When developing a navigation system with an IMU and star tracker, the attitude alignment between the two sensors (relative pose) must be estimated so that measurements from both sensors can be converted to a common reference frame. Often such navigation systems use an on-board computer to combine the IMU's gyro and accelerometer measurements with the star tracker's attitude measurements using a Kalman filter. The proposed alignment method adds three states to this navigation filter to estimate the error in a given coarse alignment quaternion. On ground, the alignment quaternion can be accurately measured by navigating with the on-board system while the star tracker points towards the night sky and takes measurements. The vehicle is rotated in several positions to image the night sky from different angles. Each change of orientation provides the navigation system with additional measurements of gravity, Earth rate, angle slews and star tracker quaternions, all of which help improve the filter's estimate of the alignment quaternion. Alignment can also be done during the mission, however gravity and Earth rate measurements may not be available. Alignment accuracy depends on the IMU's error characteristics and the star tracker's accuracy. This technique is used for the Hybrid Navigation System and is compared to a traditional indoor alignment method which uses a checkerboard optical target.
[1]
P. Savage.
Strapdown Inertial Navigation Integration Algorithm Design Part 1: Attitude Algorithms
,
1998
.
[2]
M. Shuster.
A survey of attitude representation
,
1993
.
[3]
Gerd Hirzinger,et al.
Optimal Hand-Eye Calibration
,
2006,
2006 IEEE/RSJ International Conference on Intelligent Robots and Systems.
[4]
Arthur Gelb,et al.
Applied Optimal Estimation
,
1974
.
[5]
Stephan Theil,et al.
Attitude Determination for the SHEFEX 2 Mission Using a Low Cost Star Tracker
,
2009
.
[6]
James R. Wertz,et al.
Spacecraft attitude determination and control
,
1978
.
[7]
P. Savage.
STRAPDOWN INERTIAL NAVIGATION INTEGRATION ALGORITHM DESIGN. PART 2: VELOCITY AND POSITION ALGORITHMS
,
1998
.
[8]
Jay A. Farrell,et al.
Aided Navigation: GPS with High Rate Sensors
,
2008
.