Design of h ^∞-optimal feedback controllers for model following

In this paper we look for feedback that minimizes the model following error of a linear, single variable feedback system represented by its frequency response. Plant deviation from the reference model is measured in a weighted H^∞ norm Concluding remarks are drawn concerning the requirements on a process to perform as close to a reference model as possible within the constraints of physical realizability. The design is carried out in the framework of minimax optimal sensitivity theory pioneered by Zames (1981) and developed by Zames and Francis (1981) and their results are tailored to fit the reference model problem considered in this paper.