Throughout 2025 we collaborated with our partners at Software Republique, Orange, Renault, Keolis, Atos, Lab TBMouve, Bordeaux Metropole, University of Bordeaux and Allianz France in a series of field trials in the beautiful city of Bordeaux. The goal for us at Eye-Net was to test our mobile-based, multicloud, collision-avoidance technology in real urban conditions. We wanted to understand how our technology behaves across different apps and data networks, with varied user groups, in real world complex traffic scenarios.
By running the SDK simultaneously inside various live applications and collecting large volumes of anonymized data, the team was able to analyze various inputs and further optimize safety systems’ performance – outside controlled environments.
We’re still extracting the data from the trials and working to generate technical reports. Here are some very interesting initial technology insights:
1. One Network, Multiple Apps
One of the most important insights from the Bordeaux field trials was that Eye-Net’s safety network does not need to live inside a single app.
During the trial, Eye-Net’s SDK was integrated into several different applications at the same time – including an app by Orange, Keolis’ TBM app for cyclists, and the Atos app installed inside busses. Despite being separate products, all of these apps were able to communicate within the same safety network.
In practice, this means that going forward users don’t have to download one “central” app in order to be protected. Instead, safety becomes something that could exist across apps people were already using in their daily lives, whether for public transport, connectivity, or cycling.
2. Smarter Timing for Collision Alerts
Another major focus of the Bordeaux trials was improving the timing of collision alerts. Throughout the trial, Eye-Net analyzed a wide range of real-world scenarios, including beyond-line-of-sight situations, busy intersections, crossroads, roundabouts, and other complex urban layouts. These are exactly the environments where timing matters most, and where too-early or too-late alerts reduce trust in safety systems.
Using the data collected in the field, the team fine-tuned the precise moment alerts are issued: how early a warning should appear, how to adapt it to different movement patterns, and how to avoid unnecessary alerts that could distract users.
This optimization happened on two levels. On a macro level, large volumes of input data from the trials were reviewed to identify patterns across the entire network. On a micro level, the team gathered direct feedback and personal impressions from users who experienced the alerts in real situations.
The result is a more precise and context-aware alert system, one that can adapt to different traffic environments and be applied to many other cities and use cases beyond Bordeaux.
3. Turning Millions of Data Points into Long-Term Safety Insights
Over the course of the trial, Eye-Net collected tens of millions of anonymized data points. These included information such as speed, angle, location, accuracy, and movement patterns across the network.
This scale of data enables Eye-Net to go beyond real-time collision prevention. While instant alerts are critical, the data also opens the door to deeper, long-term insights about how people and vehicles actually move through cities. Our product Eye-Net View provides a dashboard that showcases this anonymized data. With this level of data, Eye-Net can analyze safety challenges across entire urban networks, identify recurring risk points, and highlight patterns that are difficult to see at street level.
These insights can then support smart-city planning, infrastructure improvements, and data-driven safety decisions.
In this way, Eye-Net’s value continues well after the alert itself.
From Trial to Scalable Ecosystems
The Bordeaux field trials demonstrate how smartphone-based V2X technology can evolve through real-world learning. By combining a multi-app network, data-driven alert time optimization, and large-scale insights, Eye-Net’s next phase focuses on scaling the technology and integrating it across multiple applications to create a protective network. This initiative aims to create a comprehensive layer that improves users’ quality of life, while serving as a scalable model, globally.
