Corresponding author: Zadorozhnaya Natalia M., Ph. D, Associated Professor, Bauman Moscow State Technical University, Moscow, 105005, Russian Federation,
e-mail: zanatalie@ya.ru

Received on January 16, 2017
Accepted on January 27, 2017

The on-board satellite attitude and orbit control systems (AOCS) equipped with the inertial measurement units (IMU) and star trackers (ST) are the most versatile and reliable. Simultaneous processing of IMU and ST measurements is common for the foreign scientific spacecraft, which makes the basis of the operating principle of both devices. IMU functions in a wide angular rate range and doesn't depend on the external conditions. However, in case the measurement correction by another source of information is not available, the accuracy of the attitude determination might be reduced. The mathematical fusion of IMU and ST measurements allows AOCS to estimate with high accuracy and frequency the attitude and the angular rate parameters. This paper presents a suboptimal algorithm based on reduced Kalman filter and designed for IMU and ST measurements fusion. The Kalman filter state vector was chosen as a three dimensions vector. This vector is the Rodrigues parameters of the attitude between the attitude estimated by ST and the attitude estimated by IMU angular rate integration. Such a state vector has a minimal dimension for that purpose. Also it allows us to estimate the attitude parameters without additional information lag for the stabilization subsystems. Another feature of the algorithm is a different computation frequency of the state vector estimation and gain matrix. The gain matrix and the covariance matrix are 5 times slower than the state vector estimation. This algorithm performs an efficient decrease of the noise equivalent angle. Also this algorithm is not characterized by a computational complexity and significant AOCS computer requirements, which are 2,7 times lower compared with the full Kalman filter. The algorithm can function with various IMU and ST obtained by the state vector provided by the Rodrigues parameters of the attitude between ST attitude and IMU attitude. The algorithm is a part of AOCS software, verified on AOCS test bench. The test bench includes models of SED26 star tracker and GIVUS KIND34-020 IMU. Due to the new algorithm and with this equipment AOCS attitude error is lower than 5 arc. seconds.

For citation:

Ryabogin N. V., Sokolov V. N., Zadorozhnaya N. M. Inertial Measurement Units and Star Tracker Measurements Fusion on the Basis of the Rodrigues Parameters Estimation, Mekhatronika, Avtomatizacia, Upravlenie, 2017, vol. 18, no. 5, pp. 351—359.
DOI: 10.17587/mau. 18.351-359

To the contents