iscover
Shield AI and Sedaro have successfully demonstrated trusted autonomous satellite operations in low-Earth orbit, marking the first time Shield AI’s Hivemind autonomy software has operated in space.
Shield AI, Sedaro and NOVI Space have demonstrated autonomous satellite operations in orbit, extending Shield AI’s Hivemind artificial intelligence technology from airborne platforms into the space domain.
During a series of on-orbit experiments, Hivemind operated aboard a NOVI satellite in low-Earth orbit, working with Sedaro’s Autonomy Framework for the Edge (SAFE) to manage multiple mission objectives.
The demonstration involved the NOVI spacecraft acting as the leader of a six-satellite constellation, comprising the physical satellite and five virtual spacecraft. Hivemind and SAFE jointly optimised imaging tasking, spacecraft health, battery state of charge and satellite pointing.
Todd Wesley, vice president of Hivemind engineering at Shield AI, said: “By integrating Hivemind with Sedaro’s EDS’ and SAFE framework, we demonstrated how operators can move from manually operating distributed assets to autonomous mission tasking in a seamless workflow.”
Over a 24-hour experiment, the system generated and executed 189 commands that were approved through Sedaro’s SAFE framework.
A second experiment used Sedaro’s Edge Deployable Simulators to perform similar optimisation tasks and increased the satellite’s initiation of Iridium communications contacts by 8 percentage points.
Robbie Robertson, CEO and co-founder of Sedaro, said: “By providing a predictive, onboard model of the spacecraft and its environment, SAFE and our edge-deployable simulators let operators validate autonomous decisions before they execute.”
The demonstration is significant because satellite constellations supporting communications, Earth observation and national security missions remain heavily dependent on human operators and ground-based control.
Shield AI and Sedaro said trusted autonomy could reduce operator workload while allowing increasingly complex constellations to respond more rapidly to changing mission requirements.
“By integrating Hivemind with Sedaro’s EDS’ and SAFE framework, we demonstrated how operators can move from manually operating distributed assets to autonomous mission tasking in a seamless workflow,” Wesley said.
Robertson said establishing trust would be critical to deploying autonomous systems in space, with SAFE providing predictive models that allow operators to validate autonomous decisions before they are executed.
The demonstration also showed that autonomy could potentially be added to satellites that were not originally designed around autonomous operations, including for intelligence, surveillance and reconnaissance missions.
For Shield AI, the test represents a major expansion of Hivemind beyond aircraft and into orbital operations. The company and Sedaro established a strategic partnership in 2025 to develop and demonstrate autonomous space operations, with Hivemind selected as Sedaro’s preferred autonomy software for on-orbit demonstrations.
Sedaro’s SAFE technology was developed with support from SpaceWERX, the US Space Force’s innovation arm, in partnership with the Air Force Research Laboratory and Space Systems Command.
Dr Sean Phillips, technology adviser, Space Control Branch, Air Force Research Laboratory, said: “This effort provides a significant milestone to the integration of autonomy and satellite operations to meet future mission needs.”
As satellite constellations become larger and more complex, the demonstration highlights the potential for AI-enabled autonomy to shift space operations from continuous human control towards supervised, distributed and increasingly self-directed mission management.
Want to see more stories from trusted news sources?
Make Defence Connect a preferred news source on Google.
Click here to add Defence Connect as a preferred news source.