Zonotopic and gaussian state estimator: the predictor and filter algorithms
Carregando...
Data
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Federal de Minas Gerais
Descrição
Tipo
Artigo de evento
Título alternativo
Primeiro orientador
Membros da banca
Resumo
In the literature, the state estimator for linear systems called zonotopic and Gaussian Kalman estimator was introduced by Combastel in 2015. It was proposed as a predictor and presented in the one-step form. The difference between a predictor and a filter is the measurement sequence used to estimate states. Predictors use the past measurements, while filters use both past and current measurements. Also, in the one-step form, the states are estimated at once. In the two-step form, the states are estimated according to the forecast and data-assimilation steps, that is, the predictive/corrective structure is made explicit. This paper revisits such estimator, here called ZGKP, and proposes two novelties. First, the two-step form is introduced for the ZGKP. After, the filter version to the estimator is proposed, here called ZGKF, in the two equivalent forms. The advantages by using the filter in comparison to the predictor are discussed.
Abstract
Assunto
Sistemas lineares, Kalman, Filtragem de
Palavras-chave
Kalman filter; set-based estimation; linear systems; predictors; filters
Citação
Departamento
Curso
Endereço externo
https://proceedings.science/sbai-2019/trabalhos/zonotopic-and-gaussian-state-estimator-the-predictor-and-filter-algorithms?lang=pt-br