Predicting and avoiding road crashes based on Artificial Intelligence (AI) and big data

Proposals must build an AI-enabled digital twin of traffic and infrastructure that combines historical, current and forecast data — from crowdsourcing, infrastructure sensors, weather and traffic conditions — to flag safety-critical situations before crashes happen and support real-time interventions. Work must analyse the technical challenges of gathering and combining big data from multiple road-transport sensors, account for and remove bias in the AI models used, and separately address the non-technical issues of privacy, data ownership and other legal barriers to sharing that data. Proposals must also apply FAIR data principles for interoperability, build on relevant Common European data spaces, and demonstrate that the predicted risks and interventions are workable in real time. The work programme separately recommends international cooperation, naming projects or partners from the US, Japan, Singapore and Australia.