Predictive control for adaptive optics using neural networks

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

Research output: Contribution to journalArticlepeer-review

22 Scopus citations


Adaptive optics (AO) has become an indispensable tool for ground-based telescopes to mitigate atmospheric seeing and obtain high angular resolution observations. Predictive control aims to overcome latency in AO systems: the inevitable time delay between wavefront measurement and correction. A current method of predictive control uses the empirical orthogonal functions (EOFs) framework borrowed from weather prediction, but the advent of modern machine learning and the rise of neural networks (NNs) offer scope for further improvement. Here, we evaluate the potential application of NNs to predictive control and highlight the advantages that they offer. We first show their superior regularization over the standard truncation regularization used by the linear EOF method with on-sky data before demonstrating the NNs' capacity to model nonlinearities on simulated data. This is highly relevant to the operation of pyramid wavefront sensors (PyWFSs), as the handling of nonlinearities would enable a PyWFS to be used with low modulation and deliver extremely sensitive wavefront measurements.

Original languageEnglish (US)
Article number019001
JournalJournal of Astronomical Telescopes, Instruments, and Systems
Issue number1
StatePublished - Jan 1 2021


  • adaptive optics
  • neural networks
  • wavefront sensors

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Control and Systems Engineering
  • Instrumentation
  • Astronomy and Astrophysics
  • Mechanical Engineering
  • Space and Planetary Science


Dive into the research topics of 'Predictive control for adaptive optics using neural networks'. Together they form a unique fingerprint.

Cite this