NASA tests generative AI for spacecraft autonomy on Mars rover and ISS
NASA's Jet Propulsion Laboratory used Anthropic's Claude models in December to help plan two Mars drives for the Perseverance rover, with human planners reviewing routes before upload. In May, NASA and IBM deployed a compressed AI model on the International Space Station and a satellite to identify floods and clouds from orbit, the first such demonstration in space. In July, astronauts tested a…
Key points
- NASA JPL used Anthropic Claude to plan Perseverance rover drives in December with human oversight
- NASA and IBM demonstrated first compressed AI model for Earth observation on ISS and satellite in May
- Icarus Robotics trains free-flying ISS robots using zero-G data because Earth physics models fail in orbit
Former NASA autonomy chief Robert Ambrose notes engineers have historically avoided non-deterministic systems because outcomes vary based on decision paths. He says testing complexity was managed by automating the testing itself. For distant missions like a potential Europa plume dive, real autonomy becomes essential due to communication delays. Commercial startup Icarus Robotics is developing free-flying robots for the ISS, starting with teleoperation to gather microgravity training data since Earth-trained models fail in zero-G. University of Texas professor Ufuk Topcu argues risk management, not elimination, should guide deployment as commercial space accelerates innovation cycles.
Generative AI Gives Spacecraft the Autonomy Engineers Once Feared
IEEE Spectrum AI · 28 September 2026
Loading the full article…
This text was published by IEEE Spectrum AI and written by Jackie Snow. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Robotics & Physical AI
All →- Point One Navigation’s FUSE 2026 event announces robotics platform built on GPS accuracy · 1 src
- China blocks SoftBank’s $5.375B ABB Robotics acquisition · 3 src
- ABB Robotics and NVIDIA launch physical AI education program at GTC Berlin · 1 src
- Physicl raises $10M seed to build realistic 3D sims for robot training · 2 src
- iQIYI reconstructs lost 1938 sci-fi film with NadouPro AI tools · 1 src
Comments
via GitHub Discussions