Artificial Intelligence for Improving the Proactive Intelligent Safety of Pedestrians

Background and aims: Road traffic injuries are the main threat to public health and a seriouscause of concern across the world, especially in low and high-income countries. Pedestrian injuriesand fatalities, as vulnerable road users, have increased nearly every year. Recently, simultaneouslywith increased urbanization, Artificial Intelligence (AI) opened up new opportunities forintelligent traffic safety and has been widely used in order to deal with road traffic injuries anddeaths. The high vulnerability of pedestrians in road traffic crashes highlights the need to exploreand implement effective interventions in the realm of AI to protect pedestrians. The aim of thisstudy was to study the applications of AI for improving pedestrians’ safety.Method: This study has been done according to Arkesy and O’Malley’s framework according toa review of related literature from January ۲۰۰۰ to February ۲۰۲۳. Six main steps were conductedin this study as follows: identifying the research question, identifying relevant studies, study selection/screening, charting the data, collecting, summarizing, and reporting results, and providingpractical recommendations. To explore the relevant studies, the databases of PubMed, ScienceDirect, Scopus, and Web of Science were searched using predefined keywords. Literature screeningand selection were done according to Preferred Reporting Items for Systematic Reviews andMeta-Analyses, which extended to Scoping Reviews (PRISMA-ScR). Endnote software edition۷ was used for data management.Results: The findings of the literature review and screening showed that there are different approachesfor the application of AI in order to deal with road traffic accidents and prevent pedestrianinjuries and deaths. Applications of AI in preventing pedestrian accidents and improvingtheir safety were in various fields, including the main categories of ۱) pedestrian identificationand detection systems using the Vehicle-to- Pedestrian (V۲P) wireless communication technology,adaptability of autonomous vehicles (AVs) to communicate with pedestrians, and visibilityrelated systems, ۲) pedestrian accident prediction using the approaches like vehicle to everything(V۲X) systems, and ۳) pedestrian-assistance devices to help them by providing navigation andorientation information.Conclusion: AI involves a variety of technologies that facilitates proactive intelligence in pedestriansafety and is widely used in this area. Some of these technologies such as Vehicle-to- Pedestriancommunication systems, or autonomous vehicles can be used in limited countries becauseof the need for high technology and strong infrastructure. Although all countries should movetowards using different approaches to pedestrian safety promotion, it seems the use of pedestrians-assistant devices is more suitable for countries with limited needed infrastructures. Thesefindings can be useful for road safety policy-makers, technologists, and inventors in the design ofsystems and devices to promote pedestrian safety.

کلیدواژه ها:

artificial intelligence ، public health ، pedestrians ، safety ، proactive safety