Wavefront Recovery for Multiple Sun Regions in Solar SCAO Scenarios with Deep Learning Techniques
Authors:
- Sergio Luis Suárez GómezSGSergio Luis Suárez Gómez
- Francisco García RiesgoFRFrancisco García Riesgo
- Saúl Pérez FernándezSFSaúl Pérez Fernández
- Francisco Javier Iglesias RodríguezFRFrancisco Javier Iglesias Rodríguez
- Enrique Díez AlonsoEAEnrique Díez Alonso
- Jesús Daniel Santos RodríguezJRJesús Daniel Santos Rodríguez
- Francisco Javier De Cos JuezFJFrancisco Javier De Cos Juez
Open Access
- 23 March 2023
- journal article
- research article
- Published by MDPI AG in Mathematics
- Vol. 11 (7) , 1561
Abstract
The main objective of an Adaptive Optics (AO) system is to correct the aberrations produced in the received wavefronts, caused by atmospheric turbulence. From some measures taken by ground-based telescopes, AO systems must reconstruct all the turbulence traversed by the incoming light and calculate a correction. The turbulence is characterized as a phenomenon that can be modeled as several independent, random, and constantly changing layers. In the case of Solar Single-Conjugated Adaptive Optics (Solar SCAO), the key is to reconstruct the turbulence on-axis with the direction of the observation. Previous research has shown that ANNs are a possible alternative when they have been trained in the Sun’s regions where they must make the reconstructions. Along this research, a new solution based on Artificial Intelligence (AI) is proposed to predict the atmospheric turbulence from the data obtained by the telescope sensors that can generalize recovering wavefronts in regions of the sun completely unknown previously. The presented results show the quality of the reconstructions made by this new technique based on Artificial Neural Networks (ANNs), specifically the Multi-layer Perceptron (MLP).Keywords
Funding Information
- European Union’s Horizon 2020 (824135, SV-PA-21-AYUD/2021/51301, MCIU-22-PID2021-127331NB-I00)
- Fundación para el Fomento en Asturias de la Investigación Científica Aplicada y la Tecnología (824135, SV-PA-21-AYUD/2021/51301, MCIU-22-PID2021-127331NB-I00)
- Ministerio de Ciencia, Innovación y Universidades (824135, SV-PA-21-AYUD/2021/51301, MCIU-22-PID2021-127331NB-I00)
This publication has 18 references indexed in Scilit:
- Deep learningNature, 2015
- Bayesian Learning2015
- The GREGOR adaptive optics systemAstronomische Nachrichten, 2012
- Solar Adaptive OpticsLiving Reviews in Solar Physics, 2011
- European Solar Telescope: Progress statusAstronomische Nachrichten, 2010
- The 1-meter Swedish solar telescope2003
- Solar adaptive optics2000
- Modal wave-front reconstruction with Zernike polynomials and Karhunen–Loève functionsJournal of the Optical Society of America A, 1996
- Principles of Adaptive OpticsPhysics Today, 1992
- Zernike polynomials and atmospheric turbulence*Journal of the Optical Society of America, 1976