Iterative motion feedforward tuning: A data-driven approach based on instrumental variable identification

Abstract

a b s t r a c tFeedforward control can significantly enhance the performance of motion systems through compensation of known disturbances. This paper aims to develop a new procedure to tune a feedforward controller based on measured data obtained in finite time tasks. Hereto, a suitable feedforward parametrization is introduced that provides good extrapolation properties for a class of reference signals. Next, connections with closed-loop system identification are established. In particular, instrumental variables, which have been proven very useful in closed-loop system identification, are selected to tune the feedforward controller. These instrumental variables closely resemble traditional engineering tuning practice. In contrast to pre-existing approaches, the feedforward controller can be updated after each task, irrespective of noise acting on the system. Experimental results confirm the practical relevance of the proposed method.

On inversion-based approaches for feedforward and ILC

“…In feedforward, generally high performance in terms of the error e is pursued. Typically, this is enforced by minimizing the energy in the error signal through minimizing e 2 [23,4]. In ILC, the main concern is to guarantee convergence in the error to ensure stability over trails.…”

“…The aim of this paper is to investigate, compare, and develop inversion techniques for the purpose of both feedforward and learning control. The model to be inverted includes the closed-loop process sensitivity in iterative learning control (ILC) [1,2], the closed-loop complementary sensitivity in repetitive control [3], or the open-loop system in inverse model feedforward [4]. For nonminimum-phase or strictly proper systems, such inversion is not always straightforward.…”

Constrained Iterative Feedback Tuning for Robust Control of a Wafer Stage System

“…Remark 1 Though C f b and C f f can be optimized jointly, in this paper the feedforward controller C f f will not be subject [3], [30].…”

“…This has led to the development of databased approaches in [30], and [13] for the multivariable case, which induce significant performance enhancements. It has recently been shown in [3] that the IFT algorithm employed in [30] for feedforward control has a direct system identification interpretation. In fact, for the case of feedforward control, the IFT algorithm deals with a closed-loop identification problem, but comes at the expense of efficiency.…”

“…In fact, for the case of feedforward control, the IFT algorithm deals with a closed-loop identification problem, but comes at the expense of efficiency. In [3], more efficient algorithms have been proposed rendering the use of IFT superfluous for feedforward design. Still, IFT seems appealing for feedback design.…”