Tracking Error Constrained Super-twisting Sliding Mode Control for Robotic Systems

Tracking Error Constrained Super-twisting Sliding Mode Control for Robotic Systems

In this paper, we address a robust super-twisting sliding mode control that ensures finite-time convergence and the prescribed tracking error constrained performance of a robotic system. A tracking error-transformed variable based on a super-twisting sliding mode surface is constructed for achieving the transient and steady-state time performances of the position of the robotic manipulator, and satisfying the properties of an ordinary sliding mode control. The proposed controller performs the constrained positioning of the robot manipulator without using a complex modeling process or intelligent nonlinear approximations for unknown robot dynamics and uncertainty. The problem of difficult separation of an unknown nonlinear function into known and unknown parts in the conventional variable super-twisting algorithm (STA) is overcome by designing a convenient adaptive law for the unknown function. Thus, the proposed control approach provides both finite-time convergence and constrained tracking error boundedness within a prescribed position range of the robotic manipulator. The effectiveness of the proposed robust control scheme is demonstrated through its experimental application to an industrial articulated robot manipulator and a XY robotic manipulator.

In this paper, we address a robust super-twisting sliding mode control that ensures finite-time convergence and the prescribed tracking error constrained performance of a robotic system. A tracking error-transformed variable based on a super-twisting sliding mode surface is constructed for achieving the transient and steady-state time performances of the position of the robotic manipulator, and satisfying the properties of an ordinary sliding mode control. The proposed controller performs the constrained positioning of the robot manipulator without using a complex modeling process or intelligent nonlinear approximations for unknown robot dynamics and uncertainty. The problem of difficult separation of an unknown nonlinear function into known and unknown parts in the conventional variable super-twisting algorithm (STA) is overcome by designing a convenient adaptive law for the unknown function. Thus, the proposed control approach provides both finite-time convergence and constrained tracking error boundedness within a prescribed position range of the robotic manipulator. The effectiveness of the proposed robust control scheme is demonstrated through its experimental application to an industrial articulated robot manipulator and a XY robotic manipulator.