Application of smart technologies in safety of vulnerable road users: A review

Road safety is the most important feature of a modern city, and it affects almost everyone in the community, especially vulnerable road users (VRUs). This paper comprehensively examines existing scientific literature regarding contemporary methodologies for collecting data in safety studies involving VRUs. The objective is to compile a comprehensive list of data collection methods, recognize potential applications of emerging technologies, and categorize them based on a novel taxonomy. A preferred reporting items for systematic reviews and meta analyses (PRISMA) flowchart is used to conduct the systematic literature search by setting some inclusion and exclusion criteria. Different keyword searches are used in Scopus and Web of Science databases, followed by relevant references and citation analysis to find eligible papers subject to a full-text peer review. Finally, the identified papers are categorized and analyzed based on the technology type they used. 8 374 and 109 papers have been identified from the initial search and the forward and backward snowballing, respectively. 167 documents have been selected to carry out full-text reviews, with 135 finally included in the study. The technology employed in safety research for VRUs, including cameras, sensors, trackers, mobile phones, social media, drones, and eye-tracking devices has also been included in the classification of identified documents. Commonly employed methods for collecting data on VRUs include camera-based, sensor-based, and tracker-based approaches. The mobile phone-based approach has been least common for collecting data on pedestrians’ safety because of distractions. In recent years, social media-based, drone, and eye-tracking approaches have become widely utilized for collecting and analyzing data. Recently, multiple approaches have been employed for data collection. The documents predominantly have addressed the movements, behaviors, emotions, and route choices of pedestrians. Similarly, documents related to cyclists have been mainly concerned with obstacle detection, analysis of cyclists’ behavior, and guiding cyclists.