ObjectiveAfter the deployment of satellite antennas in orbit, they are prone to deformation and vibration due to various factors, which can affect observational performance. Single measurement methods struggle to meet the demands for high-precision in-orbit measurement of the antenna's surface shape in terms of both performance and environmental adaptability. To address this, this paper proposes a high-precision satellite panel deformation measurement method that combines a camera and a scanning laser rangefinder.
MethodsFirst, a mathematical model for the fusion measurement of a camera and scanning laser rangefinder was established, consisting of a camera, a two-dimensional turntable, and a laser rangefinder, and the coordinate system transformation relationships between each module were analyzed. Next, a high-precision pointing method for the laser rangefinder based on BSLO-d bio-inspired optimization was developed. This method uses the BSLO optimization algorithm to globally optimize the two large rotation angles of the turntable, achieving preliminary autonomous alignment of the laser axis, and then uses image plane information to precisely align the laser axis. Subsequently, a 3D coordinate calculation method for the fusion of the camera and scanning laser rangefinder was proposed. This method utilizes range data and image plane coordinates, along with camera extrinsic parameters, the calibrated laser exit point, and the laser axis direction vector to compute the 3D coordinates of target points. Finally, the feasibility of the pointing angle calculation and 3D coordinate measurement methods was verified through experiments.
Results and DiscussionsThe proposed method demonstrated high precision in both pointing angle calculation and 3D coordinate measurement. From the BSLO-d pointing angle optimization result (
Fig.5) and the pointing error experimental results (
Fig.6), it is evident that the BSLO-d pointing method can quickly and accurately find the optimal rotation angle with minimal pointing error. By calculating the pointing angles for seven target points using BSLO-d and comparing them with the angles obtained through manual alignment, the results showed that the root mean square errors (RMSE) for the horizontal and vertical angles were 0.435 mrad and 0.787 mrad, respectively (
Tab.1). In terms of 3D coordinate measurement, using the V-STARS high-precision photogrammetric measurement system as the ground truth, the measurement errors for 28 target points in the
X,
Y, and
Z directions were statistically analyzed. The results showed that the RMSE in the
X,
Y, and
Z directions were 0.948 mm, 0.268 mm, and 0.127 mm, respectively (
Tab.2). From the error charts for the
X,
Y, and
Z coordinates of the 28 points (
Fig.7), it is clear that this fusion measurement method not only improves measurement precision but also demonstrates high stability. Additionally, the method allows for full-field measurement of the target points. Furthermore, from the experimental setup diagram (
Fig.4), it can be seen that the measurement system is compact, easy to assemble, and calibrate.
ConclusionsThis study addresses the autonomous measurement of satellite antenna deformation after deployment in orbit by proposing a 3D coordinate system that fuses a camera with a rotating scanning laser rangefinder. The system accurately integrates data from multiple sensors, and the feasibility and accuracy of the method were experimentally validated. The method successfully achieves autonomous alignment of the laser rangefinder through visual guidance via a non-orthogonal two-dimensional turntable, and computes the 3D coordinates of target points using distance and image plane data. Experimental results demonstrate that this method is fast, accurate, and provides high precision in both pointing angle calculation and 3D coordinate measurement. Additionally, the system is compact, easy to assemble, and calibrate. Thus, this method offers a new solution for high-precision in-orbit measurement of satellite antenna surfaces and provides a new technical path for other application scenarios.