A European Space Agency study has identified a significant shift in the positioning requirements of automated vehicles, concluding that the next major challenge is no longer simply achieving greater location accuracy.
The research, conducted by Stuttgart-based Acitoflux GmbH under ESA’s Navigation Innovation and Support Programme, or NAVISP, found that availability, robustness and the ability to verify positioning data are becoming more important as advanced driver assistance systems and autonomous-driving technologies move toward wider deployment.
The NAVISP EL1-115 bis project examined positioning, navigation and timing technologies across the emerging ADAS and autonomous-driving ecosystem. Acitoflux presented the final results to ESA on July 3, 2026. ESA describes it as the first completed NAVISP study focused on this specific subject.
Rather than treating GNSS accuracy as the primary performance metric, the study asked a more demanding question: can an automated vehicle determine whether its position estimate remains trustworthy when operating conditions deteriorate?
PNT Trust Becomes Critical
Modern high-end GNSS systems can already provide centimeter-level positioning under favorable conditions, particularly when RTK corrections are available.
That does not guarantee that the same positioning solution will remain dependable everywhere a vehicle is expected to operate.
Urban canyons can introduce severe multipath errors. Tunnels can remove satellite visibility entirely. Trees, buildings and vehicle structures can attenuate signals. Construction areas can invalidate map assumptions, while intentional or accidental radio-frequency interference can degrade GNSS without immediately making the failure obvious.
For human drivers, a temporary navigation error may be inconvenient. For an automated vehicle making steering, braking or route decisions, an undetected positioning error can become a safety problem.
This is why the industry’s challenge is increasingly moving from answering “Where am I?” to answering “How certain am I that this position is correct?”
That distinction is likely to shape the next generation of automotive positioning systems.
331 Companies Evaluated
The Acitoflux analysis examined a broad technology landscape rather than focusing only on GNSS receiver manufacturers.
The project screened 331 companies operating across PNT and automated-driving technologies, including vehicle manufacturers, Tier 1 and Tier 2 suppliers, GNSS correction providers, mapping companies, localization specialists, timing companies and V2X technology developers.
Researchers initially evaluated 49 potential use cases. Twenty-four use cases that had not yet reached Technology Readiness Level 9 were selected for more detailed analysis.
ESA says the work combined a review of the state of the art, identification of AD and ADAS applications and a structured gap analysis before producing recommendations for future projects.
The assessment covered ten PNT dimensions spanning technical performance, signal quality, resilience and the economic practicality of deploying different technologies.
GNSS Sensor Fusion Expands
One of the study’s clearest conclusions is that autonomous navigation is unlikely to depend on a single positioning technology.
Multi-frequency and multi-constellation GNSS is increasingly becoming the baseline rather than a premium feature.
At the same time, tightly integrated GNSS and inertial navigation systems are emerging as a core architecture for automated mobility. Vehicle motion constraints, wheel-speed measurements and other onboard information can help maintain a usable trajectory when satellite measurements temporarily deteriorate.
RTK technology is already considered sufficiently mature for many operational applications, while PPP and PPP-RTK remain important areas for further development.
Camera localization, LiDAR, simultaneous localization and mapping, inertial sensors and wheel odometry are also becoming more important.
Crucially, these technologies should not necessarily be viewed as competitors to GNSS.
Their strongest role may be as independent or semi-independent sources of positioning evidence that allow the vehicle to cross-check its GNSS solution.
Position Verification Matters
This approach changes the architecture of an autonomous positioning system.
Instead of continuously producing one coordinate and assuming it is correct, future systems will increasingly need to calculate confidence, detect inconsistencies between sensors and identify when navigation performance has fallen outside an acceptable operating envelope.
A camera-based localization system, for example, may confirm that GNSS is placing the vehicle in the correct lane.
An inertial measurement unit can detect whether a sudden GNSS position jump is physically plausible.
Wheel odometry can maintain short-term movement estimates through a tunnel.
LiDAR or radar localization can provide another independent reference when satellite reception becomes unreliable.
The result is not simply better positioning accuracy. It is greater positioning integrity.
For automated driving, that difference is fundamental.
New PNT Research Areas
The study generated several potential directions for future ESA-supported research.
ESA’s published results include concepts covering low Earth orbit augmentation for urban lane-level positioning, assured geolocation for teleoperation fallback and validation frameworks for automated mobility under low-visibility and heavy-rain conditions.
Other areas identified during the analysis include fleet-generated GNSS risk information, improved integration of PNT equipment beneath vehicle glass and national or cross-border systems for mapping interference and positioning resilience.
Such infrastructure could eventually allow vehicles to know not only their location but also the expected reliability of available positioning technologies along a route.
A vehicle approaching a known GNSS interference zone, for example, could increase dependence on inertial, vision or map-based localization before GNSS performance deteriorates.
Resilient PNT Gains Momentum
The findings also fit a much broader trend across ESA’s current navigation research portfolio.
NAVISP’s 2026 activities include projects covering resilient PNT, complementary sensors, interference-resistant GNSS receivers, 5G and 6G positioning and alternative positioning architectures. ESA added another five Element 1 activities during 2026 specifically in fields including assured PNT, complementary sensors and disruptive PNT technologies.
ESA is also supporting positioning systems that combine authenticated Galileo signals with 5G measurements and anti-spoofing techniques, demonstrating how navigation architectures are moving toward multiple independent layers rather than relying exclusively on satellite measurements.
Why This Matters
The most important conclusion from the study may be that the automotive positioning industry is approaching the point where nominal accuracy becomes a poor way to compare systems.
A receiver advertising 2-centimeter accuracy is not necessarily more useful to an autonomous vehicle than a 5-centimeter system if the second solution can reliably detect when its measurements are compromised.
That changes what manufacturers should optimize.
Continuity, integrity monitoring, interference detection, sensor redundancy and confidence estimation could become as commercially important as the positioning specification printed on a receiver datasheet.
There is also a potentially important role for fleet intelligence.
Millions of connected vehicles could collectively become a distributed GNSS monitoring network. If vehicles anonymously report unusual signal behavior, multipath conditions or localization discrepancies, manufacturers and infrastructure providers could build continuously updated PNT risk maps.
That would move positioning resilience beyond individual vehicle hardware and turn it into a network-level service.
For autonomous driving, this appears to be the logical next stage. The industry already knows how to calculate very accurate positions. The harder problem is proving that those positions remain trustworthy every second the vehicle depends on them.
About Acitoflux
Acitoflux GmbH is a German innovation and venture-building company headquartered in Stuttgart, with activities focused on technology sectors including future mobility, New Space and life sciences.
The company describes itself as an independent international innovation-capital partner and lists operations in Stuttgart and Hamburg along with studios or activities in Barcelona, Copenhagen, Paris and Tokyo. Its LinkedIn profile places the organization in the 11 to 50 employee category and states that the business was founded in 2018.
The German commercial register records Acitoflux GmbH under HRB 768722 in Stuttgart. The company was formally registered on March 20, 2019, with share capital of €25,000.
For ESA’s ADAS Technology and PNT project, Acitoflux assessed 331 companies and 49 automated-driving use cases before narrowing the analysis to 24 non-TRL9 applications, providing ESA with a structured view of where future automotive PNT research is likely to have the greatest impact.
