Slingshot Aerospace has developed an artificial intelligence (AI) system for DARPA to localize anomalous capabilities within large satellite constellations.
The “Agatha” solution is part of an effort to screen low-earth orbit satellites to increase awareness in space.
The California-based company explained the technology was developed through more than six decades of simulated constellation data and was tested to detect outliers in operational, real-world commercial constellations.
Results from these analyses were then authenticated with respective satellite operators. Several of the covered satellites were also confirmed on their mission, hardware, and operational parameters.
Agatha AI
According to Slingshot, Agatha evaluates high-resolution astrometric, photometric, and contextual information from the firm’s database and global sensor network.
The platform uses additional sources to analyze the location and time of communications between the satellites and ground-based capabilities.
“Agatha represents a breakthrough…as its ability to find these needles in the haystack is something no human, or team of humans, could possibly execute,” Slingshot Aerospace Data Science and AI Director Dr. Dylan Kesler explained.
“Identifying malfunctioning or potentially nefarious objects and their objectives within large satellite constellations is a complex challenge that required 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.”
Future Use in Talks
Slingshot assembled the Agatha system from March 2023 to January 2024.
Currently, the company is engaged with the US government and space firms to discuss the software’s applications in the future.
“The ability to quickly identify anomalies – whether a malfunctioning spacecraft or an intentionally nefarious ‘wolf in sheep’s clothing’ – is an increasingly important aspect of maintaining safety and security in space and on Earth,” Kesler stated.
“Having worked previously in the BioTech world and with gene editing technologies like CRISPR, I know that tools like Agatha and approaches like inverse reinforcement learning almost certainly would have helped us find anomalies in the oceans of genomic data we analyzed.”
“We expect its use to grow exponentially in the years to come to solve a variety of problems, not just in space.”
