From Sim to 6DOF: Deep Learning for Real-Time Satellite Pose Estimation from Resolved Ground-Based Imagery
Year: 2026
Type: article
Abstract: This work presents the first complete system for automated six degrees of freedom (6DOF) satellite pose estimation from spatially resolved, ground-based, adaptive optics (AO)-corrected imagery, addressing a key challenge in Space Domain Awareness (SDA). The approach eliminates the need for human labeling by directly regressing satellite orientation and position from blurry, noisy, and deeply-shadowed imagery. A multi-stage deep neural network pipeline localizes the satellite, predicts pose, and optionally applies temporal smoothing. Networks are trained exclusively on fully synthetic imagery generated from a CAD model, yet generalize effectively to real data, bridging the Sim2Real domain gap. On 137 real, human-labeled test images of Seasat, the model achieved a mean rotation error of 5 degrees and a mean image-plane translation error of 21 cm. Qualitative evaluation of additional real Seasat imagery rated 178 of 199 predicted poses as “ground truth equivalent” or “high confidence match,” with zero catastrophic failures. The system was extended to seven degrees of freedom (7DOF) for satellites with articulating components and demonstrated on real Hubble Space Telescope (HST) imagery, achieving 5.9 degrees rotation error, 48 cm translation error, and 6 symmetry-adjusted solar array error on a 249-frame pass with temporal filtering. Across 586 real test images from Seasat and HST (captured over multiple decades under diverse conditions) the system consistently performed well. On a high-fidelity wave optics (HFWO) synthetic test set of Seasat, the model achieved 8.4 degrees mean rotation error, 34 cm image-plane translation error, and 1.4% range error at r0=6 cm and 1,031 km range. It outperformed human labeling in both accuracy (48% lower rotation error) and speed (800× faster at 7.1 Hz inference), while requiring < 40 hours on a single A100 GPU to train. The approach was also demonstrated for ARGOS, a smaller satellite with highly symmetric geometry. A GIQE-based image quality metric was introduced to forecast pose accuracy. General-purpose models like GPT-4o and Depth Anything V2 failed across most SDA tasks, but rapid gains in vision-language models warrant continued monitoring. These results establish a new operational baseline for practical, real-time satellite pose estimation from AO SDA imagery.
Source: Aerospace
Authors: Thomas Dickinson, Dawson Friesenhahn, Justin Fletcher, Derek Walvoord, Dennis Montera, Michael Gartley
Language: English
Subfield: Aerospace Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,896
OpenAlex ID: W7084203275
FWCI:
Cites:
Cited by:
Related to: