On-orbit calibration of multi-view photogrammetry system for satellite antenna measurement

Abstract To ensure the normal operation of the payload on the FY-4M geostationary microwave satellite, the satellite is equipped with a four-camera multi-view photogrammetry system for real-time measurement of antenna surface deformations on orbit. During on-orbit operation, the system parameters may drift, and due to limitations in camera field of view, spatial arrangement constraints, and the large antenna aperture, the observable geometric constraints are limited, making conventional calibration methods difficult to apply directly. To address this issue, this study proposes an on-orbit self-calibration method for close-range photogrammetry systems aimed at measuring large-scale satellite antenna surfaces. Taking the FY-4M application as the design and validation context, stable control points under on-orbit conditions are extracted based on the antenna’s deployable structure and the thermal expansion characteristics of its components. The initial extrinsic parameters of the cameras are quickly estimated using four control points, and the relative distance between camera stations is incorporated as a virtual observation to impose a soft scale constraint, forming a complete BA model for on-orbit self-calibration. The Levenberg–Marquardt algorithm is employed for nonlinear optimization, and adaptive weights based on the reprojection error are constructed to enhance the accuracy and robustness of the solution. The method is validated under ground laboratory conditions using both simulations and a prototype multi-view photogrammetry system developed by the Shanghai Institute of Technical Physics Chinese Academy of Sciences, in which on-orbit parameter variations are simulated. Results demonstrate that the proposed approach can effectively calibrate the photogrammetry system under weakly constrained conditions, suppress the impact of parameter drift on antenna surface measurement accuracy, and exhibit strong engineering applicability and practical value. The method also provides a useful reference for the calibration of other large-scale space structure photogrammetry systems.