Pointing Accuracy Improvements for the South Pole Telescope with Machine Learning

In this paper, we present improvements to the pointing accuracy of the South Pole Telescope (SPT) using machine learning. The ability of the SPT to point accurately at the sky is limited by its structural imperfections, which are impacted by the extreme weather at the South Pole. Pointing accuracy is particularly important during SPT participation in observing campaigns with the Event Horizon Telescope (EHT), which requires stricter accuracy than typical observations with the SPT. We compile a training dataset of historical observations of astronomical sources made with the SPT-3G and EHT receivers on the SPT. We train two XGBoost models to learn a mapping from current weather conditions to two telescope drive control arguments — one which corrects for errors in azimuth and the other for errors in elevation. Our trained models achieve root mean squared errors on withheld test data of 2[Formula: see text]14 in cross-elevation and 3[Formula: see text]57 in elevation, well below our goal of 5[Formula: see text] along each axis. We deploy our models on the telescope control system and perform further in situ test observations during the EHT observing campaign in April 2024. Our models result in significantly improved pointing accuracy: for sources within the range of input variables where the models are best trained, average combined pointing error improved 33%, from 15[Formula: see text]9 to 10[Formula: see text]6. These improvements, while significant, fall shy of our ultimate goal, but they serve as a proof of concept for the development of future models. Planned upgrades to the EHT receiver on the SPT will necessitate even stricter pointing accuracy which will be achievable with our methods.

Authors

Institutions

Publication Details

Journal
Journal of Astronomical Instrumentation
Published
2025-05-09
DOI
https://doi.org/10.1142/s2251171725500011
Citations
1
Primary Topic
Adaptive optics and wavefront sensing
Type
article
Field-Weighted Citation Impact
1.25
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Pointing Accuracy Improvements for the South Pole Telescope with Machine Learning

Chunyu Lu, T. Natoli, Krzysztof Zagórski, K. Fichman et al.
1 citations
Journal of Astronomical Instrumentation
Adaptive optics and wavefront sensing
1.25
article

Pointing Accuracy Improvements for the South Pole Telescope with Machine Learning

Chunyu Lu, T. Natoli, Krzysztof Zagórski, K. Fichman, T. de Haan, E. Hivon, N. W. Halverson, Karen Lévy, J. Stephen, P. Chaubal, Abhishek S. Maniyar, A. Coerver, S. Guns, C. L. Reichardt, F. Kéruzoré, J. E. Carlstrom, K. R. Dibert, M. Archipley, A. J. Anderson, E. Schiappucci, F. Guidi, P. Paschos, M. Rahimi, L. Bryant, T.-L. Chou, C. Tandoi, M. Millea, J. A. Sobrin, B. A. Benson, Daniel P. Marrone, M. Korman, E. Camphuis, Y. Wan, L. E. Bleem, Kedar A. Phadke, J. D. Vieira, K. R. Ferguson, G. P. Holder, Z. Pan, F. R. Bouchet, C. Chang, J. E. Ruhl, M. Rouble, A. Doussot, A. Foster, C. Umiltà, P. M. Chichura, L. Balkenhol, K. Prabhu, D. Dutcher, Y. Omori, T. M. Crawford, C. Daley, Cynthia Trendafilova, N. Goeckner-Wald, N. Whitehorn, Behzad Ansarinejad, J. C. Hood, C.-L. Kuo, F. Bianchini, J. Veitch-Michaelis, B. Thorne, Y. Nakato, W. L. Holzapfel, F. Menanteau, G. I. Noble, M. Dobbs, A. N. Bender, A. Hryciuk, R. Gualtieri, A. E. Gambrel, A. Rahlin, Alexander W. Pollak, Arati Chokshi, A. A. Stark, W. L. K. Wu, J. A. Zebrowski, K. Kornoelje, A. Vitrier, Fan Ge, K. Benabed, A. E. Lowitz, R. W. Gardner, S. Padin, S. Galli, M. R. Young, L. Knox, W. Everett, N. Huang, Junhan Kim, J. Montgomery, Wei Quan, C. Feng, A. R. Khalife, E. S. Martsen, M. Doohan
article en
1 citations

Abstract

In this paper, we present improvements to the pointing accuracy of the South Pole Telescope (SPT) using machine learning. The ability of the SPT to point accurately at the sky is limited by its structural imperfections, which are impacted by the extreme weather at the South Pole. Pointing accuracy is particularly important during SPT participation in observing campaigns with the Event Horizon Telescope (EHT), which requires stricter accuracy than typical observations with the SPT. We compile a training dataset of historical observations of astronomical sources made with the SPT-3G and EHT receivers on the SPT. We train two XGBoost models to learn a mapping from current weather conditions to two telescope drive control arguments — one which corrects for errors in azimuth and the other for errors in elevation. Our trained models achieve root mean squared errors on withheld test data of 2[Formula: see text]14 in cross-elevation and 3[Formula: see text]57 in elevation, well below our goal of 5[Formula: see text] along each axis. We deploy our models on the telescope control system and perform further in situ test observations during the EHT observing campaign in April 2024. Our models result in significantly improved pointing accuracy: for sources within the range of input variables where the models are best trained, average combined pointing error improved 33%, from 15[Formula: see text]9 to 10[Formula: see text]6. These improvements, while significant, fall shy of our ultimate goal, but they serve as a proof of concept for the development of future models. Planned upgrades to the EHT receiver on the SPT will necessitate even stricter pointing accuracy which will be achievable with our methods.

Journal of Astronomical InstrumentationVol. 14(01n02)
Canadian Institute for Advanced Research (CA), Northwestern University (US), California Institute of Technology (US), Argonne National Laboratory (US), Centre National de la Recherche Scientifique (FR), Fermi National Accelerator Laboratory (US), University of Arizona (US), High Energy Accelerator Research Organization (JP), Korea Advanced Institute of Science and Technology (KR), University of Illinois Urbana-Champaign (US), University of California, Los Angeles (US), The University of Melbourne (AU), University of Toronto (CA), University of Colorado Boulder (US), Princeton University (US), Université Paris Cité (FR), Commissariat à l'Énergie Atomique et aux Énergies Alternatives (FR), Université Paris-Saclay (FR), Institut d'Astrophysique de Paris (FR), SLAC National Accelerator Laboratory (US), Sorbonne Université (FR), University of Chicago (US), Astrophysique, Instrumentation et Modélisation (FR), Kavli Institute for Particle Astrophysics and Cosmology (US), Canadian Institute for Theoretical Astrophysics (CA), Laboratory for Atmospheric and Space Physics (US), Center for Astrophysics Harvard & Smithsonian (US), CEA Paris-Saclay (FR), National Center for Supercomputing Applications (US), McGill University (CA), Case Western Reserve University (US), University of California, Davis (US), Michigan State University (US), University of California, Berkeley (US), Stanford University (US)
Affordable and clean energy
Openalex Percentile: Top 21%
Adaptive optics and wavefront sensing
1.25
DOISourceOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.