Robustness of uncertain discrete linear repetitive processes with disturbance attenuation

Repetitive processes are a class of 2D systems that can be used to model physical systems and also there are applications, such as iterative learning control, where using a repetitive processes setting for design has advantages over alternatives. In most cases, it is discrete dynamics that are basis for design and often there will be a need to deal with uncertainty in the process model and the effects of disturbances. This paper develops new algorithms for these tasks that are computed using linear matrix inequalities when the uncertainty is modeled using two standard representations and disturbance attenuation is measured by an norm. The extension of these algorithms to control law design is also developed.