-
Differentiable Optics with dLux II: Optical Design Maximising Fisher Information
Authors:
Louis Desdoigts,
Benjamin Pope,
Michael Gully-Santiago,
Peter Tuthill
Abstract:
The design of astronomical hardware operating at the diffraction limit requires optimization of physical optical simulations of the instrument with respect to desired figures of merit, such as throughput or astrometric accuracy. These systems can be high dimensional, with highly nonlinear relationships between outputs and the adjustable parameters of the hardware. In this series of papers we prese…
▽ More
The design of astronomical hardware operating at the diffraction limit requires optimization of physical optical simulations of the instrument with respect to desired figures of merit, such as throughput or astrometric accuracy. These systems can be high dimensional, with highly nonlinear relationships between outputs and the adjustable parameters of the hardware. In this series of papers we present and apply dLux, an open-source end-to-end differentiable optical modelling framework. Automatic differentiation enables not just efficient high-dimensional optimization of astronomical hardware designs, but also Bayesian experimental design directly targeting the precision of experimental outcomes. Automatic second derivatives enable the exact and numerically stable calculation of parameter covariance forecasts, and higher derivatives of these enable direct optimization of these forecasts. We validate this method against analytic theory and illustrate its utility in evaluating the astrometric precision of a parametrized telescope model, and the design of a diffractive pupil to achieve optimal astrometric performance for exoplanet searches. The source code and tutorial software are open source and publicly available, targeting researchers who may wish to harness dLux for their own optical simulation problems.
△ Less
Submitted 16 June, 2024; v1 submitted 12 June, 2024;
originally announced June 2024.
-
Differentiable Optics with dLux I: Deep calibration of Flat Field and Phase Retrieval with Automatic Differentiation
Authors:
Louis Desdoigts,
Benjamin Pope,
Jordan Dennis,
Peter Tuthill
Abstract:
The sensitivity limits of space telescopes are imposed by uncalibrated errors in the point spread function, photon-noise, background light, and detector sensitivity. These are typically calibrated with specialized wavefront sensor hardware and with flat fields obtained on the ground or with calibration sources, but these leave vulnerabilities to residual time-varying or non-common path aberrations…
▽ More
The sensitivity limits of space telescopes are imposed by uncalibrated errors in the point spread function, photon-noise, background light, and detector sensitivity. These are typically calibrated with specialized wavefront sensor hardware and with flat fields obtained on the ground or with calibration sources, but these leave vulnerabilities to residual time-varying or non-common path aberrations and variations in the detector conditions. It is therefore desirable to infer these from science data alone, facing the prohibitively high dimensional problems of phase retrieval and pixel-level calibration. We introduce a new Python package for physical optics simulation, dLux, which uses the machine learning framework JAX to achieve GPU acceleration and automatic differentiation (autodiff), and apply this to simulating astronomical imaging. In this first of a series of papers, we show that gradient descent enabled by autodiff can be used to simultaneously perform phase retrieval and calibration of detector sensitivity, scaling efficiently to inferring millions of parameters. This new framework enables high dimensional optimization and inference in data analysis and hardware design in astronomy and beyond, which we explore in subsequent papers in this series.
△ Less
Submitted 16 June, 2024; v1 submitted 12 June, 2024;
originally announced June 2024.
-
The James Webb Interferometer: Space-based interferometric detections of PDS 70 b and c at 4.8 $μ$m
Authors:
Dori Blakely,
Doug Johnstone,
Gabriele Cugno,
Anand Sivaramakrishnan,
Peter Tuthill,
Ruobing Dong,
Benjamin J. S. Pope,
Loïc Albert,
Max Charles,
Rachel A. Cooper,
Matthew De Furio,
Louis Desdoigts,
René Doyon,
Logan Francis,
Alexandra Z. Greenbaum,
David Lafrenière,
James P. Lloyd,
Michael R. Meyer,
Laurent Pueyo,
Shrishmoy Ray,
Joel Sánchez-Bermúdez,
Anthony Soulain,
Deepashri Thatte,
Thomas Vandal
Abstract:
We observed the planet-hosting system PDS 70 with the James Webb Interferometer, JWST's Aperture Masking Interferometric (AMI) mode within NIRISS. Observing with the F480M filter centered at 4.8 $μ$m, we simultaneously fit a geometric model to the outer disk and the two known planetary companions. We re-detect the protoplanets PDS 70 b and c at an SNR of 21 and 11, respectively. Our photometry of…
▽ More
We observed the planet-hosting system PDS 70 with the James Webb Interferometer, JWST's Aperture Masking Interferometric (AMI) mode within NIRISS. Observing with the F480M filter centered at 4.8 $μ$m, we simultaneously fit a geometric model to the outer disk and the two known planetary companions. We re-detect the protoplanets PDS 70 b and c at an SNR of 21 and 11, respectively. Our photometry of both PDS 70 b and c provide evidence for circumplanetary disk emission through fitting SED models to these new measurements and those found in the literature. We also newly detect emission within the disk gap at an SNR of $\sim$4, at a position angle of $207^{+11}_{-10}$ degrees, and an unconstrained separation within $\sim$200 mas. Follow-up observations will be needed to determine the nature of this emission. We place a 5$σ$ upper limit of $Δ$mag = 7.56 on the contrast of the candidate PDS 70 d at 4.8 $μ$m, which indicates that if the previously observed emission at shorter wavelengths is due to a planet, this putative planet has a different atmospheric composition than PDS 70 b or c. Finally, we place upper limits on emission from any additional planets in the disk gap. We find an azimuthally averaged 5$σ$ upper limit of $Δ$mag $\approx$ 7.5 at separations greater than 125 mas. These are the deepest limits to date within $\sim$250 mas at 4.8 $μ$m and the first space-based interferometric observations of this system.
△ Less
Submitted 19 April, 2024;
originally announced April 2024.
-
The Near Infrared Imager and Slitless Spectrograph for the James Webb Space Telescope -- IV. Aperture Masking Interferometry
Authors:
Anand Sivaramakrishnan,
Peter Tuthill,
James P. Lloyd,
Alexandra Z. Greenbaum,
Deepashri Thatte,
Rachel A. Cooper,
Thomas Vandal,
Jens Kammerer,
Joel Sanchez-Bermudez,
Benjamin J. S. Pope,
Dori Blakely,
Loïc Albert,
Neil J. Cook,
Doug Johnstone,
André R. Martel,
Kevin Volk,
Anthony Soulain,
Étienne Artigau,
David Lafrenière,
Chris J. Willott,
Sébastien Parmentier,
K. E. Saavik Ford,
Barry McKernan,
M. Begoña Vila,
Neil Rowlands
, et al. (14 additional authors not shown)
Abstract:
The James Webb Space Telescope's Near Infrared Imager and Slitless Spectrograph (JWST-NIRISS) flies a 7-hole non-redundant mask (NRM), the first such interferometer in space, operating at 3-5 \micron~wavelengths, and a bright limit of $\simeq 4$ magnitudes in W2. We describe the NIRISS Aperture Masking Interferometry (AMI) mode to help potential observers understand its underlying principles, pres…
▽ More
The James Webb Space Telescope's Near Infrared Imager and Slitless Spectrograph (JWST-NIRISS) flies a 7-hole non-redundant mask (NRM), the first such interferometer in space, operating at 3-5 \micron~wavelengths, and a bright limit of $\simeq 4$ magnitudes in W2. We describe the NIRISS Aperture Masking Interferometry (AMI) mode to help potential observers understand its underlying principles, present some sample science cases, explain its operational observing strategies, indicate how AMI proposals can be developed with data simulations, and how AMI data can be analyzed. We also present key results from commissioning AMI. Since the allied Kernel Phase Imaging (KPI) technique benefits from AMI operational strategies, we also cover NIRISS KPI methods and analysis techniques, including a new user-friendly KPI pipeline. The NIRISS KPI bright limit is $\simeq 8$ W2 magnitudes. AMI (and KPI) achieve an inner working angle of $\sim 70$ mas that is well inside the $\sim 400$ mas NIRCam inner working angle for its circular occulter coronagraphs at comparable wavelengths.
△ Less
Submitted 7 November, 2022; v1 submitted 31 October, 2022;
originally announced October 2022.
-
Phase Retrieval and Design with Automatic Differentiation
Authors:
Alison Wong,
Benjamin Pope,
Louis Desdoigts,
Peter Tuthill,
Barnaby Norris,
Chris Betters
Abstract:
The principal limitation in many areas of astronomy, especially for directly imaging exoplanets, arises from instability in the point spread function (PSF) delivered by the telescope and instrument. To understand the transfer function, it is often necessary to infer a set of optical aberrations given only the intensity distribution on the sensor - the problem of phase retrieval. This can be import…
▽ More
The principal limitation in many areas of astronomy, especially for directly imaging exoplanets, arises from instability in the point spread function (PSF) delivered by the telescope and instrument. To understand the transfer function, it is often necessary to infer a set of optical aberrations given only the intensity distribution on the sensor - the problem of phase retrieval. This can be important for post-processing of existing data, or for the design of optical phase masks to engineer PSFs optimized to achieve high contrast, angular resolution, or astrometric stability. By exploiting newly efficient and flexible technology for automatic differentiation, which in recent years has undergone rapid development driven by machine learning, we can perform both phase retrieval and design in a way that is systematic, user-friendly, fast, and effective. By using modern gradient descent techniques, this converges efficiently and is easily extended to incorporate constraints and regularization. We illustrate the wide-ranging potential for this approach using our new package, Morphine. Challenging applications performed with this code include precise phase retrieval for both discrete and continuous phase distributions, even where information has been censored such as heavily-saturated sensor data. We also show that the same algorithms can optimize continuous or binary phase masks that are competitive with existing best solutions for two example problems: an Apodizing Phase Plate (APP) coronagraph for exoplanet direct imaging, and a diffractive pupil for narrow-angle astrometry. The Morphine source code and examples are available open-source, with a similar interface to the popular physical optics package Poppy.
△ Less
Submitted 2 July, 2021;
originally announced July 2021.
-
Periodic Astrometric Signal Recovery through Convolutional Autoencoders
Authors:
Michele Delli Veneri,
Louis Desdoigts,
Morgan A. Schmitz,
Alberto Krone-Martins,
Emille E. O. Ishida,
Peter Tuthill,
Rafael S. de Souza,
Richard Scalzo,
Massimo Brescia,
Giuseppe Longo,
Antonio Picariello
Abstract:
Astrometric detection involves a precise measurement of stellar positions, and is widely regarded as the leading concept presently ready to find earth-mass planets in temperate orbits around nearby sun-like stars. The TOLIMAN space telescope[39] is a low-cost, agile mission concept dedicated to narrow-angle astrometric monitoring of bright binary stars. In particular the mission will be optimised…
▽ More
Astrometric detection involves a precise measurement of stellar positions, and is widely regarded as the leading concept presently ready to find earth-mass planets in temperate orbits around nearby sun-like stars. The TOLIMAN space telescope[39] is a low-cost, agile mission concept dedicated to narrow-angle astrometric monitoring of bright binary stars. In particular the mission will be optimised to search for habitable-zone planets around Alpha Centauri AB. If the separation between these two stars can be monitored with sufficient precision, tiny perturbations due to the gravitational tug from an unseen planet can be witnessed and, given the configuration of the optical system, the scale of the shifts in the image plane are about one millionth of a pixel. Image registration at this level of precision has never been demonstrated (to our knowledge) in any setting within science. In this paper we demonstrate that a Deep Convolutional Auto-Encoder is able to retrieve such a signal from simplified simulations of the TOLIMAN data and we present the full experimental pipeline to recreate out experiments from the simulations to the signal analysis. In future works, all the more realistic sources of noise and systematic effects present in the real-world system will be injected into the simulations.
△ Less
Submitted 24 June, 2020;
originally announced June 2020.