A new article written by members of the Optoelectronics Laboratory of PUCV was published in the well known scientific journal “Photonics Research”.
Felipe Guzmán, Jorge Tapia, Camilo Weinberger, Nicolás Hernández, and Professor Esteban Vera from the School of Electrical Engineering at PUCV, along with Jorge Bacca from the Universidad Industrial de Santander in Bucaramanga, Colombia, and Professor Benoit Neichel from the Laboratory of Astrophysics in Marseille (LAM), France, published “Deep Optics preconditioner for modulation-free pyramid wavefront sensing”.
The research conducted by the Optolab members, including graduate students, postdoctoral researchers, and professors from Pontificia Universidad Católica de Valparaíso, was published in the scientific journal Photonics Research, which belongs to the distinguished Optica Publishing Group and has an impact factor of 7.6.
As a result of this research, the discovery of an optical element called a preconditioner, which enhances the performance of a traditional wavefront sensor such as the pyramid, enabling better estimation and correction of light turbulence and atmospheric aberrations, was established using deep learning tools.
The use of deep learning tools in wavefront sensing is a recent development that has not been extensively explored, so results like those presented in this paper validate its effectiveness in improving existing processes.
The designed instrument is an optical diffractive element that enhances the performance of the wavefront sensor, unlike other advances that focus on training multiple layers of digital convolutional networks.
“The major advantage it offers, being an optical layer rather than a neural network, is that it achieves good results without requiring significant computational power. Additionally, the optical design occurs physically, allowing preconditioning to happen at the speed of light,” notes Felipe Guzmán, the paper’s lead author.
Also part of the team was the Signal Processing group from UIS in Colombia, led by Jorge Bacca, due to their expertise in designing optical elements related to image sensing problems. Subsequently, collaborative work on physical aspects was done with Benoit Neichel, representing the LAM group due to their extensive experience in designing and implementing wavefront sensors.
Jorge Tapia, a postdoctoral researcher at Optolab, commented on this collaborative work, stating, “The practical difficulties had to do with the fact that the team was dispersed in different parts of the world, yet it worked well, and it is gratifying to achieve this multidisciplinary work.”
The proposed methodology is of great relevance to the scientific community, as it could potentially be applied to the design of a new generation of wavefront sensors for use in next-generation telescopes such as the Extremely Large Telescope (ELT).
You can read the full paper here, and we invite you to check out our Instagram post where we explain the research.