Forecasting Wavefront Corrections in an Adaptive Optics System. (arXiv:2112.01437v1 [astro-ph.IM])
<a href="http://arxiv.org/find/astro-ph/1/au:+Hafeez_R/0/1/0/all/0/1">Rehan Hafeez</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Archinuk_F/0/1/0/all/0/1">Finn Archinuk</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Fabbro_S/0/1/0/all/0/1">Sébastien Fabbro</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Teimoorinia_H/0/1/0/all/0/1">Hossen Teimoorinia</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Veran_J/0/1/0/all/0/1">Jean-Pierre Véran</a>
We use telemetry data from the Gemini North ALTAIR adaptive optics system to
investigate how well the commands to the wavefront correctors (Tip/Tilt and
high-order turbulence) can be forecasted in order to reduce lag error and
improve delivered image quality. We show that a high level of reduction ($sim$
5 for Tip-Tilt and $sim$ 2 for high-order modes) can be achieved by using a
“forecasting module” based on a linear auto-regressive models with only a few
coefficients ($sim$ 30 for Tip-Tilt and $sim$ 5 for high-order modes) to
complement the existing integral servo-controller. Updating this module to
adapt to evolving observing conditions is computationally inexpensive and
requires less than 10 seconds worth of telemetry data. We also use several
machine learning models to evaluate whether further improvements could be
achieved with a more sophisticated non-linear model. Our attempts showed no
perceptible improvements over linear auto-regressive predictions, even for
large lags where residuals from the linear models are high, suggesting that
non-linear wavefront distortions for ALTAIR at the Gemini North telescope may
not be forecasted with the current setup.
We use telemetry data from the Gemini North ALTAIR adaptive optics system to
investigate how well the commands to the wavefront correctors (Tip/Tilt and
high-order turbulence) can be forecasted in order to reduce lag error and
improve delivered image quality. We show that a high level of reduction ($sim$
5 for Tip-Tilt and $sim$ 2 for high-order modes) can be achieved by using a
“forecasting module” based on a linear auto-regressive models with only a few
coefficients ($sim$ 30 for Tip-Tilt and $sim$ 5 for high-order modes) to
complement the existing integral servo-controller. Updating this module to
adapt to evolving observing conditions is computationally inexpensive and
requires less than 10 seconds worth of telemetry data. We also use several
machine learning models to evaluate whether further improvements could be
achieved with a more sophisticated non-linear model. Our attempts showed no
perceptible improvements over linear auto-regressive predictions, even for
large lags where residuals from the linear models are high, suggesting that
non-linear wavefront distortions for ALTAIR at the Gemini North telescope may
not be forecasted with the current setup.
http://arxiv.org/icons/sfx.gif
