Artificial intelligence is set to transform how South Africa’s leading astronomical observatories operate through a new partnership between UK and South African researchers.
The Science and Technology Facilities Council Hartree Centre is collaborating with the South African Astronomical Observatory to deliver the Intelligent Observatory programme. The initiative aims to integrate AI and advanced digital technologies into telescope operations, allowing observatories to automate many routine processes and respond more quickly to events in space.
The programme is designed to make telescopes more autonomous, enabling them to monitor their own systems, process observation data instantly, and capture sudden astronomical events such as exploding stars in real time.
Balancing the stars
Operating an observatory involves managing a range of variables, including unexpected weather changes and potential equipment faults. These disruptions can affect the quality of data collected from distant stars and galaxies.
Until recently, much of the work at the South African Astronomical Observatory relied on small teams manually checking systems and reviewing data. As more astronomers gain access to the facility, researchers say operations must become faster and more automated.
Visiting scientists often need quick access to manuals, research papers, and maintenance records. At the same time, observational data and sensor readings must be corrected for atmospheric distortions and monitored for signs of potential technical issues.
Automating exploration across the Universe
The Intelligent Observatory programme is funded through the UK Research and Innovation International Science Partnership Fund. It brings together specialists in artificial intelligence, software engineering, and telescope operations to address these challenges.
By embedding AI into daily observatory operations, telescopes will be able to oversee their own performance, process observation data rapidly, and produce high quality datasets. Researchers say this will allow scientists to spend more time analysing discoveries rather than managing technical systems.
The collaboration is developing several new tools, including a connected monitoring system that tracks telescope instruments and alerts staff to possible faults before they cause downtime.
Automated data processing tools are also being created to convert raw observations into refined datasets, correcting them for atmospheric interference and imperfections in telescope instruments.
Another feature is an AI powered search platform that allows astronomers to quickly locate relevant logs, technical records, and research papers without manually searching through large collections of documents.
Powering AI for astronomy
The Hartree Centre, based at STFC’s Daresbury Laboratory in the Liverpool City Region, is one of the UK’s leading supercomputing facilities working with industry and the public sector. The centre specialises in artificial intelligence, high performance computing, and advanced data analytics.
The international collaboration with the South African Astronomical Observatory was initiated by Hartree Centre researchers Dr Adriano Agnello and Dr Rob Firth, both former astronomers who now focus on AI research.
Smarter telescopes, faster discoveries
The teams are developing software that can monitor telescope systems, detect problems early, and convert raw observational data into usable scientific insights with minimal human intervention.
The technology can summarise nightly observations, identify technical glitches or weather disruptions, and help ensure that astronomers collect the best possible data.
Advanced AI models are also being used to predict local weather conditions, detect potential equipment faults, and adjust observations to account for atmospheric effects.
Dr Adriano Agnello, principal AI researcher at the Hartree Centre, said the project demonstrates how artificial intelligence can support scientific discovery by making telescope operations more efficient.
He said the new tools represent a significant step forward in applying AI to observatory management and could also support wider research and industrial innovation.
An AI assistant for astronomers
Researchers have also developed a smart text search platform hosted at the Hartree Centre that uses large language models to analyse and interpret technical documents.
The system is designed to reduce incorrect responses and includes simple feedback features to improve reliability. It allows astronomers to find accurate information quickly, particularly during busy nights of observing when time is limited.
The team plans to expand the platform by adding documents from the South African Large Telescope and incorporating additional AI techniques to enhance its capabilities.
A new era of inspiration
Professor Stephen Potter, head of astronomy at the South African Astronomical Observatory, said the collaboration represents a major shift in how astronomical facilities operate.
He said combining the Hartree Centre’s expertise in artificial intelligence and high performance computing with SAAO’s experience in telescope operations is helping embed advanced AI into the core systems that support observation and data analysis.
According to Potter, the partnership strengthens South Africa’s position in innovative astronomy while creating opportunities for students and researchers across Africa to engage with advanced AI driven science.
Broader than astronomy
Researchers say observatories provide ideal testing environments for emerging technologies that could also benefit other industries.
Developments such as smart sensors, predictive maintenance systems, and automated data processing could be applied in sectors including manufacturing, energy, and transport.
The Intelligent Observatory programme demonstrates how artificial intelligence can enhance scientific infrastructure while accelerating discoveries in astronomy.
The project is also helping train the next generation of scientists and engineers. By making information more accessible and integrating AI tools into research environments, the initiative offers students, including those from historically disadvantaged institutions, valuable experience in data science and artificial intelligence.
Source: https://www.ukri.org/news/uk-south-africa-partnership-uses-ai-to-make-telescopes-smarter/
