Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userNonlinear wavefront reconstruction with convolutional neural networks for Fourier-based wavefront sensors
66
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
Fourier-based wavefront sensors, such as the Pyramid Wavefront Sensor (PWFS), are the current preference for high contrast imaging due to their high sensitivity. However, these wavefront sensors have intrinsic nonlinearities that constrain the range where conventional linear reconstruction methods can be used to accurately estimate the incoming wavefront aberrations. We propose to use Convolutional Neural Networks (CNNs) for the nonlinear reconstruction of the wavefront sensor measurements. It is demonstrated that a CNN can be used to accurately reconstruct the nonlinearities in both simulations and a lab implementation. We show that solely using a CNN for the reconstruction leads to suboptimal closed loop performance under simulated atmospheric turbulence. However, it is demonstrated that using a CNN to estimate the nonlinear error term on top of a linear model results in an improved effective dynamic range of a simulated adaptive optics system. The larger effective dynamic range results in a higher Strehl ratio under conditions where the nonlinear error is relevant. This will allow the current and future generation of large astronomical telescopes to work in a wider range of atmospheric conditions and therefore reduce costly downtime of such facilities.
rate research
Read More
Wavefront sensors encode phase information of an incoming wavefront into an intensity pattern that can be measured on a camera. Several kinds of wavefront sensors (WFS) are used in astronomical adaptive optics. Amongst them, Fourier-based wavefront sensors perform a filtering operation on the wavefront in the focal plane. The most well known example of a WFS of this kind is the Zernike wavefront sensor, and the pyramid wavefront sensor (PWFS) also belongs to this class. Based on this same principle, new WFSs can be proposed such as the n-faced pyramid (which ultimately becomes an axicone) or the flattened pyramid, depending on whether the image formation is incoherent or coherent. In order to test such novel concepts, the LOOPS adaptive optics testbed hosted at the Laboratoire dAstrophysique de Marseille has been upgraded by adding a Spatial Light Modulator (SLM). This device, placed in a focal plane produces high-definition phase masks that mimic otherwise bulk optic devices. In this paper, we first present the optical design and upgrades made to the experimental setup of the LOOPS bench. Then, we focus on the generation of the phase masks with the SLM and the implications of having such a device in a focal plane. Finally, we present the first closed-loop results in either static or dynamic mode with different WFS applied on the SLM.
Increasing interest in astronomical applications of non-linear curvature wavefront sensors for turbulence detection and correction makes it important to understand how best to handle the data they produce, particularly at low light levels. Algorithms for wavefront phase-retrieval from a four-plane curvature wavefront sensor are developed and compared, with a view to their use for low order phase compensation in instruments combining adaptive optics and Lucky Imaging. The convergence speed and quality of iterative algorithms is compared to their step-size and techniques for phase retrieval at low photon counts are explored. Computer simulations show that at low light levels, preprocessing by convolution of the measured signal with a gaussian function can reduce by an order of magnitude the photon flux required for accurate phase retrieval of low-order errors. This facilitates wavefront correction on large telescopes with very faint reference stars.
The Adaptive Optics Lucky Imager (AOLI) is a new instrument under development to demonstrate near diffraction limited imaging in the visible on large ground-based telescopes. We present the adaptive optics system being designed for the instrument comprising a large stroke deformable mirror, fixed component non-linear curvature wavefront sensor and photon-counting EMCCD detectors. We describe the optical design of the wavefront sensor where two photoncounting CCDs provide a total of four reference images. Simulations of the optical characteristics of the system are discussed, with their relevance to low and high order AO systems. The development and optimisation of high-speed wavefront reconstruction algorithms are presented. Finally we discuss the results of simulations to demonstrate the sensitivity of the system.
Focal plane wavefront sensing (FPWFS) is appealing for several reasons. Notably, it offers high sensitivity and does not suffer from non-common path aberrations (NCPA). The price to pay is a high computational burden and the need for diversity to lift any phase ambiguity. If those limitations can be overcome, FPWFS is a great solution for NCPA measurement, a key limitation for high-contrast imaging, and could be used as adaptive optics wavefront sensor. Here, we propose to use deep convolutional neural networks (CNNs) to measure NCPA based on focal plane images. Two CNN architectures are considered: ResNet-50 and U-Net which are used respectively to estimate Zernike coefficients or directly the phase. The models are trained on labelled datasets and evaluated at various flux levels and for two spatial frequency contents (20 and 100 Zernike modes). In these idealized simulations we demonstrate that the CNN-based models reach the photon noise limit in a large range of conditions. We show, for example, that the root mean squared (rms) wavefront error (WFE) can be reduced to < $lambda$/1500 for $2 times 10^6$ photons in one iteration when estimating 20 Zernike modes. We also show that CNN-based models are sufficiently robust to varying signal-to-noise ratio, under the presence of higher-order aberrations, and under different amplitudes of aberrations. Additionally, they display similar to superior performance compared to iterative phase retrieval algorithms. CNNs therefore represent a compelling way to implement FPWFS, which can leverage the high sensitivity of FPWFS over a broad range of conditions.
We investigate the improvements in Shack-Hartmann wavefront sensor image processing that can be realised using total variation minimisation techniques to remove noise from these images. We perform Monte-Carlo simulation to demonstrate that at certain signal-to-noise levels, sensitivity improvements of up to one astronomical magnitude can be realised. We also present on-sky measurements taken with the CANARY adaptive optics system that demonstrate an improvement in performance when this technique is employed, and show that this algorithm can be implemented in a real-time control system. We conclude that total variation minimisation can lead to improvements in sensitivity of up to one astronomical magnitude when used with adaptive optics systems.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
