New ESA demonstrators are intended not only to advance early PNT technologies, but to validate them under real conditions in space. This reflects more than classic space research: Europe is preparing for navigation architectures that are meant to be more capable, more adaptable and more resilient than today's individual systems.
Position, navigation and timing remain invisible to many digital processes – until they are missing or can no longer be trusted. Communication networks require precise time references. Transport and logistics systems depend on reliable position information. Energy, financial and security infrastructures synchronise their operations via external time signals. Autonomous systems, too, must not only know where they are, but must also be able to judge how dependable that information is.
The growing dependence on PNT services therefore raises a fundamental question: how can new technologies be tested in a way that demonstrates their performance not only in the laboratory, but within a complete system?
The European Space Agency (ESA) intends to make greater use of the orbit as a test environment for exactly this purpose.
From technology idea to system experience
With its call for future PNT demonstrators, ESA is looking for disruptive and early-stage technologies that can be developed further for later in-orbit demonstration missions. The call is part of a broader ESA portfolio that includes, among other things, software-defined PNT payloads, the use of artificial intelligence and quantum-technology approaches. The aim is not to consider new system concepts in isolation, but to examine them on their path towards a possible application under realistic conditions.
The underlying ESA contribution was published on 23 June 2026. This is therefore not about a single new satellite mission, but about building a continuous capability for innovation and validation.
That distinction is decisive.
An algorithm can deliver excellent results in the laboratory. A new frequency can show advantages under controlled conditions. A software-defined payload can be flexibly reconfigurable. For later use in critical applications, however, numerous further questions must be answered:
- How does the technology behave in the interplay between space, ground and user segment?
- What errors arise from real orbit, timing and propagation conditions?
- How reliably can performance be monitored?
- How quickly can disturbances or implausible states be detected?
- What dependencies arise with respect to other system components?
- What information does a user need in order to assess the quality of a service?
In-orbit demonstrations therefore create more than technical maturity. They generate system knowledge.
PNT is evolving from an individual system into an architecture
The initiative is linked to a broader shift in satellite-based navigation. Future PNT services will most likely not rely on a single constellation, a single frequency band or a single sensor technology alone.
ESA's Celeste mission already illustrates this development in concrete terms. The planned LEO-PNT demonstration constellation comprises eleven satellites and is intended to test new signals in different frequency bands. In particular, it will examine how a navigation layer in low Earth orbit can work together with Galileo, EGNOS and other GNSS. ESA explicitly describes this approach as a multi-layered architecture designed to increase robustness and resilience.
LEO-PNT is not simply a "second GNSS".
From the user's perspective, satellites in low Earth orbit move across the sky more quickly, can enable stronger signals at ground level and can provide additional geometric information. At the same time, new requirements arise regarding orbit determination, time synchronisation, transitions between satellites, receiver architectures and the assessment of service quality.
The added value therefore does not stem from an additional signal source alone. It stems from combining different characteristics.
An additional PNT layer only increases resilience if it has error and failure characteristics that are sufficiently independent of the existing systems. If several services are affected by the same source of interference, the same reference or the same processing chain, apparent redundancy can lead to a false sense of security.
New frequencies are an option – not an automatic solution
Alongside established L-band signals, Celeste is also intended to investigate further frequency bands. The S-band, for example, could enable new links between navigation and communication. The C-band, owing to its signal characteristics, promises additional resistance to certain types of interference. In this context, ESA explicitly refers to jamming, spoofing, authentication and new signal structures.
Yet even a stronger or differently structured signal does not fully solve the resilience problem.
Every new frequency brings its own requirements in terms of antennas, hardware, regulation, power demand and integration. It must also be clarified which user groups can actually receive new signals, and how the transitions between different PNT sources are managed.
For critical infrastructure, what matters is therefore not only whether a new signal is technically available. What is decisive is whether it can be embedded into a comprehensible operational architecture.
This includes:
- safety processes with configuration and incident management,
- continuous quality and integrity monitoring,
- traceable trust values,
- independent plausibility checks,
- prepared fallback procedures,
- safe transitions into degraded operating states.
A technology can be successfully demonstrated in orbit and still not yet be immediately suitable for safety-critical applications. Between technical feasibility and operational usability lies a development stage in its own right.
How tangible this step from technical demonstration to operational usability is can also be seen from the user's perspective. The automotive industry set out its expectations of the LEO-PNT constellation now being built in the VDA position paper on LEO-PNT: reliable, resilient PNT services are regarded there as a basic prerequisite for automated and autonomous driving, and the sector explicitly names the need for a dedicated safety-of-life PNT service. At the same time, the paper stresses that the benefits of LEO-PNT remain theoretical as long as they are not validated in application-oriented campaigns – with systematically measured and published parameters such as availability, integrity and latency. This is precisely where the planned demonstrations come in.
Software-defined payloads change the development process
ESA's approach of investigating reconfigurable and software-defined PNT payloads is particularly strategic. Such systems could adapt signals, processing methods or operating modes after launch, rather than being restricted to a largely fixed configuration for the entire duration of the mission.
This opens up new possibilities:
New signal formats could be tested step by step. Insights from operations could feed directly into modified configurations. Systems could respond to new requirements or changing interference environments.
At the same time, new risks emerge.
As reconfigurability grows, so do the requirements for cybersecurity, configuration control, verification and evidence keeping. A flexible system must at all times make clear which state it is in, which software version is active and how a change affects users and dependent services.
Resilience, in this context, does not mean adaptability alone. It also means being able to control changes and to assess their consequences.
The orbit becomes part of the development cycle
Traditionally, the launch of a satellite often marked the transition from a development project to operations. With future PNT demonstrators, this logic changes. The orbit itself becomes an environment for development, learning and validation.
Celeste, for example, invites manufacturers, application developers, research institutions and organisations from the field of public safety to test LEO-PNT technologies with real satellite signals. The demonstration phase is thus intended not only to test the space segment, but also to generate experience on the user and application side.
This openness is relevant because PNT performance is not created solely within the satellite. It is the result of a complete chain:
signal generation, transmission, reception, processing, quality assessment and operational use.
Only when this chain is considered as a whole can the limits of the system be recognised.
A service may, for example, achieve high nominal accuracy while its error detection is too slow for a critical application. A signal may be available while its integrity cannot be sufficiently assessed. A receiver may use several constellations even though all its input data are impaired by the same local interference.
Such interdependencies become more visible in real experiments than in isolated component tests.
The strategic question is not: which technology wins?
In debates about the future of PNT, technologies are often set against one another: classic GNSS versus LEO-PNT, radio navigation versus inertial methods, terrestrial versus satellite-based systems.
For resilience, this framing falls short.
The decisive question is not which single system replaces all the others. What matters is how different technologies can be combined into an architecture whose overall state remains observable and manageable.
In doing so, operators of critical applications do not need to control every technical detail of a satellite system. They do, however, need to know:
- which external services their operations depend on,
- how the quality of these services is monitored,
- which common causes of failure exist,
- how deviations are detected and communicated,
- what consequences a failure has for the mission,
- which alternative operating modes are available.
In-orbit demonstrators can deliver valuable insights for this. They do not, however, replace application-related risk and system analysis on the ground.
From demonstrator to resilient infrastructure
The current ESA call shows that Europe increasingly understands PNT innovation as an ongoing process. New technologies are to be moved more quickly from early development stages into real test environments. At the same time, navigation, communication, quantum technology, artificial intelligence and software-defined systems are being brought more closely together.
For Europe's technological capacity to act, this is an important step. Strategic sovereignty does not arise solely from developing components in Europe. It also arises from the ability to test new system concepts oneself, to understand their limits and to assess their suitability for critical applications independently.
The orbit thus becomes the proving ground for a new generation of PNT systems.
Whether this actually results in more resilient infrastructure is not, however, decided in space alone. It is decided at the interface between signal, system and operations.
The DiMOS perspective
An additional PNT service does not, by itself, create resilience.
Resilience arises when different sources are deliberately combined, continuously monitored and used within clear operational processes. Future in-orbit demonstrators should therefore not be assessed solely in terms of accuracy, availability or signal strength.
Equally important are:
- the detectability of faulty states,
- independence from common causes of failure,
- the quality of the integrity information,
- integration into monitoring and mission control,
- and the behaviour of the overall system under degraded conditions.
The strategic value of the ESA initiative therefore lies not only in individual new technologies. It lies in the opportunity to learn, at an early stage, how they can be turned into robust PNT architectures.
