filtering,adaptive neural network,disturbance observer,barrier Lyapunov function(BLF),finite-time control(FTC)technology and backstepping method.The main research results of this paper are as follows:(1)An adaptive neural network-based finite-time output constraint command filtered impedance control method is proposed for the robotic manipulators with output constrained.Firstly,by combining FTC technology with logarithmic BLF,the tracking errors of control system have the faster convergence rate without violating the output constraint.Then,the adaptive neural network technology is introduced to deal with the uncertainty in the mathematical model of the robotic manipulators.In addition,the command filtering technology and error compensation mechanism are used to solve the problem of "explosion of complexity" in the process of backstepping controller design,so that improve the control effect.The simulation results can prove that the proposed control method can achieve better tracking performance without violating the output constraints.(2)A disturbance observer-based finite-time output constraint control method is proposed for the robotic manipulators with unknown nonlinear disturbance and input saturation.Firstly,the hyperbolic tangent function is used to approximate the input saturation to obtain a smooth control input.Secondly,a disturbance observer is designed to handle the influence of unknown nonlinear disturbance,which reduce the practical application cost and design difficulty of the robotic manipulators.The simulation results show that the proposed control method can estimate and compensate the unknown nonlinear disturbance,and achieve better control performance without violating output constraints.(3)Furthermore,a disturbance observer-based finite-time full state constraints control method for the robotic manipulators is proposed.This method considers the full state constraints of the robotic manipulators and the possible reduction of the control performance due to the use of hyperbolic tangent function to approximate the input saturation nonlinearity.Firstly,by combining FTC technology with logarithmic BLF,the control system realizes the finite-time full state constraint control of robotic manipulators,the tracking errors of control system have the faster convergence rate without violating the full state constraints.Then,an improved smoothing function is used to approximate the saturated input,reducing the approximation error and improving the control effect.The simulation results show that the proposed control method can achieve better tracking control performance without violating the full state constraints.