Self-Calibration for Star Sensors.

Area of Science:

• Spacecraft attitude determination
• Optical sensor calibration
• Astro-navigation systems

Background:

• Star sensors are crucial for spacecraft attitude determination.
• On-orbit calibration is essential but often relies on pre-identified star patterns.
• Existing methods face challenges with intrinsic parameter accuracy and computational efficiency.

Purpose of the Study:

• To develop an on-orbit self-calibration method for star sensors independent of star identification.
• To enhance the accuracy of intrinsic parameter estimation, particularly focal length.
• To improve the convergence speed of the calibration process.

Main Methods:

• Derived self-calibration equations based on the invariance of interstar angles between image frames.
• Introduced a constant constraint on the optical path to minimize estimation bias.
• Employed a scaled nonlinear least squares method for efficient equation solving.

Main Results:

• Demonstrated a two-orders-of-magnitude reduction in focal length estimation bias with the proposed constraint.
• Achieved convergence within 10 iterations using the scaled nonlinear least squares method.
• Validated that calibrated parameters are directly usable in conventional star map identification.

Conclusions:

• The proposed method effectively addresses the chicken-and-egg problem in star sensor calibration.
• The technique offers a robust and efficient solution for on-orbit intrinsic parameter determination.
• This self-calibration approach enhances the reliability and applicability of star sensor data.