Disturbances and model uncertainties in Iterative Learning Control

A high performance feedforward signal for systems that perform repetitive tasks can be derived using Iterative Learning Control (ILC). Disturbances present in the feedback control loop, such as measurement and load disturbances, enter the learning process and deteriorate the performance of ILC. Model uncertainties also influence the learning process and the minimum achievable error of ILC. This paper presents an expression for the tracking error of an arbitrary iteration that shows the influence of measurement disturbances, load disturbances and model uncertainties throughout the learning process. The tracking error is proven to be determined to a great extend by both the disturbances of the last two iterations and the model uncertainties.