Data-Based Approaches to Trackability Compensation for Learning Control Systems

This article aims to propose two data-based trackability compensation strategies for iterative learning control (ILC) systems to improve their tracking ability. By designing the input-output tests for the sample data collection, a data-based criterion for trackability is exploited, under which the trackability sets for ILC systems can be further determined. From the task level and set level, two classes of compensation strategies are developed by adopting the interconnection techniques, respectively, to modify the trackablity sets for specific ILC systems subject to different requirements. Consequently, the tracking ability of ILC systems is enhanced, based on which the better tracking performance can be achieved. The developed theoretical results are supported by illustrative simulations.