--- title: Google Cloud and NASA: Testing robotic systems for deep... description: Google Public Sector and NASA are building a proof-of-concept robotic system to assist astronauts on deep space missions, including Mars trips where round-trip... canonical: https://daily.dev/posts/google-cloud-and-nasa-testing-robotic-systems-for-deep-space-missions-rw9m0xdpa twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Google Cloud and NASA: Testing robotic systems for deep space missions | daily.dev og:description: Google Public Sector and NASA are building a proof-of-concept robotic system to assist astronauts on deep space missions, including Mars trips where round-trip... og:url: https://daily.dev/posts/google-cloud-and-nasa-testing-robotic-systems-for-deep-space-missions-rw9m0xdpa og:image: https://api.daily.dev/og/posts/rw9m0xdPa.png og:image:alt: Google Cloud and NASA: Testing robotic systems for deep space missions og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Google Cloud and NASA: Testing robotic systems for deep space missions **[Google Cloud](https://daily.dev/sources/gcp)** · 5 min read · 0 upvotes · 0 comments ## Summary Google Public Sector and NASA are building a proof-of-concept robotic system to assist astronauts on deep space missions, including Mars trips where round-trip communication delays can reach 44 minutes. The robot handles cargo and logistics and can switch to medical support tasks, like preparing an ultrasound machine. It uses the open-source Pi0 vision-language-action model from Physical Intelligence, which hit an 85% baseline success rate in a pick-and-place test at Johnson Space Center's iMETRO facility. Google Cloud G4 instances fine-tuned the 4-billion-parameter Pi0 model in 2.5 hours, and the team tested about 15 edge computing devices for low-power operation in space. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What success rate did the Pi0 robotic model achieve in NASA's pick-and-place tests? The Pi0 vision-language-action model achieved an 85% baseline success rate in early lab tests at Johnson Space Center's iMETRO facility. The test involved a complex pick-and-place maneuver to locate and move a medical bag, using an eye-to-hand camera configuration for spatial awareness. Pi0 is a 4-billion-parameter open-source model developed by Physical Intelligence on Google Cloud infrastructure. _daily.dev surfaces developments like this for engineers tracking real-world robotics and VLA model performance._ ### How long does it take to fine-tune the Pi0 4-billion-parameter model on Google Cloud G4 instances? Fine-tuning the 4-billion-parameter Pi0 model took just 2.5 hours on Google Cloud G4 instances. This training produced models that could then run locally on compact, energy-efficient edge devices, which matters for spacecraft that cannot rely on continuous connections to Earth-based data centers. _Teams evaluating training infrastructure for edge-deployed AI models follow benchmarks like this on daily.dev._ ### Why does round-trip communication latency matter for astronaut medical care on a Mars mission? Round-trip communication latency on a Mars mission can reach up to 44 minutes, making real-time medical guidance from Earth nearly impossible. This is why NASA and Google developed a Clinical Decision Support System and an accompanying robotic system to give astronauts autonomous medical and logistics support without depending on Earth-based communication. _Engineers designing for high-latency, disconnected environments follow edge AI case studies on daily.dev._ ## Similar posts on daily.dev - [Autonomous AI in Space Exploration Raises New Stakes](https://daily.dev/posts/autonomous-ai-in-space-exploration-raises-new-stakes-1urhcxm2o) · IEEE Spectrum · 0 upvotes · 0 comments - [Google Unveils Project Suncatcher, Envisioning AI Models Running in Space](https://daily.dev/posts/google-unveils-project-suncatcher-envisioning-ai-models-running-in-space-lbabw1skb) · InfoQ · 0 upvotes · 0 comments --- Tags: [#gcp](https://daily.dev/tags/gcp), [#robotics](https://daily.dev/tags/robotics), [#edge-computing](https://daily.dev/tags/edge-computing), [#nasa](https://daily.dev/tags/nasa) [View this post on daily.dev](https://daily.dev/posts/google-cloud-and-nasa-testing-robotic-systems-for-deep-space-missions-rw9m0xdpa) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Google Cloud and NASA: Testing robotic systems for deep space missions","url":"https://daily.dev/posts/google-cloud-and-nasa-testing-robotic-systems-for-deep-space-missions-rw9m0xdpa","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/google-cloud-and-nasa-testing-robotic-systems-for-deep-space-missions-rw9m0xdpa"},"datePublished":"2026-10-27T13:21:16.771Z","dateModified":"2026-10-27T13:21:47.153Z","description":"Google Public Sector and NASA are building a proof-of-concept robotic system to assist astronauts on deep space missions, including Mars trips where round-trip...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/fe2a2f5849ad43dd3ea9a4168b8c1060?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/fe2a2f5849ad43dd3ea9a4168b8c1060?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Google Cloud","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Google Cloud","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/gcp","url":"https://daily.dev/sources/gcp"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/google-cloud-and-nasa-testing-robotic-systems-for-deep-space-missions-rw9m0xdpa","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"gcp,robotics,edge-computing,nasa","timeRequired":"PT5M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Google Cloud","item":"https://daily.dev/sources/gcp"},{"@type":"ListItem","position":3,"name":"Google Cloud and NASA: Testing robotic systems for deep space missions"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/google-cloud-and-nasa-testing-robotic-systems-for-deep-space-missions-rw9m0xdpa#faq","mainEntity":[{"@type":"Question","name":"What success rate did the Pi0 robotic model achieve in NASA's pick-and-place tests?","acceptedAnswer":{"@type":"Answer","text":"The Pi0 vision-language-action model achieved an 85% baseline success rate in early lab tests at Johnson Space Center's iMETRO facility. The test involved a complex pick-and-place maneuver to locate and move a medical bag, using an eye-to-hand camera configuration for spatial awareness. Pi0 is a 4-billion-parameter open-source model developed by Physical Intelligence on Google Cloud infrastructure. daily.dev surfaces developments like this for engineers tracking real-world robotics and VLA model performance."}},{"@type":"Question","name":"How long does it take to fine-tune the Pi0 4-billion-parameter model on Google Cloud G4 instances?","acceptedAnswer":{"@type":"Answer","text":"Fine-tuning the 4-billion-parameter Pi0 model took just 2.5 hours on Google Cloud G4 instances. This training produced models that could then run locally on compact, energy-efficient edge devices, which matters for spacecraft that cannot rely on continuous connections to Earth-based data centers. Teams evaluating training infrastructure for edge-deployed AI models follow benchmarks like this on daily.dev."}},{"@type":"Question","name":"Why does round-trip communication latency matter for astronaut medical care on a Mars mission?","acceptedAnswer":{"@type":"Answer","text":"Round-trip communication latency on a Mars mission can reach up to 44 minutes, making real-time medical guidance from Earth nearly impossible. This is why NASA and Google developed a Clinical Decision Support System and an accompanying robotic system to give astronauts autonomous medical and logistics support without depending on Earth-based communication. Engineers designing for high-latency, disconnected environments follow edge AI case studies on daily.dev."}}]} ```