Iuliia Shatokhina

Preprocessed cumulative reconstructor with domain decomposition: a fast wavefront reconstruction method for pyramid wavefront sensor

We present a fast method for the wavefront reconstruction from pyramid wavefront sensor (P-WFS) measurements. The method is based on an analytical relation between pyramid and Shack-Hartmann sensor (SH-WFS) data. The algorithm consists of two steps--a transformation of the P-WFS data to SH data, followed by the application of cumulative reconstructor with domain decomposition, a wavefront reconstructor from SH-WFS measurements. The closed loop simulations confirm that our method provides the same quality as the standard matrix vector multiplication method. A complexity analysis as well as speed tests confirm that the method is very fast. Thus, the method can be used on extremely large telescopes, e.g., for eXtreme adaptive optics systems.

Convolution- and Fourier-transform-based reconstructors for pyramid wavefront sensor

In this paper, we present two novel algorithms for wavefront reconstruction from pyramid-type wavefront sensor data. An overview of the current state-of-the-art in the application of pyramid-type wavefront sensors shows that the novel algorithms can be applied in various scientific fields such as astronomy, ophthalmology, and microscopy. Assuming a computationally very challenging setting corresponding to the extreme adaptive optics (XAO) on the European Extremely Large Telescope, we present the results of the performed end-to-end simulations and compare the achieved AO correction quality (in terms of the long-exposure Strehl ratio) to other methods, such as matrix-vector multiplication and preprocessed cumulative reconstructor with domain decomposition. Also, we provide a comparison in terms of applicability and computational complexity and closed-loop performance of our novel algorithms to other methods existing for this type of sensor.

Fast algorithm for wavefront reconstruction in XAO/SCAO with pyramid wavefront sensor

We present a fast wavefront reconstruction algorithm developed for an extreme adaptive optics system equipped with a pyramid wavefront sensor on a 42m telescope. The method is called the Preprocessed Cumulative Reconstructor with domain decomposition (P-CuReD). The algorithm is based on the theoretical relationship between pyramid and Shack-Hartmann wavefront sensor data. The algorithm consists of two consecutive steps - a data preprocessing, and an application of the CuReD algorithm, which is a fast method for wavefront reconstruction from Shack-Hartmann sensor data. The closed loop simulation results show that the P-CuReD method provides the same reconstruction quality and is significantly faster than an MVM.

Wavefront reconstruction for ELT-sized telescopes with pyramid wavefront sensors

The new generation of ground-based telescopes relies on real-time adaptive optics systems to compensate for atmospheric perturbations arising during the imaging process. Pyramid wavefront sensors are planned to be part of many instruments currently under development for ELT-sized telescopes. The high number of correcting elements to be controlled in real-time and the segmented pupils of the ELTs lead to unprecedented challenges posed to the control algorithms. Based on various (approximate) models, several algorithms were developed in the last decades for linear and non-linear wavefront correction from pyramid sensor data. Among those, we emphasize interaction-matrix-based approaches, Fourier domain methods, iterative algorithms, and algorithms based on the inversion of the Finite Hilbert transform. We briefly present the core ideas of the algorithms and provide the necessary theoretical background like, e.g., the Fourier domain filters, or the direct inversion formulas. We give a detailed comparison of the presented methods with respect to underlying pyramid sensor models, the computational complexities, and reconstruction qualities. The performance of our algorithms is demonstrated in the context of an XAO system on the EPICS instrument and a SCAO system on the METIS instrument on the ELT. In the simulations, realistic features as the ELT spiders and the hexagonal M4 geometry are partially taken into account.

Single conjugate adaptive optics for METIS

METIS is the Mid-infrared Extremely large Telescope Imager and Spectrograph, one of the first generation instruments of ESO’s 39m ELT. All scientific observing modes of METIS require adaptive optics (AO) correction close to the diffraction limit. Demanding constraints are introduced by the foreseen coronagraphy modes, which require highest angular resolution and PSF stability. Further design drivers for METIS and its AO system are imposed by the wavelength regime: observations in the thermal infrared require an elaborate thermal, baffling and masking concept. METIS will be equipped with a Single-Conjugate Adaptive Optics (SCAO) system. An integral part of the instrument is the SCAO module. It will host a pyramid type wavefront sensor, operating in the near-IR and located inside the cryogenic environment of the METIS instrument. The wavefront control loop as well as secondary control tasks will be realized within the AO Control System, as part of the instrument. Its main actuators will be the adaptive quaternary mirror and the field stabilization mirror of the ELT. In this paper we report on the phase B design work for the METIS SCAO system; the opto-mechanical design of the SCAO module as well as the control loop concepts and analyses. Simulations were carried out to address a number of important aspects, such as the impact of the fragmented pupil of the ELT on wavefront reconstruction. The trade-off that led to the decision for a pyramid wavefront sensor will be explained, as well as the additional control tasks such as pupil stabilization and compensation of non-common path aberrations.

Dealing with spiders on ELTs: using a Pyramid WFS to overcome residual piston effects

In the design of the future generation ELTs the support structures for the secondary mirror (also known as spiders) lead to a piston on each of the pupil segments created by the spiders, known as ”island effect”. In this talk we focus on fast and stable reconstruction methods to cope with the island effect. We present and compare wavefront reconstruction algorithms and highlight their performance in a METIS- like AO system. We focus on FEWHA (Finite Element-Wavelet Hybrid Algorithm), Poke Matrix Inversion using a set of predefined DM influence functions and new methods for a direct segment piston estimation in combination with the P-CuReD (Preprocessed Cumulative Reconstructor with Domain decomposition). The results are backed up by Octopus (the full AO end-to-end simulator from ESO) simulations highlighting stable Strehl ratios for our simulation setting.