Disturbance Observer Based Model Prediction Control for a 2-DOF Nanopositioning Stage.

Min Ming,Zhao Feng,Jie Ling,Xiaohui Xiao
DOI: https://doi.org/10.1109/iecon.2019.8927729
2019-01-01
Abstract:To compensate the nonlinear effects of a nanopositioning stage and its model uncertainties, this paper presents a composite controller by integrating disturbance observer (DOB) and model prediction control (MPC). The nonlinearity and dynamic cross-coupling effects are treated as unknown disturbances to the system, and a DOB is used to partially estimate and compensate the disturbance to improved performance. A MPC is commonly used to track a reference signal, which minimizes the steady-state tracking error as well as increases the closed-loop bandwidth. Inspired by existing MPC schemes to improve the positioning performance of the piezo-actuated stage, a two-input two-output (TITO) form of the MPC controller is designed in this work with the estimated residual error being used to modify the reference in real time for better performance. The novelty of the proposed TITO MPC lies not only the compensation for nonlinear effects but also the rejection of cross-coupling dynamics effects in a 2-DOF nanopositioning system. The effectiveness of the proposed controller is validated through experimental investigations on a commercial nanopositioner platform. Experimental results show that the proposed method can improve the tracking performance of the piezo-actuated stage, compared to the traditional MPC.
What problem does this paper attempt to address?