WASHINGTON \u2014 Shield AI has flown its Hivemind autonomy software aboard a satellite for the first time, extending technology developed for autonomous aircraft into space.<\/p>\n
Shield AI, Sedaro and NOVI Space said Aug. 24 that Hivemind ran aboard a NOVI satellite in low Earth orbit during a series of experiments that combined artificial intelligence with onboard simulation software designed to check commands before they were carried out.<\/p>\n
During one 24-hour test, the satellite executed 189 commands generated by Hivemind and approved by Sedaro\u2019s autonomy software, the companies said. The experiment went beyond using artificial intelligence for onboard data processing, testing whether software could make decisions across several mission demands.<\/p>\n
Shield AI is a San Diego-based defense technology company known for its AI autonomy software that allows vehicles to perceive their environment, make decisions and carry out missions with limited or no real-time human control. Shield AI and Sedaro announced a partnership in December 2025 to adapt the technology for orbital missions.<\/p>\n
The latest experiment used a single NOVI spacecraft as the leader of a simulated six-satellite constellation. The other five satellites were virtual rather than actual spacecraft in orbit, an important distinction from demonstrating autonomous control of a physical six-satellite fleet.<\/p>\n
Hivemind and Sedaro\u2019s Autonomy Framework for the Edge, known as SAFE, balanced imaging assignments against factors including spacecraft health, battery charge and pointing requirements across the test architecture, the companies said.<\/p>\n
Giving software greater authority aboard satellites also raises the problem of how to verify that an autonomous command is safe before it is executed.<\/p>\n
Sedaro says its technology is intended to provide that check.<\/p>\n
The Arlington, Virginia-based company\u2019s software uses detailed models of the spacecraft and its environment that can run onboard the satellite. When Hivemind proposes an action, Sedaro\u2019s system can simulate the consequences before the command is carried out, providing a verification layer between the AI software and the spacecraft.<\/p>\n
\u201cThe hardest part of autonomous space operations is trust,\u201d Sedaro\u2019s chief executive and co-founder Robbie Robertson said. Sedaro\u2019s software allows operators to \u201cvalidate autonomous decisions before they execute.\u201d<\/p>\n
The companies did not say that the satellite operated entirely without human oversight. Their announcement said the experiments were conducted with NOVI operator verification and validation, leaving some human involvement in the test process.<\/p>\n
NOVI, also based in Arlington, builds small satellites and onboard computing systems capable of running AI and machine-learning applications in orbit. The spacecraft used for the experiment was not originally designed around Hivemind, according to the companies. The autonomy software was integrated after the satellite was already in orbit.<\/p>\n
NOVI\u2019s LIME spacecraft launched on SpaceX\u2019s Transporter-12 rideshare mission in January 2025 and has since been used for several experiments involving communications, onboard algorithms and AI and machine-learning applications.<\/p>\n
In an earlier 2025 test, Sedaro demonstrated that its spacecraft models could run onboard the NOVI satellite. The latest experiment added Hivemind as the software generating mission commands.<\/p>\n
Autonomy software could be added to existing satellites rather than requiring spacecraft designed around it from the outset, Sedaro said. \u201cThis demonstration showed how trusted autonomy capabilities can be layered onto existing and future constellations.\u201d<\/p>\n
The work builds on a $1.76 million SpaceWERX Phase 2 Small Business Innovation Research award to Sedaro for satellite autonomy technology, with involvement from the Air Force Research Laboratory and Space Systems Command.<\/p>\n
Sean Phillips, technology adviser for AFRL\u2019s Space Control Branch, called the experiment a step toward integrating autonomy into future satellite operations.<\/p>\n
\n\tRelated<\/p>\n","protected":false},"excerpt":{"rendered":"WASHINGTON \u2014 Shield AI has flown its Hivemind autonomy software aboard a satellite for the first time, extending…\n","protected":false},"author":2,"featured_media":149828,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[24,25,73290,73291,4711],"class_list":["post-149827","post","type-post","status-publish","format-standard","has-post-thumbnail","category-ai","tag-ai","tag-artificial-intelligence","tag-sedaro","tag-small-satellite-conference","tag-sn"],"_links":{"self":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/149827","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/comments?post=149827"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/149827\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media\/149828"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media?parent=149827"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/categories?post=149827"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/tags?post=149827"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}