The mission reliability of satellites depends as much on how they are managed from the ground as on how they operate in space. It’s here that Sopra Steria has decided to focus its expertise by managing the security of data exchanges between the ground and satellite, explains Nicolas Frouvelle, Marketing and Communications Director for Space at Sopra Steria's CS Group.
In space, every line of code commits to years of autonomous operation, billions in investment and sometimes human lives. Contrary to popular belief, the reliability of a satellite does not rely solely on the robustness of its onboard components; instead, it is first built on the ground, in the systems that observe it, control it and anticipate its behaviour.
The space sector is undergoing a profound transformation. Hundreds of new satellites are sent into orbit each year, constellations of several thousand craft are reshaping the space economy and exploration missions are venturing ever further. In this context, software complexity is exploding.
“The major challenge of the next five years is knowing how to handle this incredible increase in the number of satellites and the volume of data that it generates,” says Nicolas Frouvelle. Faced with this challenge, Sopra Steria’s teams have built a deep and differentiating expertise – the ground segment, where mission reliability is actually determined.
The ground segment: Where the mission comes to life
A satellite, once in orbit, does not operate alone. It continuously sends a stream of data to ground reception stations (telemetry) and, in return, receives commands that guide its manoeuvres, activate its equipment and correct its trajectory. It is at this level that CS Group, part of Sopra Steria, operates, with software that processes these exchanges in real time.
“We secure the information exchange between the satellite and the ground, whether from the ground to the satellite or from the satellite to the ground and we secure our control centre as much as possible,” explains Frouvelle. This includes continuously verifying the reliability of incoming data, detecting anomalies in the data streams and the ability to isolate a failing data provider without interrupting operations. If a satellite component begins transmitting inconsistent information, for example, due to intense solar activity, the ground systems detect it, isolate the problem and adapt.
This role in no way diminishes the importance of onboard software, developed by specialised equipment manufacturers whose core business it is. Instead, it complements it. The robustness of a space mission is born from this pairing.
OREKIT: When open source becomes a global standard
At the heart of CS Group’s software expertise in orbitography lies OREKIT (Orbit Extrapolation Kit), an open-source orbital mechanics library initiated by Luc Maisonnobe, a former CS Group engineer. Its first operational use dates back to 2008, during the validation of the final approach of the automated transfer vehicle (ATV) Jules Verne to the International Space Station. Since then, the library has continued to grow.
OREKIT, recognised as the second most widely used Flight Dynamics Core Library in the world, provides the functions necessary for orbit calculation and propagation – predicting where a satellite will be in the coming hours and days, reconstructing its actual trajectory from received data and anticipating critical events such as passes over a ground station, eclipse periods, or risks of collision with debris. These capabilities are integrated into Sopra Steria’s commercial products, including GOSMIC (Ground Operation System and Mission Intelligence Centre), its modular ground segment.
The decision to make the library open source transformed an internal tool into a global strategic asset. “OREKIT now benefits from developments proposed by international experts,” says Frouvelle.
The Project Management Committee that governs OREKIT now brings together experts from Sopra Steria (CS Group), Thales Alenia Space, Airbus Defence and Space and the European Space Agency, along with several European and American institutional, industrial, and New Space partners. The Indian and Japanese space agencies use it. Various constellation operators use it to manage their fleet mechanics. It is a global success.
The library is freely accessible, but the expertise required to master it and integrate it into operational systems is not. Whether it’s training, support or custom developments, Sopra Steria will use its position as a technical benchmark to offer high value-added services. “We sometimes learn that a competitor has won a contract by proposing a solution based on OREKIT. While this does not directly benefit us, we are happy that it still enhances the product’s reputation,” says Frouvelle.
The constellation revolution: A new operational paradigm
The rise of New Space has radically changed the software equation as human management resources shrink. “Previously, several engineers were assigned to a single satellite,” says Frouvelle. “Now, actions must be pooled. This requires a paradigm shift in operations.”
This transformation demands open and scalable software architectures capable of maintaining their performance whether managing 10 or several hundred satellites at once. The robustness tests carried out by Sopra Steria on GOSMIC, simulating fleets of several hundred satellites, are designed to validate this scaling up. It also calls for new operational agility, concentrating human and computing resources where the need is immediate, depending on the criticality of the moment.
Regulatory considerations are also entering the equation. As the number of satellites grows, manoeuvrability and deorbiting requirements are being strengthened. SpaceX, for example, lowered part of the Starlink constellation by 100 kilometres to accelerate the natural disintegration of satellites at the end of their life. Ground software will have to integrate these new compliance constraints, in real time, for each craft in a fleet.
The next frontier: AI to anticipate anomalies
Artificial intelligence opens a new perspective for supervising space systems, going beyond reactive detection and instead becoming predictive. Sopra Steria is working on integrating AI-based anomaly detection capabilities into its control centre solutions. The objective is to detect a malfunction even before a formal alert is triggered, by continually analysing the exchange streams between the satellite and the ground.
“By using AI to analyse the way these data packets arrive, we will be able to identify in advance that an anomaly that is going to occur,” says Frouvelle. This predictive approach complements existing detection mechanisms without replacing them. Humans remain in the loop for critical decisions, with AI enhancing their ability to anticipate rather than merely react.
It is on this combination of AI and human expertise that Sopra Steria builds its vision of space software – proven orbital expertise, open architectures and AI tools serving operators. It’s an approach focused on operational mastery rather than technological hype and of which OREKIT, nearly 20 years after its launch, remains perhaps the most enduring symbol.