Implementation of closed-loop wavefront control

Closed-loop focal plane wavefront control with the SCExAO instrument

Frantz Martinache1, Nemanja Jovanovic2,3 and Olivier Guyon2,4,5

1 Laboratoire Lagrange, Université Côte d’Azur, Observatoire de la Côte d’Azur, CNRS, Parc Valrose, Bât. H. Fizeau, 06108 Nice, France
2 National Astronomical Observatory of Japan, Subaru Telescope, 650 North A’Ohoku Place, Hilo, HI 96720, USA
3 Department of Physics and Astronomy, Macquarie University, 2109 Sydney, Australia
4 Steward Observatory, University of Arizona, Tucson, AZ 85721, USA
5 College of Optical Sciences, University of Arizona, Tucson, AZ 85721, USA

Received:11 March 2016
Accepted:29 April 2016


Aims. This article describes the implementation of a focal plane based wavefront control loop on the high-contrast imaging instrument SCExAO (Subaru Coronagraphic Extreme Adaptive Optics). The sensor relies on the Fourier analysis of conventional focal-plane images acquired after an asymmetric mask is introduced in the pupil of the instrument.

Methods. This absolute sensor is used here in a closed-loop to compensate for the non-common path errors that normally affects any imaging system relying on an upstream adaptive optics system.This specific implementation was used to control low-order modes corresponding to eight zernike modes (from focus to spherical).

Results. This loop was successfully run on-sky at the Subaru Telescope and is used to offset the SCExAO deformable mirror shape used as a zero-point by the high-order wavefront sensor. The paper details the range of errors this wavefront-sensing approach can operate within and explores the impact of saturation of the data and how it can be bypassed, at a cost in performance.

Conclusions. Beyond this application, because of its low hardware impact, the asymmetric pupil Fourier wavefront sensor (APF-WFS) can easily be ported in a wide variety of wavefront sensing contexts, for ground- as well space-borne telescopes, and for telescope pupils that can be continuous, segmented or even sparse. The technique is powerful because it measures the wavefront where it really matters, at the level of the science detector.

Key words:instrumentation: adaptive optics / methods: data analysis / techniques: high angular resolution / techniques: interferometric

© ESO, 2016


Several approaches to high contrast imaging have now clearly demonstrated the power of focal-plane based image analysis. Most prominently, non-redundant aperture masking (NRM) interferometry (Tuthill et al. 2000), relying on interferometric calibration tricks in the focal plane has led to high contrast detections (of the order of 1000:1) in a regime of angular separation (typically between 0.5 and a few λ/D) that is still unmatched in practice by techniques like coronagraphy (Kraus & Ireland 2012; Sallum et al. 2015). As the generation of extreme adaptive optics (XAO) instruments is coming online, more advanced wavefront control schemes developed in the context of space-borne coronagraphy like speckle nulling (Bordé & Traub 2006) or the general framework of electric field conjugation (Give’On 2009) are being ported on-sky (Martinache et al. 2014; Cady et al. 2013). Nevertheless, it remains remarkable that such a venerable approach (the original masking idea by Fizeau was actually first tested in the 1870s), has remained relevant for well over a century. This is really a tribute to the deep understanding that interferometry has brought to the process of image formation.

More recently, it has been shown that the same self-calibrating tricks used in masking interferometry could in fact be applied to regular (i.e. unmasked) images, assuming AO-correction with residual wavefront errors ≤ 1 radian RMS. The notion of closure-phase (Jennison 1958), was indeed generalized and shown to be a special case of a wider family of self-calibrating observable quantities coined kernel-phases (Martinache 2010), since they form the basis for the null-space (or kernel) of a linear operator. This generalization also opened the way for a focal-plane-based wavefront sensing approach, relying this time on the eigen-phases of the same linear operator. While this problem is generally degenerate, one way to break this degeneracy proved to be simple, and involved masking a small but non-negligible fraction of the pupil to introduce some level of asymmetry. The principles of this asymmetric pupil Fourier wavefront sensor (APF-WFS) were described by Martinache (2013), and exploited by Pope et al. (2014) to show how it could be used, for instance, to cophase a segmented mirror. This paper further expands on the possible applications of this wavefront sensor, since it has now been implemented as part of the SCExAO instrument (Jovanovic et al. 2015b), to compensate for a non-common path error unseen by its upstream pyramid wavefront sensor.

Implementation of closed-loop wavefront control