A stellar gyroscope is a star based attitude propagator that is capable of propagating a spacecraft’s attitude in three degrees of freedom by tracking the motion of the stars in an imager's field of view. The modeling and algorithm development has been done by the Space Systems Laboratory at the University of Kentucky. This paper discusses a realization of the stellar gyroscope concept on a CubeSat attitude determination and control system (ADCS) designed by SSBV Space & Ground Systems UK. The stellar gyroscope can be used to measure attitude changes from a known initial condition without drift while sufficient stars are common across frames, because absolute attitude changes are measured and not angular rates. Algorithms to perform the star detection, correspondence, and attitude propagation are presented in this paper. The Random Sample Consensus (RANSAC) approach is applied to the correspondence problem which is challenging due to spurious false-star detections, missed stars, stars leaving the field of view, and new stars entering the field of view. The CubeSat attitude determination and control system described in this paper uses a stellar gyroscope, implemented using inexpensive optics and sensor, to augment a MEMS gyroscope attitude propagation algorithm to minimize drift in the absence of an absolute attitude sensor. The MEMS device provides the high frequency measurement updates required by the control system, and the stellar gyroscope, at a lower update rate, resets the drift accumulated in the MEMS inertial gyroscope integrator. This in effect could allow sun-sensing satellites to maintain a high quality attitude estimate in eclipse, where the sun sensors can no longer contribute in absolute attitude estimates. This paper describes an algorithm to solve the relative attitude problem by identifying the change in attitude between two star field images. RANSAC is applied to solve the correspondence problem in the presence of false star detections and misses. The camera and attitude determination and control system are described, prototype hardware is used to generate night-sky datasets of known attitude changes to demonstrate the performance of the algorithm, and a simulation is developed to evaluate the stellar gyroscope’s ability in limiting the drift of an attitude propagator based on MEMS gyroscope rates. The CubeSat ADCS system developed by SSBV is an experiment on TechDemoSat-1, to be launched in early 2013.
[1]
James R. Wertz,et al.
Spacecraft attitude determination and control
,
1978
.
[2]
Robert C. Bolles,et al.
Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography
,
1981,
CACM.
[3]
Marcel J. Sidi,et al.
Spacecraft Dynamics and Control: A Practical Engineering Approach
,
1997
.
[4]
Bong Wie,et al.
Space Vehicle Dynamics and Control
,
1998
.
[5]
Taejung Kim,et al.
Automatic satellite image registration by combination of matching and random sample consensus
,
2003,
IEEE Trans. Geosci. Remote. Sens..
[6]
John L. Crassidis,et al.
Spacecraft Angular Rate Estimation Algorithms For Star Tracker-Based Attitude Determination
,
2003
.
[7]
K. Gromov,et al.
Toward a stellar gyroscope for spacecraft attitude determination
,
2004
.
[8]
A. Katake,et al.
Development and Testing of the StarCam SG100: A Stellar Gyroscope
,
2008
.
[9]
Jan-Michael Frahm,et al.
A Comparative Analysis of RANSAC Techniques Leading to Adaptive Real-Time Random Sample Consensus
,
2008,
ECCV.
[10]
Tom Dzamba,et al.
Towards Star Tracker Only Attitude Estimation
,
2010
.
[11]
S. Rawashdeh.
PASSIVE ATTITUDE STABILIZATION FOR SMALL SATELLITES
,
2010
.
[12]
S. Rawashdeh.
Development of a Drift-Free Stellar Gyroscope
,
2011
.
[13]
J. Lumpp,et al.
Design of a Stellar Gyroscope for visual attitude propagation for small satellites
,
2012,
2012 IEEE Aerospace Conference.