Improved star identification algorithms and techniques for monochrome and color star trackers

• Ano de defesa: 2017
• Autor(a) principal: Márcio Afonso Arimura Fialho
• Orientador(a): Leonel Fernando Perondi, Daniele Mortari
• Banca de defesa: Ronan Arraes Jardim Chagas, Antônio Gil Vicente de Brum, Josiel Urbaninho de Arruda
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Instituto Nacional de Pesquisas Espaciais (INPE)
• Programa de Pós-Graduação: Programa de Pós-Graduação do INPE em Engenharia e Gerenciamento de Sistemas Espaciais
• País: BR
• Link de acesso: http://urlib.net/sid.inpe.br/mtc-m21b/2017/08.10.22.44

This work presents modeling, implementation, testing and simulation of a real-time attitude estimator using extended Kalman filter (FKE), applied to a three-axis air-bearing platform controlled by gas jet actuators. The goal of this work is to obtain an attitude determination algorithm that can replace the attitude estimate from the embedded inertial unit, currently in use by the platform electronics. Two models of estimators were implemented in MATLAB environment: in the first one, the state vector is composed only by the quaternions of the attitude, and in the second, the state vector is composed by quaternions and biases of the three gyroscopes. These estimators were tested using data from the sensors of the platforms inertial unit, and the results were compared with the attitude estimation sent by the internal attitude solution of the inertial unit. Finally, tests of the two FKE algorithms were performed in the simulation model of the air-bearing platform. Both Kalman filter-based estimators were successful in the attitude determination process, both in the tests using real data and in the simulation of control of the air-bearing platform. The proposed bias estimation functioned in a degraded way for the cases tested. We then hypothesized the reasons for this performance, and the convenience of estimating them in the proposed application.