![Representative Image](http://media.assettype.com/cdomagazine%2F2024-05%2Fcab9e83b-1d9a-4828-be7a-806a2da8faaa%2Fpia04234orig.jpg?w=480&auto=format%2Ccompress&fit=max)
The Defense Advanced Research Projects Agency (DARPA) is collaborating with Slingshot Aerospace, Inc., to develop an advanced autonomous artificial intelligence (AI) system designed to detect irregular spacecraft amidst vast satellite constellations.
Slingshot, which is renowned for its AI-driven solutions in satellite tracking, space traffic management, and space simulation, is developing Agatha, which identifies anomalous spacecraft within large satellite constellations.
Agatha underwent training utilizing over six decades' worth of simulated constellation data meticulously crafted by Slingshot Aerospace. Employing state-of-the-art methodologies in AI data analysis, Agatha integrates innovative techniques such as inverse reinforcement learning (IRL).
This approach enables Agatha to assess behaviors, discern policies, and infer intentions of the objects under its surveillance. Unlike conventional methods focused solely on pinpointing individual outlier maneuvers, Agatha, through IRL, delves deeper into strategic inquiries, unraveling the underlying motivations behind satellite behaviors.
“Agatha represents a breakthrough in how AI can deliver unparalleled space domain awareness, as its ability to find these needles in the haystack is something no human, or team of humans, could possibly execute,” said Dr. Dylan Kesler, Director of Data Science and AI, Slingshot Aerospace.
“Identifying malfunctioning or potentially nefarious objects and their objectives within large satellite constellations is a complex challenge that requires us to reach beyond traditional approaches and develop a novel and scalable AI algorithm. Our Agatha model has also proven its ability to deliver high-quality insights that provide ‘explainability’ or context for why specific objects were flagged,” Kesler added.