On structure selection for multivariable repetitive-predictive controllers

It is well known from the internal model control principle that in order to follow a periodic reference signal with zero steady-state error, the generator for the reference must be included in the stable closed-loop control system. The subject area is repetitive control systems where a periodic signal is to be followed and the starting point is previous work where a frequency decomposition approach was used in a model predictive control setting and termed repetitive-predictive control. This paper extends the previous work to multi-input and multi-output systems and investigates the issue of controller complexity with respect to performance loss and robustness to measurement noise. The results developed are applied to repetitive-predictive control of a robot arm where the model has been constructed from measured frequency response data.