All-parameter calibration method of the on-orbit multi-view dynamic photogrammetry system.

Area of Science:

• Space Engineering
• Photogrammetry
• Optical Measurement

Background:

• Photogrammetry (PG) provides accurate data for evaluating large space structures.
• On-orbit Multi-view Dynamic Photogrammetry Systems (OMDPS) require spatial reference data for camera calibration and orientation.

Purpose of the Study:

• To propose a multi-data fusion calibration method for OMDPS to address the lack of spatial reference data.
• To enable full-parameter calibration for OMDPS, including camera interior parameters (CIP), exterior parameters (CEP), and lens distortion parameters (LDP).

Main Methods:

• Developed a multi-camera relative position model using star and scale bar targets to resolve reference camera position constraints.
• Employed a two-norm matrix and weight matrix to adjust the Jacobian matrix, overcoming bundle adjustment failures and inaccuracies.
• Optimized all system parameters simultaneously through the proposed algorithm.

Main Results:

• Ground-based experiments with 333 spatial targets measured by OMDPS and V-star System (VS).
• OMDPS achieved root-mean-square errors (RMSE) of less than 0.0538 mm (in-plane Z-direction) and 0.0428 mm (Z-direction).
• Out-of-plane Y-direction RMSE was less than 0.1514 mm, validating the system's accuracy.

Conclusions:

• The developed multi-data fusion calibration method effectively addresses OMDPS calibration challenges.
• The photogrammetry system demonstrates significant potential for precise on-orbit measurement tasks.
• The method ensures simultaneous optimization of all system parameters for enhanced accuracy.