| In this dissertation, as the object of study, flexible joint robot is introduced. The problems of the dynamic parameter identification, modeling analysis and fuzzy control are studied, which combined the influence of joint surface dynamic variety, parameter uncertainty, modeling error, unknown disturbance, fragile of controller and input delay. According to theoretical analysis and experiment, several approachs of parameter identification and control are proposed. The conclusions and results offered are helpful to the research of parameter identification and control for flexible joint robot. The detail work would be shown as follow:Firstly, a approach to identify joint parameters of the robot in motion state is proposed. Considered robot’s flexible joint as elastic torsion axis, combined the travelling wave approach of structures, the wave equations of robot mechanism is obtained. The vibration torques of system in motion state are established. The vibration equations are received according to the vibration torques and vibration angles. Joint dynamic stiffness and damping in the vibration equations are identified by using the RBF neural network which could approximate nonlinear function with arbitrary precision. Parameters identification experiment of a3-DOF robot driven follow fixed trajectory is done and the experimental result shows that the approach is feasible and effective.Secondly, a fuzzy robust control approach for flexible joint robot is proposed. The dynamic model of flexible joint robot is analyzed, inevitable parameter uncertainty is considered, the uncertainty Takagi and Sugeno(T-S) fuzzy model is employed to approximate the flexible joint robot.(1) A fuzzy robust controller is developed based on parallel distributed compensation principle(PDC). The fuzzy control system of flexible joint robot is constructed. The sufficient conditions for the stability of closed-loop control system for the flexible joint robot are proposed by using Lyapunov function combined with linear matrix inequality (LMI).(2) The problem of constraint on input and output is considered, the sufficient conditions for the stability of closed-loop control system for the flexible joint robot are proposed by using Lyapunov function combined with linear matrix inequality (LMI).The simulation experiment is given to illustrate the robust of the fuzzy robust approach of flexible joint robot according to MATLAB software. Thirdly, fuzzy H-inf control, fuzzy passive control and fuzzy dissipative control approachs for flexible joint robot are proposed. The dynamic model of flexible joint robot is analyzed, the inevitable parameter uncertainty and unknown disturbance are considered, the Takagi and Sugeno(T-S) fuzzy model is employed to approximate the flexible joint robot.(1) The fuzzy controller is developed based on parallel distributed compensation principle(PDC). H-inf performance and passive control are introduced to restrain the influence of parameter uncertainty and unknown disturbance to flexible joint robot control system,(2) Dissipative control is introduced to restrain the influence of unknown disturbance to flexible joint robot control system for increasing the flexibility of the controller.The fuzzy control system of flexible joint robot is constructed. The sufficient conditions for the stability of closed-loop control system for the flexible joint robot are proposed by using Lyapunov function combined with linear matrix inequality (LMI). The simulation experiment is given to illustrate the robust and influence of unknown disturbance of the control approach of flexible joint robot according to MATLAB software.Fourthly, fuzzy non-fragile control and fuzzy non-fragile dissipative control approach for flexible joint robot are proposed. The dynamic model of flexible joint robot is analyzed, the fragile of controller, parameter uncertainty, unknown disturbance and modeling error are considered, the Takagi and Sugeno(T-S) fuzzy model is employed to approximate the flexible joint robot.(1) The fuzzy non-fragile controller is developed based on parallel distributed compensation principle(PDC), additive perturbation and multiplicative perturbation of controller are considered in the design of controller. H-inf performance and passive control are introduced to restrain the influence of parameter uncertainty, unknown disturbance and modeling error to flexible joint robot control system.(2) Dissipative control is introduced to restrain the influence of parameter uncertainty, unknown disturbance and modeling error to control system of flexible joint robot for increasing the flexibility of the controller.The fuzzy control system of flexible joint robot is constructed. The sufficient conditions for the stability of closed-loop control system for the flexible joint robot are proposed by using Lyapunov function combined with linear matrix inequality (LMI). The simulation experiment is given to illustrate the effectiveness and robust of the control approach according to MATLAB software.Finally, fuzzy input delay dissipative control and fuzzy non-fragile input delay dissipative control approach for flexible joint robot are proposed. The dynamic model of flexible joint robot is analyzed, input delay, the fragile of controller, parameter uncertainty, unknown disturbance and modeling error are considered, the Takagi and Sugeno(T-S) fuzzy model is employed to approximate the flexible joint robot.(1) The fuzzy input delay dissipative controller is developed based on parallel distributed compensation principle(PDC). Dissipative control is introduced to restrain the influence of parameter uncertainty, unknown disturbance and modeling error to flexible joint robot control system.(2) Fuzzy non-fragile input delay dissipative controller is developed based on dissipative control, restrain the influence of controller fragile to flexible joint robot control system,The fuzzy control system of flexible joint robot is constructed. The sufficient conditions for the stability and controller gain matrix of closed-loop control system for the flexible joint robot are proposed by using Lyapunov function combined with linear matrix inequality (LMI). The simulation experiment is given to illustrate the effectiveness and robust of control approach of flexible joint robot according to MATLAB software. |
PDF Full Text Request