How a Robot Focused 5,000 Radio Eyes express
An orchestra where musicians play in different cities, and the conductor hears them with a fraction-of-a-second delay. That's roughly how radio telescope networks like the Very Long Baseline Array work — a dozen antennas from Hawaii to the Virgin Islands catch the whispers of distant galaxies. To assemble a single image from disparate recordings, you need to synchronize the signals perfectly — accounting for air humidity, atmospheric jitter, and off-kilter clocks. For decades, humans did this calibration, spending hours on each object. For the SMILE project, which requires processing about 5000 sources, this approach became a bottleneck.
AVICA, the new tool, takes on the role of a patient sound engineer. It opens archival files recorded in different formats over 30 years and finds the reference tone on its own — a bright spot in the sky against which all antennas can be aligned. Like a piano tuner, the algorithm catches the beating of signals (interference fringes) and picks the cleanest note. Then the program automatically applies a chain of corrections, turning raw data into a finished image. The whole process for one object takes about 30 minutes on a standard server.
Behind the scenes, the ALFRD system conducts the work: it launches tasks, monitors progress, and logs results in real time. This approach has for the first time allowed a giant archive of very long baseline interferometry data to be fully automatically processed without any manual calibration. In the future, AVICA's ability to choose reference signals on its own will be included in the public rPICARD package, and any radio telescope will be able to work blind, without a human at the controls. The orchestra has finally learned to play in harmony by itself.
Original abstract
as written by the authors arXiv:2604.17448
Calibrating large volumes of Very Long Baseline Interferometry (VLBI) data traditionally requires significant human intervention at every stage. While the Common Astronomy Software Applications (CASA) package is the standard data reduction tool across major radio observatories, no existing CASA-based pipeline operates in a fully automated manner across the heterogeneous data formats produced by the Very Long Baseline Array (VLBA) over three decades of operations. The Search for Milli-Lenses (SMILE) project, requiring the calibration of ~5000 VLBA sources, makes such blind automation a practical necessity. We introduce the Automated VLBI pipeline in CASA (AVICA), which automates the calibration of archival VLBA data. AVICA extends the CASA-based rPICARD framework by automating preprocessing of FITS-IDI and Measurement Set data formats, calibrator and reference antenna selection via FFT-based fringe detection, and execution of the full calibration workflow. Progress tracking is handled by ALFRD (Automated Logical Framework for executing Dynamic scripts), which orchestrates pipeline execution and records results in real time. AVICA was validated on 1000 NRAO archival sources spanning 1995-2023, covering 1372 band-separated observations across the S, C, X, U, and K bands. Calibrated output was produced for 978 sources (97.8%), with 22 failures due to corrupted or incomplete data. Mean per-source execution time was ~30 minutes using MPI parallelization with up to 20 cores. AVICA demonstrates that fully blind calibration of heterogeneous archival VLBA data is achievable with CASA. The automated calibrator and reference antenna selection will be incorporated into a future rPICARD release, extending blind calibration to any supported array. AVICA and ALFRD are available as open-source Python packages.
The VLBA network is so sensitive that it can detect a continent shifting by a couple of centimeters — scientists use it to measure the motion of tectonic plates.
In Carl Sagan's novel "Contact," the first signal from an extraterrestrial civilization is caught precisely by a network of radio telescopes similar to the VLBA.