Every day, millions of children walk and cycle to school, often along roads that were never designed with their safety in mind. Identifying risks and prioritising improvements is essential, but traditional road safety assessments can be time-consuming and resource-intensive. Through the Google.org-supported AI&Me: Leveraging AI Tools for Road Safety Impact in Vietnam, iRAP and its partners are changing that by harnessing artificial intelligence and expanding the scope of Star Rating for Schools assessments to better reflect how children travel today.
A major breakthrough of the project has been the development of AI-powered systems capable of automatically identifying road safety features from street-level imagery and earth observation imagery. Using machine learning, the system can recognise elements such as crossings, traffic signs, street lighting, speed management measures, and intersections, translating them into the attributes required for SR4S assessments. This dramatically reduces the need for manual coding while enabling assessments to be carried out at a much larger scale. We are performing assessments in more than 900 schools in target project provinces in Vietnam, Vĩnh Long et An Giang.
The technology is helping transform SR4S from a location-based assessment tool into a platform capable of analysing entire school routes. New functionality developed through the Google project supports bulk data uploads and continuous assessments, creating opportunities to assess hundreds of schools more efficiently and consistently than ever before.
At the same time, SR4S is evolving to better support sustainable mobility. Recognising that many children cycle to school, the programme is being expanded to incorporate the bicycle Star Rating model from iRAP’s latest methodology, Version 3.10. This enhancement allows SR4S to assess cycling risks alongside pedestrian risks, providing a more complete picture of safety around schools and helping decision-makers identify improvements that benefit all young road users.
The introduction of bicycle safety assessment is particularly significant as cities around the world seek to encourage active travel, reduce congestion, and improve public health. By integrating cycling into school safety assessments, SR4S can now support evidence-based investments that make walking and cycling safer and more attractive options for children and families.
The three new SR4S enhancements are currently undergoing internal testing and are expected to be piloted externally next year. If you are interested in participating in the pilot programme or learning more about these developments, please get in touch.
The project demonstrates how digital innovation can accelerate road safety action. By bringing together AI-driven attribute detection, continuous assessment capabilities, and the new bicycle Star Rating model, SR4S is becoming a next-generation platform designed to protect children wherever their journey begins.