Belgian spacetech company EDGX has completed its first in-orbit demonstration of STERNA, an AI-enabled edge computing system designed for satellite constellations, launched aboard SpaceX’s Transporter-16 mission. With two hosted payloads now operating in space, EDGX can process data directly in orbit—an essential capability for the next generation of commercial, government, and defence satellite networks.
STERNA brings high-performance, NVIDIA-based computing power onto satellites, allowing data to be analysed in space instead of being sent back to Earth for processing. The platform is built to handle demanding workloads in orbit, with adaptive power usage ranging from 10W to 45W to maintain continuous performance under varying thermal and energy conditions. It is also engineered for durability, with an expected operational lifespan of up to seven years in space.
This achievement marks an important step forward for Europe’s space-based computing ecosystem and follows EDGX’s €2.3 million seed funding round in June 2025.
EDGX CEO Nick Destrycker highlighted the significance of the launch, noting that it represents both a milestone for the company and a step forward for Europe’s role in space computing. He explained that embedding powerful computing capabilities directly in orbit allows satellites to evolve from simple data collectors into systems capable of making real-time decisions. He added that the company’s goal is to provide dependable, scalable computing infrastructure in space, with this mission serving as the foundation.
He also emphasized that the future of the space industry will increasingly rely on in-orbit computing, enabling satellites to function as intelligent, software-driven platforms that can process information at the source.
By delivering NVIDIA-level performance in space, EDGX is paving the way for advanced, software-defined satellites capable of running complex AI applications—from Earth observation analysis to real-time signal intelligence—without needing to transmit large volumes of raw data back to Earth. This approach reduces delays, lowers bandwidth demands, and enables faster decision-making for ground operators.
Ultimately, this capability removes the traditional bottleneck of transmitting vast datasets for ground-based processing, allowing satellite operators to provide quicker, more efficient, and data-driven services. In defence contexts, this can offer a significant operational edge by shortening the time between detection and response in critical situations.