MSBAI has introduced OrbitGuard, an innovative AI copilot designed for space traffic management, during an event in Las Vegas at CES 2025. This cutting-edge system transforms the overwhelming influx of space data into actionable insights in nearly real-time, ushering in a new era of awareness in the space domain.
“Space traffic is growing at an unprecedented rate, creating an urgent need for advanced traffic control systems that can keep pace,” said Allan Grosvenor, CEO of MSBAI. “Our team is thrilled about OrbitGuard’s potential – what began as a Space Force initiative has evolved into a groundbreaking solution at a crucial moment. With commercial space activities expanding exponentially, the next evolution of OrbitGuard will be vital for ensuring safe, efficient operations in an increasingly crowded orbital environment.”
OrbitGuard addresses a pressing challenge in Space Domain Awareness (SDA) with innovative solutions. While traditional systems can take months to convert a single uncorrelated track into actionable intelligence, OrbitGuard revolutionizes this process. By harnessing cutting-edge AI, it reduces debris cataloging time to almost real-time and significantly enhances maneuver detection accuracy in geostationary orbit.
The remarkable development of OrbitGuard was made possible through pioneering R&D on the world’s fastest supercomputer, Frontier, at the Oak Ridge Leadership Computing Facility (OLCF). Thanks to a prestigious allocation from the ALCC program, research teams at MSBAI routinely utilized between 100 to 2,000 compute nodes to uncover and refine the advanced AI models behind OrbitGuard. This foundational research, combined with extensive testing and deployment on Department of Defense High Performance Computing Modernization Program (HPCMP) supercomputers, has successfully turned OrbitGuard from an innovative concept into a tangible solution.
The architecture of OrbitGuard is built upon state-of-the-art AI techniques, including our Patch Time Series Transformer (Patch TST) technology. This approach aligns with the latest trends in AI research, shifting from traditional tokenization to more adaptable and efficient patch-based methodologies. The system combines this with:
- Joint Embedding Architectures for advanced temporal understanding
- Multi-framework AutoML for scalable model evolution
- End-to-end Adaptable Workflows powered by modern Reinforcement Learning
MSBAI has successfully fulfilled its obligations under Space Force contract FA864923P1191 by completing the development of the initial OrbitGuard capabilities. This achievement directly aligns with the requirements and collaborative guidance provided by AFRL/RVSX and the DRAGON Army. Additionally, the company has effectively integrated OrbitGuard services at Space Systems Command’s SDA TAP Lab through node-RED integration, conducting live demonstrations on-site in Colorado Springs.
Live demonstrations showcasing OrbitGuard’s capabilities are currently taking place throughout CES 2025, highlighting the promising future of safe and intelligent autonomous maneuvering in space.