Sistemas de Control basados en Reset

Abstract

A reset control system is formed by a linear controller endowed with certain resetting mechanism, that sets to zero the state. This zeroing of some of the state coordinates is applied when certain condition holds. The condition that triggers resetting is usually the zero crossing of the tracking error. The idea of re-set control dates back to Clegg Integrator (CI) in 1958, and was adopted in the 70's by I. Horowitz, who emphasized the ability of reset systems to overcome fundamental limitations of linear systems with delays or with poles or zeros in the right half-plane. Left aside for some time, the idea was recovered in the last decade and was given a new impulse by several research teams that contribute with advances in analysis, design and application of reset control systems. This interest appears in the context of the present boom of hybrid systems: a reset system is a particular case of hybrid system that, in spite of its simplicity, enables the possibility of overcoming linear limitations. The objetive of this work is to present, in the form of a tutorial, the current panorama in the field. First, a historical review is introduced, on the three main approaches to reset control: the classical approach (based on zero-crossing of the tracking error), the impulsive approach (where reset is applied on a temporal basis) and the hybrid approach (where some sector reset condition is defined). Next, a summary of results is presented on analysis topics (relation to fundamental limitations, stability) and finally several design proposals are discussed, focused to the tuning of PI+CI compensators.