Machine learning-driven approaches for precision antenna alignment

Nimesh A. Patel; Ramprasad Rao; Robert Christensen; Garrett Keating; Kristen Laguana; Adam Mills; Angelu Ramos; Shelbi Schimpf; Rachel Smith; Sheng-Feng Yen; Patrick Leiker; Timothy Norton

doi:10.1117/12.3018888

25 July 2024 Machine learning-driven approaches for precision antenna alignment

Nimesh A. Patel, Ramprasad Rao, Robert Christensen, Garrett Keating, Kristen Laguana, Adam Mills, Angelu Ramos, Shelbi Schimpf, Rachel Smith, Sheng-Feng Yen, Patrick Leiker, Timothy Norton

Author Affiliations +

Proceedings Volume 13101, Software and Cyberinfrastructure for Astronomy VIII; 131010R (2024) https://doi.org/10.1117/12.3018888
Event: SPIE Astronomical Telescopes + Instrumentation, 2024, Yokohama, Japan

Abstract

The Submillimeter Array (SMA) requires precise full-sky blind pointing for its eight 6m antennas, aiming for an error within 3′′, a fraction of the 34′′ FWHM beam at 345 GHz. SMA’s typical 2–3′′ rms pointing accuracy is crucial for efficient array operation, especially with 4 to 6 antenna relocations across 23 pads in various configurations each semester. Traditional calibration using optical guidescopes for mount model errors has shifted to interferometric pointing measurements on quasars, for full model acquisition and baseline calibration. Following every array reconfiguration, mechanical imperfections in antenna mounting lead to significant deviations in azimuth encoder offset and axis tilt parameters, complicating pointing accuracy. To overcome this, a three-layer feed-forward neural network, trained on over ten years of data for each antenna-pad configuration, predicts post-reconfiguration changes. This approach, currently under evaluation and refinement, aims to expedite re-calibration, indicating potential substantial reductions in calibration time and enhanced operational efficiency.

Conference Presentation

Citation Download Citation

Nimesh A. Patel, Ramprasad Rao, Robert Christensen, Garrett Keating, Kristen Laguana, Adam Mills, Angelu Ramos, Shelbi Schimpf, Rachel Smith, Sheng-Feng Yen, Patrick Leiker, and Timothy Norton "Machine learning-driven approaches for precision antenna alignment", Proc. SPIE 13101, Software and Cyberinfrastructure for Astronomy VIII, 131010R (25 July 2024); https://doi.org/10.1117/12.3018888

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

;

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE