Artificial intelligence is no longer just powering chatbots. The European Space Agency (ESA) said it is now using AI and machine learning to improve rocket manufacturing.
The European Space Agency (ESA) said it is using artificial intelligence and machine learning to develop launch vehicles that are faster, more precise, and more cost-efficient in the future. ESA’s Future Launchers Preparatory Programme (FLPP) carries out the initiative in partnership with German aerospace company MT Aerospace.
“Artificial Intelligence, or AI, promises many benefits in all domains, and rocketry is no different,” ESA said in a statement, noting that the technology could lead to better manufacturing processes and even entirely new material shapes for future rockets and spacecraft.
One of the key techniques being improved by AI is shot peen forming, a process that shapes metal through repeated impacts from fast-firing metal balls rather than heat. The method is used to create dome-shaped fuel tank heads for the Ariane 6 rocket.
Predicting how the metal will bend during the process has been traditionally difficult. But machine learning models are now being used to forecast metal deformation more accurately, improving precision and reducing waste.
ESA is also applying AI to friction welding, a technique that joins metals using a rapidly spinning pin rather than conventional arc welding. Machine learning helps speed up machine setup and analyze weld quality, allowing manufacturers to detect defects earlier in the production process.
Meanwhile, under the Phoebus Project, ESA used AI to support the development of carbon-fiber fuel tanks for Ariane 6. It combines machine learning with laser sensor technology to enable engineers detect defects in carbon-fiber layering in real time. This allows production to be halted immediately when problems arise, reducing errors and shortening manufacturing timelines.
“Artificial intelligence, such as machine learning, in combination with new digital technologies, is transforming launcher manufacturing,” said Daniel Chipping, ESA project manager for software-centred activities under the FLPP. “From automating complex analysis tasks to reducing tedious machine stop-starts, we are starting to see the benefits across all materials and shaping processes.”
ESA said these advances could help make Europe’s future launch systems more efficient, reliable, and cost-effective as global competition in space transportation intensifies.