Google and NASA Test Robotic Assistant to Aid Astronauts on Deep-Space Missions

Google Cloud and NASA are developing an experimental robotic assistant to assist astronauts with clinical equipment handling and routine spacecraft logistics, according to reporting by TechRadar Pro (https://www.techradar.com/pro/google-is-building-nasa-a-robot-doctor-for-use-on-future-deep-space-missions).

The collaborative project involves Google Cloud, NASA's Human Research Program, the robotics team at Johnson Space Center, and the Jet Propulsion Laboratory. It focuses on addressing the operational barriers of deep-space exploration, where physical distance makes real-time medical consultations with flight surgeons on Earth impossible. On a mission to Mars, for example, radio signals can incur round-trip communication delays of up to 44 minutes, leaving onboard crews reliant on localized support during an emergency.

The experimental hardware builds on the Crew Medical Officer Digital Assistant, an artificial intelligence support system that NASA and Google began evaluating in 2025. Rather than attempting autonomous diagnosis or surgical treatment, the robot is designed to serve as a physical assistant operating under the direct command of a human crew medical officer. In one evaluated scenario, an astronaut suffering acute flank pain could direct the machine to retrieve and position an ultrasound scanner, while a human crew member interprets the imaging and manages clinical care.

To develop the machine's manipulation capabilities, engineers have tested object handling within DeepMind's MuJoCo physics simulation environment and integrated navigation software developed at the Jet Propulsion Laboratory. In initial ground-based laboratory testing, the prototype achieved an 85% success rate locating and moving a medical supply bag.

Because acute medical crises are infrequent, the system is designed primarily as a dual-role payload and logistics handler. During standard flight operations, the robot is slated to manage tool organization, cargo transport, and compartment inspections, freeing crew hours on lean missions.

The initiative remains an early-stage proof of concept. Neither Google nor NASA has scheduled flight demonstrations or manufactured space-qualified hardware, and NASA declined to release further architectural specifications beyond confirming it is evaluating multiple collaborative AI research initiatives for potential future missions.