Adaptive Sliding Mode Force Tracking Approach for Unstructured Environment

Abstract

This manuscript presents a novel force tracking scheme that focuses on solving the problem of compliant interaction during contact between robot and unstructured environment. Constant interaction force error compensation lacks adaptive capability for robot, which cannot effectively compensate for unstructured environment. Contrarily, the proposed adaptive variable impedance control method based on sliding mode structure (AVIS) can steadily track desired force during interaction process. Sliding mode control based on approximation law is used to adaptively modify damping and stiffness parameters of impedance control online, which continuously compensates for nonlinear dynamic system with high uncertainty. Stability and convergence of AVIS are proven. Experiments on diverse force interaction tasks show that stable and robust force tracking performance can be achieved by AVIS.