STATE ESTIMATION BASED ON STOCHASTIC AND ZONOTOPIC APPROACHES: PART I - LINEAR SYSTEMS

ALESI  AUGUSTO DE PAULA; BRUNO  OTÁVIO SOARES TEIXEIRA; GUILHERME  VIANNA RAFFO

doi:10.20906/CBA2022/437

ALESI AUGUSTO DE PAULA UFMG - Universidade Federal de Minas Gerais
BRUNO OTÁVIO SOARES TEIXEIRA UFMG - Universidade Federal de Minas Gerais
GUILHERME VIANNA RAFFO UFMG - Universidade Federal de Minas Gerais

DOI: https://doi.org/10.20906/CBA2022/437

Keywords: Zonotopic itering, State estimation, Linear systems, Kalman filter

Abstract

This paper presents a comparative review on the most usual stochastic and zonotopic ltering methods in the literature for state estimation of uncertain linear systems. The mean and the condence level of the Gaussian random variable are compared to the center and uncertainty of the zonotopic variable. To achieve that, a unied notation for these approaches is proposed. On one hand, the Kalman filter is an algorithm often used for treating states as Gaussian variables, whose probability density function is simple to represent. On other
hand, the estimation of zonotopic states has owned relevance in the literature due to intrinsic properties of sets, which guarantee inclusion of the exact states into the estimated sets and improved computational burden on the computation of domains.