Google, NASA partner to build robot medic for deep-space astronaut assistance

Google's public sector division is collaborating with NASA's Human Research Program, the robotics team at Johnson Space Center, and the Jet Propulsion Laboratory to create a robotic assistant capable of aiding astronauts during forthcoming deep-space voyages. Jim Kelly, the public sector VP at Chocolate Factory, detailed in a blog post posted Tuesday that this ambitious undertaking could extend their technology far beyond the lunar surface. This robotic system is intended to support astronauts with tasks ranging from medical duties and handling supplies to performing routine upkeep.

This robotics endeavor builds upon the Crew Medical Officer Digital Assistant (CMO-DA), an AI-driven system for supporting medical decisions that Google has been developing alongside NASA since last year. CMO-DA, which utilizes the open-source local AI tool RamaLama supported by Red Hat, is currently undergoing evaluations on terrestrial computers that mirror those used aboard the ISS. Its purpose is to process medical concerns using AI models locally, a capability that will be crucial for future missions in space due to the communication delays experienced between Earth and astronauts far from home.

Kelly noted in this week's publication that "An astronaut on a future deep space mission to Mars could face severe operational challenges—including round-trip communication latencies of up to 44 minutes, making real-time medical guidance for crew nearly impossible." The vision shared by Google and NASA is for this robot to function as a supplement to the human crew medical officer, assisting with medical tasks and undertaking other work when direct bedside care is not necessary. Kelly elaborated that, "To maximize its utility for a lean crew, the robotic system is designed as a versatile 'payload' handler capable of managing cargo, logistics, and equipment." Furthermore, he clarified that "When not needed for routine maintenance like organizing tools or securing compartments, it can pivot to medical support."

As a specific illustration of its potential applications, the Google public sector VP provided an example: should someone suffer intense back pain, possibly stemming from kidney stones, the robot could retrieve and prepare an ultrasound machine, thereby allowing the crew medical officer to focus on interpreting the scan and determining the appropriate treatment. Kelly asserted that "The ultimate goal is to build a robotic system that can execute a variety of tasks via simple voice instructions, under the astronauts’ direct supervision and command." He added, "For instance, the robot might be directed to assist with routine health monitoring—such as capturing basic vitals or simple ultrasounds—under the strict guidance of a human crew medical officer."

Regarding the project's current standing, it has not progressed extensively yet. Google is investigating several technologies, including JPL's efforts to adapt satellite situational-awareness tools, which were initially used for tracking orbiting objects, to enable the robot to navigate three-dimensional environments. While early laboratory demonstrations have taken place, placing the project at the proof-of-concept stage, Google has not confirmed the existence of a fully integrated, space-ready robot. Kelly mentioned that Google utilized DeepMind's MuJoCo physics simulator to test the robot's movements, and in initial lab assessments involving locating and moving a medical bag, the system achieved an 85 percent success rate—a starting point, though not definitive proof of readiness for medical procedures. When approached for further details, Google directed inquiries toward NASA, but the space agency declined to offer precise information. The agency merely stated that it is developing various "super intelligence" projects for its numerous missions, and that it is collaborating with partners to evaluate these for future uses.

Why it matters

This development shows how AI and robotics are being applied to solve critical operational challenges in extreme environments. For developers, this highlights the practical application of local AI processing to overcome communication latency in future space endeavors.