| With the advancement and development of science and technology,the robot has been involved in all aspects of human life.Sliding mode control should be widely used in nonlinear robotic systems for its appealing advantages of the nonlinear control theory,but the potential chattering phenomenon resulted from the discontinuous control limits its application in robot control.This thesis aims to present several effective controls for the improved performance of robotic systems with sliding mode control.Three approaches are proposed within the frame of first-order and second-order sliding mode control methodology.First,to facilitate the control design,an inverse dynamic decoupling model is firstly formed,and a first-order linear sliding mode control is formulated.Second,for a fast transient and high steady-state tracking precision,a nonlinear sliding surface is constructed by introducing a nonlinear function,and a first-order nonlinear sliding mode control is proposed.The global asymptotic stability of the closed-loop system with these two first-order sliding mode controls is proven by Lyapunov's direct method.Finally,taking the advantages of high-order sliding mode control,the nonlinear sliding surface is integrated into the conventional used second-order sliding mode control,and thus an improved nonlinear second-order sliding mode control is proposed.The proposed nonlinear second-order sliding mode control keeps the appealing features of the conventional used second-order sliding mode control such as lower chattering and higher robustness to uncertainty and bounded external disturbances as well as the easiness of practical implementation.The control law designed in this research adopts two-degree-of-freedom robot as a model was simulated in Matlab,and the feasibility of the control law was verified and the control effect of the control law was compared.The validity and performance improvement of the three approaches are demonstrated with numerical simulation comparisons performed on a two-degree-of-freedom robot.The simulations results show that the designed first-order nonlinear sliding mode control can solve the global asymptotic stable tracking of robot manipulators subject to partial parametric uncertainties and bounded external disturbances.Benefitted from the introduction of the nonlinear function,the proposed first-order nonlinear sliding mode control has faster convergence speed and higher steady-state tracking accuracy than the traditional linear sliding mode control.he introduction of nonlinear design in traditional sliding mode control can be more suitable for the accuracy requirement of the robot system control.Similar to the conventional second-order sliding mode control,the proposed nonlinear second-order sliding mode control produces a smooth control action and higher robustness to uncertainty and bounded external disturbances and gains a faster transient and higher steady-state tracking precision.The proposed approach of this thesis actually provides an effective easy-going solution for high-precise motion control of nonlinear robots. |
