Trajectory Tracking Control Based On Planar 2-DOF Manipulator

In this thesis,several sliding mode variable structure control algorithms are designed for the trajectory tracking of a planar two-degree-of-freedom manipulator,and their control performance is analyzed and compared,and the effectiveness of the control algorithm is verified by numerical simulation.First part of the basic concepts and the rigid arm application,a brief introduction to the basic structure of the type of mechanical arm and each made,and the development history of mechanical arm control theory to do a literature review,mainly introduced the robust adaptive control in rigid manipulator trajectory tracking as well as the application status on the joint position tracking.In chapter 2,based on Lagrange dynamics equation,the dynamics model of planar 2-DOF manipulator is established in detail,the simplified DC motor model is designed,and the forward and inverse kinematics of the manipulator is introduced,which provides the system model foundation for the following manipulator control algorithm.Secondly,an adaptive sliding mode control(A-SMC)algorithm is designed based on Lyapunov stability theory,which enables the manipulator end to track different reference trajectory stably when the system has parameter uncertainties and external disturbances.The tracking control of the eccentric circle reference trajectory and the parallelogram reference trajectory with the same parameters is carried out respectively,and the control effect of the system without external disturbance is compared with that with external disturbance.The simulation results show that the designed A-SMC control algorithm can effectively make the manipulator end track the corresponding reference trajectory stably.In the case of external disturbance,it shows good robustness against disturbance.Then based on the Lyapunov stability theory,an RBF neural network sliding mode control(RBFNN-SMC)algorithm is designed,so that the end of the manipulator can track different reference trajectories stably when the system has parameter uncertainties and external disturbances.In order to verify the robustness of the RBFNN-SMC algorithm,the control effect of the system without external disturbance and with external disturbance is compared.The simulation results show that the designed RBFNN-SMC algorithm can effectively make the end of the manipulator track the corresponding reference trajectory stably,and has a certain degree of robustness to the system uncertainty.In order to further discuss the control performance of RBFNN-SMC,it is compared with the A-SMC algorithm.The simulation results show that the control accuracy of the closed-loop system is almost the same as RBFNN-SMC and A-SMC.For the robustness of closed-loop systems against disturbances,A-SMC is better than RBFNN-SMC.Finally,based on the RBFNN-SMC algorithm,a neural network sliding mode control(DOBNN-SMC)algorithm based on a nonlinear disturbance observer is designed,so that the end of the manipulator can be stable when the system has parameter uncertainties and external disturbances.Track different reference trajectories.In order to verify the robustness of the DOBNN-SMC algorithm,the control effect of the system against different reference trajectories under external disturbances is compared.The simulation results show that the designed DOBNN-SMC algorithm can effectively enable the end of the manipulator to track the corresponding reference trajectory stably,and has good robustness to system uncertainty.In order to discuss the control performance of DOBNN-SMC in detail,it is compared with RBFNN-SMC algorithm.The simulation results show that DOBNN-SMC can effectively improve the steady-state tracking accuracy on the basis of RBFNN-SMC control algorithm,and overcome the weakness that RBFNN-SMC is more prone to be affected by disturbance due to the lack of input information of external explicit time-varying disturbance.