A Reduction Paradigm for Output Regulation

The goal of this paper is to provide a reduction paradigm for the design of nonlinear output regulators which can be of interest for nonlinear as well as linear uncertain systems. The main motivation of the work is to provide a systematic design tool to deal with non-minimum-phase uncertain systems for which conventional high-gain stabilization paradigms are not effective. The contribution of the work is two-fold. First this work extends the nonlinear framework proposed in Marconi et al. (Systems & Control Letters 53 2004) to the case in which the so-called ``immersion into a linear system" does not hold. Furthermore, in the case the uncertain controlled dynamics are linear, we show how the proposed framework leads to a number of systematic design tools of interest in the case of non-minimum phase linear systems affected by severe uncertainties. A numerical example of output regulation of a linearized model of an inverted pendulum on a cart is presented.