Feedback-based iterative learning control for uncertain linear MIMO systems

This paper proposes a necessary and sufficient convergence condition for a feedback-based iterative learning control (ILC) system including an uncertain linear multi-input and multi-output (MIMO) plant to converge in the /spl Lscr//sub 2/-norm sense. To deal with various uncertainties, the uncertain plant is expressed by linear fractional transformation (LFT). After obtaining a convergence condition for the known part of the uncertain plant, we derive a convergence condition for the ILC system including the plant uncertainty The convergence condition is expressed in terms of the structured singular value (/spl mu/) of the nominal system and turns out to be the robust performance condition for the feedback control system.