Fuzzy Pedestrian’s Risk Perception and Notification in Fuzzy Neighborhoods

Abstract

Analyzing pedestrians’ behavior in smart cities relies on the understanding of the human cognitive abilities, including the ability to focus on the traffic environment, and the understanding of the perceived spatiotemporal object’s semantics. It relies also on understanding the flow activity on a high-level scale of what’s circulating around the pedestrian and over all the corners of the cities. There is a lot of risk on pedestrian while they are interacting with the world and sadly that can lead to accidents. We find that one of the causes of pedestrian accidents is inattention, pedestrian gets easily distracted by their digital devices, or also by events nearby. To minimize the pedestrian accidents and help increasing their assurance of safety, we propose in this paper, a model oriented over road crossing to percept risk on pedestrian, based on spatial analysis and notify the pedestrian as fast as possible using a scalable distributed architecture and a push-based notifications approach. We used a theory of risk perception in a spatial neighborhood, by constructions based on fuzzy neighborhoods and fuzzy pretopology.