Data-driven feedback-feedforward controller design from closed-loop step response data based on virtual reference feedback tuning

Abstract

This paper proposes data-driven feedback-feedforward controller design from closed-loop step response data. The proposed method directly tunes feedback controllers based on Virtual Reference Feedback Tuning (VRFT) using closed-loop step response data. Then, the feedforward controller is designed for disturbance attenuation from measurable disturbances. The feedback controller is linearly parametrized with integrator for assuring the asymptotically tracking to the step-type reference signal, and the feedforward controller is a parametrized rational transfer function. The optimal prefilter design are provided for the both feedback and feedforward controllers. The effectiveness of the proposed method is illustrated through a numerical example.