Abstract
Smart mobility initiatives encompass innovative methods to reduce traffic congestion and ecological footprint. Technologies such as IoT-connected vehicles, coupled with artificial intelligence, are emerging as a solution to collect and analyze traffic data in large-scale urban environments. This allows supporting traffic management experts in decision-making tasks to improve urban infrastructures and reduce the ecological footprint. Accurate and continuous information about traffic is necessary to implement such decision-making tasks. This is not always available because of the cost of the information: on the one hand, it is not possible to install sensor devices everywhere because of financial costs and privacy; on the other hand, a high quantity of sensors requires important computational capabilities, and centralized data analysis approach can hinder real-time or near-real-time applications and limit their practical deployment in real-world traffic management systems. This paper presents a novel privacy-aware, edge-based method for estimating traffic density without over-deploying privacy-intrusive surveillance technologies such as cameras. The objective is to reduce the cost of traffic information while providing accurate information to support traffic management experts in decision-making tasks.
Funding Information
  • Traffic prOcessing foR uRban EnvironmentS (TORRES), a Joint Research and Development Project
  • “Région de Bruxelles-Capitale-Innoviris” (2022-RDIR-59b)
  • UK Research and Innovation (UKRI) Future Leaders Fellowship: ‘Digitally Assisted Collective Governance of Smart City Commons–ARTIO’ (MR/W009560/1)
  • Alan Turing Fellowship