Comparative Study of Neural Network Frameworks for the Next Generation of Adaptive Optics Systems.

Carlos González-Gutiérrez,Alistair Basden,James Osborn,Francisco Díaz-Pernas,Mario Martínez-Zarzuela,Francisco De Cos Juez,Jesús Santos

doi:10.3390/s17061263

Abstract

Many of the next generation of adaptive optics systems on large and extremely large telescopes require tomographic techniques in order to correct for atmospheric turbulence over a large field of view. Multi-object adaptive optics is one such technique. In this paper, different implementations of a tomographic reconstructor based on a machine learning architecture named “CARMEN” are presented. Basic concepts of adaptive optics are introduced first, with a short explanation of three different control systems used on real telescopes and the sensors utilised. The operation of the reconstructor, along with the three neural network frameworks used, and the developed CUDA code are detailed. Changes to the size of the reconstructor influence the training and execution time of the neural network. The native CUDA code turns out to be the best choice for all the systems, although some of the other frameworks offer good performance under certain circumstances.

Highlights

One of the biggest issues for the observation of astronomical targets using ground-based telescopes is the optical aberration produced by the Earth’s atmosphere in the received images
The studied frameworks are not ready to handle that amount of data in a single Graphics Processor Units (GPUs), so we focused only on adaptive optics (AO) systems which can fit in one GPU
Atmospheric conditions change over time periods as small as a few seconds, and having a framework which can re-train the network with a different set of data could help to maintain AO performance

Summary

Introduction

One of the biggest issues for the observation of astronomical targets using ground-based telescopes is the optical aberration produced by the Earth’s atmosphere in the received images. For wide-field AO systems the error introduced by the atmospheric turbulence can be measured in different directions and computer tomography techniques can be used to calculate the aberration in the direction of the object of interest. This allows for the real-time compensation of the astronomical image with the deformable mirrors [3]

Objectives

Discussion

Conclusion