Norm-optimal control of time-varying discrete repetitive processes with iteration-varying reference inputs

This paper extends the norm-optimal control design methodology of iterative learning control (ILC) to the case of linear time-varying discrete repetitive processes (LTV-DRP) driven by known iteration-varying reference input signals. Our approach exploits the idea of the internal model principle for robust servomechanism design and uses a state-space model of LTV-DRPs introduced in our earlier work. A general state-feedback solution to the problem is derived that contains as special cases the solutions for time-invariant DRPs and iteration-invariant reference signals as well as the equivalent solutions for the ILC case. A numerical example illustrates the results.