Poster
28 August 2024 Congratulations, it's twins: a digital twin for your high-contrast instrument? Part 2
Sebastiaan Y. Haffert, Vincent Chambouleyron, Vincent Deo, Jared R. Males, Olivier Guyon, Rebecca Jensen-Clem, Bruce Macintosh
Author Affiliations +
Conference Poster
Abstract
The next generation of extremely large telescopes will revolutionize the field of direct imaging. Imaging Earth-like exoplanets with upcoming instruments requires extremely precise control and knowledge of the incoming wavefront because any residual will leak through the coronagraph and create speckle noise. We propose to develop a high-fidelity digital twin of the instrument. Accurate digital twins can have a far-reaching impact: these models are crucial for the inverse problem of non-linear wavefront reconstruction, and they enable the reconstruction of coronagraphic speckle fields with wavefront sensor telemetry. We propose to use advances in machine learning to create a differentiable optical model. Results from this approach have shown that the reconstruction error can approach the machine precision limit even well into the non-linear range. We will show case the capabilities for two specific systems: the unmodulated pyramid wavefront sensor and post-coronagraphic phase retrieval. We will also showcase the performance of these models on the MagAO-X instrument and the AO testbed at the Santa Cruz Extreme Adaptive Optics Lab.
(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Sebastiaan Y. Haffert, Vincent Chambouleyron, Vincent Deo, Jared R. Males, Olivier Guyon, Rebecca Jensen-Clem, and Bruce Macintosh "Congratulations, it's twins: a digital twin for your high-contrast instrument? Part 2", Proc. SPIE 13097, Adaptive Optics Systems IX, 130976O (28 August 2024); https://doi.org/10.1117/12.3020817
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
Back to Top