Control of a two-link manipulator using disturbance observer-based model predictive control

Abstract

This paper addresses position and path-tracking problems of robot manipulators subject to constraints on both the magnitude and difference of the control input. Along with the model predictive control (MPC), the disturbance observer (DOB) is used to compensate for disturbances and nominalize the plant dynamics. To constrain the total control input, which is the sum of the MPC output and DOB output, we adopt time-varying input constraints and construct a useful structure of the combined DOB-MPC system. The effectiveness of the proposed control scheme is validated through a few experiments using a real two-link manipulator.