Data-Driven and AI-Enabled Crash Investigation, Analysis, and Prevention for Safer Future Mobility

BackgroundRoad traffic crashes remain one of the most serious global public health and mobility challenges of the 21st century, causing roughly 1.19 million deaths each year and ranking as the leading cause of death among people aged 5–29. As transport systems move toward connected, automated, and electrified mobility, the volume and richness of crash-related data are growing fast: event data recorders (EDRs), on-board sensors, dashcams, roadside CCTV, telematics, and smartphones now capture detailed records of how crashes unfold. Yet the methods used to investigate, reconstruct, and learn from these data lag behind. Emerging risks such as smartphone distraction, impairment from alcohol and novel psychoactive substances, and mixed traffic between automated and human-driven vehicles demand new analytical approaches that turn crash evidence into prevention.GoalThis Research Topic aims to advance data-driven and AI-enabled methods for investigating, analysing, and ultimately preventing road traffic crashes. While crash causation is widely studied from epidemiological and engineering angles, the investigative dimension — how digital and physical crash evidence is captured, reconstructed, and translated into safer road and vehicle design — receives far less attention. We seek contributions that connect detailed crash investigation with system-level prevention under the Safe System and Vision Zero paradigms. The collection will bring together transport safety researchers, data scientists, and engineers to show how event data recorders, naturalistic driving data, computer vision, and machine learning can reveal crash mechanisms, quantify risk, and support proactive countermeasures. By linking micro-level crash reconstruction to macro-level safety policy and infrastructure design, the Topic intends to strengthen the evidence base for a future transport system in which serious injuries and deaths are eliminated.Scope and information for authorsWe welcome original research, systematic reviews, methods papers, case studies, and perspective pieces addressing: analysis of vehicle event data recorders (EDRs), dashcam, CCTV, telematics, and smartphone data for crash investigation; AI and machine learning for crash reconstruction, severity prediction, and pattern detection; computer vision, 3D laser scanning, photogrammetry, and UAV-based crash scene documentation; detection of alcohol, cannabis, opioids, and novel psychoactive substances in impaired-driving crashes; biomechanics of injury and occupant protection; safety of vulnerable road users and mixed automated–human traffic; and the translation of crash findings into infrastructure design, vehicle safety systems, and Safe System / Vision Zero policy. Studies that integrate multiple data sources, address data quality, privacy, and ethical use, or that demonstrate transferable, open, and reproducible methods are especially encouraged. Contributions from low- and middle-income countries, where the crash burden is highest, are strongly welcomed. Topic editors: TANURUP DAS, Yanyong Guo Source: https://www.frontiersin.org/research-topics/84006/data-driven-and-ai-enabled-crash-investigation-analysis-and-prevention-for-safer-future-mobility