Predictive control for adaptive optics using neural networks

Wong, Alison; Norris, Barnaby; Tuthill, P. G.; Scalzo, R.; Lozi, Julien; Vievard, Sébastien; Guyon, Olivier

doi:10.1117/1.jatis.7.1.019001

Cited by 26 publications

(17 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Predictive controllers have been studied in Guyon & Males (2017); Poyneer et al (2007); Dessenne et al (1998);van Kooten et al (2017van Kooten et al ( , 2019. Methods vary from linear filters to filters operating on single modes (such as Fourier modes) to neural network approaches (Swanson et al 2018;Sun et al 2017;Liu et al 2019;Wong et al 2021). Predictive control methods have also been studied in a closed-loop configuration.…”

Section: Related Workmentioning

confidence: 99%

Toward on-sky adaptive optics control using reinforcement learning

Nousiainen

Rajani

Kasper

et al. 2022

A&A

View full text Add to dashboard Cite

Context. The direct imaging of potentially habitable exoplanets is one prime science case for the next generation of high contrast imaging instruments on ground-based, extremely large telescopes. To reach this demanding science goal, the instruments are equipped with eXtreme Adaptive Optics (XAO) systems which will control thousands of actuators at a framerate of kilohertz to several kilohertz. Most of the habitable exoplanets are located at small angular separations from their host stars, where the current control laws of XAO systems leave strong residuals. Aims. Current AO control strategies such as static matrix-based wavefront reconstruction and integrator control suffer from a temporal delay error and are sensitive to mis-registration, that is, to dynamic variations of the control system geometry. We aim to produce control methods that cope with these limitations, provide a significantly improved AO correction, and, therefore, reduce the residual flux in the coronagraphic point spread function (PSF). Methods. We extend previous work in reinforcement learning for AO. The improved method, called the Policy Optimization for Adaptive Optics (PO4AO), learns a dynamics model and optimizes a control neural network, called a policy. We introduce the method and study it through numerical simulations of XAO with Pyramid wavefront sensor (PWFS) for the 8-m and 40-m telescope aperture cases. We further implemented PO4AO and carried out experiments in a laboratory environment using Magellan Adaptive Optics eXtreme system (MagAO-X) at the Steward laboratory.Results. PO4AO provides the desired performance by improving the coronagraphic contrast in numerical simulations by factors of 3-5 within the control region of deformable mirror and PWFS, both in simulation and in the laboratory. The presented method is also quick to train, that is, on timescales of typically 5-10 seconds, and the inference time is sufficiently small (< ms) to be used in real-time control for XAO with currently available hardware even for extremely large telescopes.

show abstract

Section: Related Workmentioning

confidence: 99%

Toward on-sky adaptive optics control using reinforcement learning

Nousiainen

Rajani

Kasper

et al. 2022

A&A

View full text Add to dashboard Cite

show abstract

Section: Related Workmentioning

confidence: 99%

Towards on-sky adaptive optics control using reinforcement learning

Nousiainen,

Rajani,

Kasper

et al. 2022

Preprint

View full text Add to dashboard Cite

Context. The direct imaging of potentially habitable Exoplanets is one prime science case for the next generation of high contrast imaging instruments on ground-based extremely large telescopes. To reach this demanding science goal, the instruments are equipped with eXtreme Adaptive Optics (XAO) systems which will control thousands of actuators at a framerate of kilohertz to several kilohertz. Most of the habitable exoplanets are located at small angular separations from their host stars, where the current control laws of XAO systems leave strong residuals. Aims. Current AO control strategies like static matrix-based wavefront reconstruction and integrator control suffer from temporal delay error and are sensitive to mis-registration, i.e., to dynamic variations of the control system geometry. We aim to produce control methods that cope with these limitations, provide a significantly improved AO correction, and, therefore, reduce the residual flux in the coronagraphic point spread function (PSF). Methods. We extend previous work in Reinforcement Learning (RL) for AO. The improved method, called PO4AO, learns a dynamics model and optimizes a control neural network, called a policy. We introduce the method and study it through numerical simulations of XAO with Pyramid wavefront sensing for the 8-m and 40-m telescope aperture cases. We further implemented PO4AO and carried out experiments in a laboratory environment using Magellan Adaptive Optics eXtreme system (MagAO-X) at the Steward laboratory.Results. PO4AO provides the desired performance by improving the coronagraphic contrast in numerical simulations by factors 3-5 within the control region of DM and Pyramid WFS, both in simulation and in the laboratory. The presented method is also quick to train, i.e., on timescales of typically 5-10 seconds, and the inference time is sufficiently small (< ms) to be used in real-time control for XAO with currently available hardware even for extremely large telescopes.

show abstract

“…22 More recent simulation work show promising results when comparing EOF to a neural network under certain assumptions and advocate for operating non-linear wavefront sensors. 23 However, more work into necessary to understand the require training for on-sky deployment and the required hardware for real-time-control. These are also discussed by J. Nousianen et al (2021) 24 and R. Swanson et al (2021) 25 while showing results for other machine learning algorithms.…”

Section: Introductionmentioning

confidence: 99%

Predictive wavefront control on Keck II adaptive optics bench: on-sky coronagraphic results

Kooten¹,

Cetre²,

Ragland³

et al. 2022

Preprint

View full text Add to dashboard Cite

The behavior of an adaptive optics (AO) system for ground-based high contrast imaging (HCI) dictates the achievable contrast of the instrument. In conditions where the coherence time of the atmosphere is short compared to the speed of the AO system, the servo-lag error can become the dominant error term of the AO system. While the AO system measures the wavefront error and subsequently applies a correction (typically taking a total of one or a few milliseconds), the atmospheric turbulence above the telescope has changed resulting in the servolag error. In addition to reducing the Strehl ratio, the servo-lag error causes a build-up of speckles along the direction of the dominant wind vector in the coronagraphic image, severely limiting the contrast at small angular separations. One strategy to mitigate this problem is to predict the evolution of the turbulence over the delay time. Our predictive wavefront control algorithm minimizes, in a mean square sense, the wavefront error over the delay and has been implemented on the Keck II AO bench. In this paper, we report on the latest results of our algorithm and discuss updates to the algorithm itself. We explore how to tune various filter parameters based on both daytime laboratory tests and on-sky tests. We show a reduction in residual-mean-square wavefront error for the predictor compared to the leaky integrator (the standard controller for Keck) implemented on Keck for three separate nights. Finally, we present contrast improvements for daytime and on-sky tests for the first time.Using the L-band vortex coronagraph for Keck's NIRC2 instrument, we find a contrast gain of up to 2 at a separation of 3 λ/D and up to 3 for larger separations (3-7λ/D).

show abstract

Predictive control for adaptive optics using neural networks

Cited by 26 publications

References 32 publications

Toward on-sky adaptive optics control using reinforcement learning

Toward on-sky adaptive optics control using reinforcement learning

Towards on-sky adaptive optics control using reinforcement learning

Predictive wavefront control on Keck II adaptive optics bench: on-sky coronagraphic results

Contact Info

Product

Resources

About