Adaptive Optics control using Model-Based Reinforcement Learning

Abstract

Reinforcement Learning (RL) presents a new approach for controlling Adaptive Optics (AO) systems for Astronomy. It promises to effectively cope with some aspects often hampering AO performance such as temporal delay or calibration errors. We formulate the AO control loop as a model-based RL problem (MBRL) and apply it in numerical simulations to a simple Shack-Hartmann Sensor (SHS) based AO system with 24 resolution elements across the aperture. The simulations show that MBRL controlled AO predicts the temporal evolution of turbulence and adjusts to mis-registration between deformable mirror and SHS which is a typical calibration issue in AO. The method learns continuously on timescales of some seconds and is therefore capable of automatically adjusting to changing conditions.

arXiv commentAccepted for publication in Optics Express

References in corpus (17)

• Adam: A Method for Stochastic Optimization
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm
• QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulation
• Limits of Adaptive Optics for high contrast imaging
• Towards Vision-Based Deep Reinforcement Learning for Robotic Motion Control
• The Structure of High Strehl Ratio Point-Spread Functions
• Fundamental limitations on Earth-like planet detection with Extremely Large Telescopes
• Open-loop tomography with artificial neural networks on CANARY: on-sky results
• Origin of the asymmetry of the wind driven halo observed in high contrast images
• Nonlinear wavefront reconstruction with convolutional neural networks for Fourier-based wavefront sensors
• Impact of time-variant turbulence behavior on prediction for adaptive optics systems
• Modelling astronomical adaptive optics performance with temporally-filtered Wiener reconstruction of slope data
• SPHERE+: Imaging young Jupiters down to the snowline
• PCS -- A Roadmap for Exoearth Imaging with the ELT
• Adaptive Optics Predictive Control with Empirical Orthogonal Functions (EOFs)
• A new calibration strategy for adaptive telescopes with pyramid WFS

Cited by in corpus (4)

• Towards on-sky adaptive optics control using reinforcement learning
• Entropy Regularized Reinforcement Learning Using Large Deviation Theory
• Transformer neural networks for closed-loop adaptive optics using non-modulated pyramid wavefront sensors
• Focal plane wavefront control with model-based reinforcement learning