Observer-based Adaptive Sliding Mode Tracking Control for Soft Bending Actuators

Abstract

This paper investigates the challenging problem of finite-time position tracking control problem for the soft bending actuators without angular velocity measurements, in presence of unknown model uncertainties and external disturbance. To improve the performance of system and simplicity of controller, this paper utilizes an accelerated super-twisting algorithm (STA) to construct a gain-adaption observer and a controller simultaneously. By the observer, the condition on the derivative of uncertainties is relaxed, and the adaptive law leads to attenuate the observer gains overestimation. Next, an observer-based adaptive STA control strategy is designed to attenuate chattering phenomenon and be robust for the system uncertainties. The finite-time convergence of the proposed control strategy is proved by the Lyapunov method. Under the proposed controller, the tracking errors and estimation errors all tend to zero in finite time in the presence of external disturbance, and all the signals and states are bounded. Simulation results verify the proposed controller.