https://doi.org/10.1140/epjs/s11734-023-01014-0
Regular Article
Performance analysis of derivative-free estimation methods from the perspective of attitude estimation influenced by real data
1
Lorena School of Engineering (EEL), University of São Paulo (USP), Estrada Municipal do Campinho, S/N. Ponte Nova, 12602-810, Lorena, São Paulo, Brazil
2
Space Mechanics and Control Division (DMC), National Institute for Space Research (INPE), Av Dos Astronautas, 1758, Jd. da Granja, 12227-010, São José dos Campos, São Paulo, Brazil
3
Gama Campus (FGA), University of Brasilia (UnB), Área Especial de Indústria, Projeção A, Setor Leste (Gama), 72444-240, Brasília, Federal District, Brazil
4
Engineering, Modeling and Applied Social Sciences Center (CECS), Federal University of ABC (UFABC), Av. dos Estados, 5001, Bangú, 09210-580, Santo André, São Paulo, Brazil
5
Mathematics Department (DM), São Paulo University (UNESP), Av. Ariberto Pereira da Cunha, 333, Pedregulho, 12516-410, Guaratinguetá, São Paulo, Brazil
6
Collaborative Laboratory (CoLAB), Center of Engineering and Product Development (CEiiA), PACT, Rua Luís Adelino Fonseca, 1, 7005-841, Évora, Portugal
Received:
8
May
2023
Accepted:
9
November
2023
Published online:
7
December
2023
The main difference between the Extended Kalman Filter (EKF) and the non-linear estimators that make use of the so-called Sigma Points is the need or not of linearizing the equations that compose the whole dynamic system, a process that requires the calculation of the Jacobian matrices composed of partial derivatives. In this study, an analysis of the Central Difference Kalman Filter (CDKF) efficiency is performed, as compared to other derivative-free estimators and the standard EKF, when real data from on-board satellite sensors are processed by the filters. However, the use of real data can generate problems, not only regarding errors and uncertainties of different natures that can lead the filter to inaccurate results, but also regarding the difficulty in validating the results due to the absence of reference values. In this case, results of the attitude estimated by filters, such as EKF, Unscented Kalman Filter (UKF), and Cubature Kalman Filter (CKF), already validated in previous papers served as the basis for the comparisons made with the CDKF. It was observed that the performance of CDKF is superior to the conventional EKF and equivalent to filters that make use of the sigma points, while still maintaining an adequate processing time for real applications.
Hélio K. Kuga, William R. Silva, Leandro Baroni, Maria C. F. P. S. Zanardi, and Paula C. P. M. Pardal have contributed equally to this work.
Copyright comment Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
© The Author(s), under exclusive licence to EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature 2023. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
