A bike, a shared e-scooter, ten minutes to the lecture hall – for plenty of students that is simply how the morning works. The car beside them has airbags, ABS and assistance systems; the person on the scooter has a helmet, at best. The EU-funded SOTERIA project asked a different question: instead of counting collisions after they happen, what would a city need to collect to see one coming? The answer runs through sensors, artificial intelligence and riders themselves – starting with whether your own near miss ever reaches anyone.
Bicycles, e-scooters and other forms of micromobility have become commonplace in urban environments, and that has put the safety of the people using them under growing scrutiny. The imbalance is easy to describe. Cars come with advanced safety technologies, and they usually move on infrastructure that was designed for them in the first place. Micromobility users do not have the same levels of protection – neither in the vehicle they are riding nor in the street layout around it.
This gap has turned into a European policy priority. The long-term goal of the EU's vision zero strategy is to move as close as possible to zero fatalities in road transport by 2050, and a target like that exposes the weakness of the usual method. Crash statistics describe what has already gone wrong: someone has to be hurt before a junction shows up in the numbers at all. SOTERIA, funded by the EU, looked at how cities can move beyond relying solely on those statistics by combining mobility, infrastructure and behavioural data to identify risks before collisions occur.
Everyone who rides through a city knows the moment – a door opening, a car turning across the lane, nothing actually happens. Nobody records it. That is precisely the material the project went after. "The project team also analysed policy and accident reports while systematically collecting information on near misses using sensor kits and smart cameras," says SOTERIA project member Marina Georgiou from Netcompany in Luxembourg. "Near misses are often the missing piece of the puzzle because they reveal dangerous situations before serious collisions occur."
The other half of the method was talking to people. Open-ended workshops let road users describe their own mobility experiences and think about solutions for everyday problems, and the participants were not only commuters. "A delivery company for example was interested in finding ways of better protecting its employees," Georgiou notes.
The approaches were validated across four Living Labs, each standing for a different mobility environment: Madrid as a dense metropolitan setting, Saxony as a mix of urban and rural contexts, Oxford as a university city with high levels of active travel, and Chania together with Igoumenitsa for the pressure of seasonal tourism and rapidly changing traffic patterns. Oxford collected evidence of safety issues for cyclists and tested sensory solutions meant to improve the safety of pedestrians and micromobility users, while Madrid trialled a prototype digital tool for more proactive road safety work, in collaboration with the local police.
Some findings were local – Oxford, for instance, lacks underpasses at intersections that need them. But a set of issues came back everywhere:
None of these is exotic engineering. They are the things a student notices on the way to campus and usually stops noticing by the second week of term.
From that common ground the team identified solutions local authorities and policymakers can actually pick up: comprehensive infrastructure upgrades and adaptations, and awareness campaigns promoting safer road use. On the technical side, SOTERIA brought together multiple sources of mobility and infrastructure data for predictive road safety analysis, explored ways of enabling secure data sharing between stakeholders, and developed AI-supported analysis, near-miss detection, safer route planning and situational awareness tools. The prototype applications include safer route planning, behavioural nudging and citizen reporting of hazards – which means a rider becomes a source of data rather than a statistic. The project runs from 1 November 2022 to 30 April 2026 under grant agreement 101077433, with a total cost of €3 319 325.01 and an EU contribution of €3 319 325.00, coordinated by NETCOMPANY S.A. in Luxembourg.
Beyond the technology, SOTERIA leaves cities with practical guidance on proactive road safety, data governance, stakeholder engagement and policy recommendations – a foundation for municipalities that want to reduce risks before collisions happen. For anyone weighing up transport, urban planning or data science, the interesting part is the job description hidden in there: road safety today is work done on datasets, workshop notes and camera feeds, not only on asphalt.
5 questions · one minute · nothing is saved
EN 
CS
DE
ES
FR
HU
IT
PL
PT
РУ
SK
TR
УК
AR
中文
