Title: GPS Attitude Determination for VELOX-CI Near-Equatorial LEO Microsatellite
Author(s): Guo X. Lee, Fuxiang Cao, Yung-Fu Tsai, Kay Soon Low, Keck Voon Ling, Eng Kee Poh, Chin Siong Lim, Shi Tong Chin, Wee Seng Lim, Bingxuan Li
Published in: Proceedings of the 29th International Technical Meeting of The Satellite Division of the Institute of Navigation (ION GNSS+ 2016)
September 12 - 16, 2016
Oregon Convention Center
Portland, Oregon
Pages: 902 - 908
Cite this article: Lee, Guo X., Cao, Fuxiang, Tsai, Yung-Fu, Low, Kay Soon, Ling, Keck Voon, Poh, Eng Kee, Lim, Chin Siong, Chin, Shi Tong, Lim, Wee Seng, Li, Bingxuan, "GPS Attitude Determination for VELOX-CI Near-Equatorial LEO Microsatellite," Proceedings of the 29th International Technical Meeting of The Satellite Division of the Institute of Navigation (ION GNSS+ 2016), Portland, Oregon, September 2016, pp. 902-908.
Full Paper: ION Members/Non-Members: 1 Download Credit
Sign In
Abstract: In this paper, an attitude determination system using three GPS antennas on the VELOX-CI microsatellite is presented. Attitude determination using GPS is typically challenging on a microsatellite due to its small form factor which limits the length of its GPS antenna baselines and the accuracy of the attitude fix. The baselines of the three GPS antennas mounted on the top platform of the VELOXCI satellite are only 26.5 cm, 26.5 cm and 47.8 cm. In addition, due to the dynamics and high velocity of a LEO satellite, the number of common satellites for antennas in a baseline is fast changing and may be insufficient for reliable integer ambiguity resolution. In order to perform the dual missions of GPS attitude determination as well as radio occultation for VELOX-CI, the centre GPS antenna is mounted facing up while the other antennas are mounted facing to the side at an angle of 25.9 degrees to the centre antenna. This configuration further reduces the number of common visible satellites between the antennas in each baseline. In order to overcome these challenges, the visibility of the GPS satellites is predicted for each antenna using the satellites’ two-line elements (TLE) for each orbit while assuming that the satellite is in the sun-pointing mode. The time frames in which there are at least four common satellites between the antennas in each baseline are identified. The on-board receivers are then scheduled from the ground station to log the GPS L1 and L2 pseudorange and carrier phase observations at a sampling rate of 1 Hz during these periods. The collected data are downloaded for post-processing on subsequent satellite ground passes. To estimate the attitude, double differences are first performed on the observables to reduce the common satellite and receiver clock errors. The linearized doubledifference observation equation is then solved using the Least squares AMBiguity Decorrelation Adjustment (LAMBDA) method for the baseline coordinates and carrier phase integer ambiguities. The known baseline lengths are used as an additional constraint for evaluating the success of ambiguity resolution. The attitude of the satellite is then estimated using direct computation or least square estimation. The estimated attitude using GPS is compared with both the predicted attitude using the satellite’s TLE and sun model as well as the estimated attitude obtained from the flight control system using the satellite’s sun-sensor, magnetometer and gyroscope. The experimental results preliminarily validated our algorithm for GPS attitude determination and demonstrated the feasibility of attitude determination using multiple GPS antennas for a LEO microsatellite.