Computing at the edge of space to accelerate EO insights
Able to capture high-definition scans of air, land and sea across a range of visible and electromagnetic spectra, Earth observation (EO) satellites collect a lot of data. While this has the potential to help monitor wildfires, track weather, manage agriculture, and much more, the desired insights lie buried in huge amounts of raw data. “Before EO data can be processed, it needs to be sent from the satellite in space back down to the ground, which can take hours,” explains Adèle Hankache, microservices and cloud-native architect specialist at Thales Alenia Space(opens in new window) in France. This transmission latency creates bottlenecks that drive up competition for download bandwidth and limit our ability to make use of satellite data. This is where the ORCHIDE(opens in new window) project comes in. Leveraging an approach known as space edge computing, the project – a consortium of four partners across France, Poland and Romania – set out to relieve the EO bottleneck by processing data on board satellites as it is acquired. “By processing data directly in space, we can reduce the amount of data being transmitted to Earth and improve response times for time-critical applications while also strengthening Europe’s technological autonomy and competitiveness in the space sector,” says Hankache, the project’s coordinator.
Lightweight, secure, unikernel design
At the heart of the project is an artificial intelligence-enabled on-board orchestration platform for the autonomous, secure and energy-efficient in-orbit processing of data. The system is unikernel-based, supporting the lightweight operating systems that are increasingly common as satellites shift to multipurpose, software-defined devices. “Because our platform is software- and hardware-agnostic, it operates independently of underlying technologies and is thus compatible across heterogeneous edge and space platforms,” notes Hankache. On the ground, developers use the project’s software development kit (SDK), a suite of diagnostic and application tools, to develop, package and prepare their applications for execution on board satellites. These tools can also be shared through an app marketplace. The project developed a simulation environment that allows the on-board system and end-user applications to be tested and validated before deployment, reducing development risks and further facilitating the adoption of space edge computing technologies.
More efficient and responsive EO services
From maritime surveillance to cloud and fire detection, once deployed, the orchestration platform can support any use case packaged and prepared with its SDK. Safe-and-secure-by-design, the ORCHIDE platform dynamically manages services, on-board resources and task priorities. Raw image data is processed while still on the satellite, and only meaningful results sent to ground, drastically improving the speed at which insights can be delivered. “Our platform makes satellite systems and services more efficient and responsive – which is particularly important for environmental monitoring and emergency response missions where time is of the essence,” concludes Hankache. The ORCHIDE project concluded in May 2026. Since then, new partners have joined the consortium to further develop the project’s key results under the EU-funded ORION project.
