Adaptive finite‐time super‐twisting sliding mode control for robotic manipulators with control backlash

Zeyu Li,Junyong Zhai,Hamid Reza Karimi
DOI: https://doi.org/10.1002/rnc.5744
IF: 3.8973
2021-08-23
International Journal of Robust and Nonlinear Control
Abstract:This article aims at the problem of trajectory tracking for industrial robotic manipulators with control backlash. An arctangent terminal sliding mode surface is developed to deal with the lumped disturbance and enhance the robustness of the system. A novel adaptive super-twisting sliding mode control method is developed to achieve fast convergence and continuous control. The chattering in control law is surmounted by using super-twisting method. The lumped disturbance with unknown upper bound is compensated with the help of the adaptive technique. Based on the Lyapunov stability theory, the sliding mode surface will be arrived in a finite time and the trajectory tracking error will converge to zero in a finite time. The feasibility of the proposed control scheme is validated through an example of a two-link robotic manipulator.
automation & control systems,engineering, electrical & electronic,mathematics, applied
What problem does this paper attempt to address?
This paper aims to solve the problem of trajectory tracking for industrial robot manipulators in the presence of control backlash. Specifically, the paper develops an arctangent terminal sliding mode surface (ATSMS) - based method to handle cumulative disturbances and enhance the robustness of the system. In addition, a novel adaptive super - twisting sliding mode control (ASTSMC) method is proposed to achieve fast convergence and continuous control. By using the super - twisting method, the chattering phenomenon in the control law is overcome, and the cumulative disturbances with unknown upper bounds are compensated by adaptive techniques. Based on Lyapunov stability theory, it is proved that the sliding mode surface will be reached within a finite time and the trajectory tracking error will converge to zero within a finite time. Finally, the effectiveness of the proposed control scheme is verified by an example of a two - link robot manipulator.