On-Orbit Calibration of Installation Matrix between Remote Sensing Camera and Star Camera Based on Vector Angle...

Area of Science:

• Geospatial science
• Remote sensing technology
• Photogrammetry

Background:

• Accurate exterior orientation elements are crucial for satellite photogrammetry without ground control points (GCPs).
• Existing methods rely on GPS for centimeter-level accuracy of line elements and star cameras for angle elements, but installation matrices require recalibration due to environmental changes.
• The variability of the installation matrix necessitates robust calibration methods for long-term on-orbit data.

Purpose of the Study:

• To develop an accurate calibration method for the installation matrix of remote sensing cameras.
• To improve geo-positioning accuracy in satellite photogrammetry by removing the coupling of attitude and installation matrix.
• To reduce coordinate system conversion errors in remote sensing applications.

Main Methods:

• Estimating the cosine angle vector invariance between a remote sensing camera and a star camera, independent of satellite attitude.
• Utilizing batch processing on long-term on-orbit data for accurate installation matrix calibration.
• Leveraging star camera data for precise direction reference in the inertial coordinate system.

Main Results:

• The proposed method effectively calibrates the installation matrix without relying on ground control points (GCPs).
• Simulation results show a remarkable improvement in planimetric geo-positioning accuracy compared to conventional methods.
• The method successfully removes the coupling between attitude and installation matrix, enhancing overall accuracy.

Conclusions:

• The developed technique provides a robust solution for accurate on-orbit calibration of satellite remote sensing cameras.
• This approach significantly enhances geo-positioning accuracy, enabling more precise mapping and analysis.
• The method offers a valuable advancement for photogrammetry applications requiring high precision without GCPs.