MP-SAFE: Multimodal-Aware Pressure-Based Traffic Signal Control for Smart Urban Intersections

Authors:
  • LBLaisa C Costa De Biase
    Laisa C Costa De Biase
  • AGAndrey L B Gomes
    Andrey L B Gomes
  • AMAugusto R Machado
    Augusto R Machado
  • GGGuilherme Garcia
    Guilherme Garcia
  • HMHenrique Y Murata
    Henrique Y Murata
  • YAYasmin Azevedo
    Yasmin Azevedo
  • CMClaudio Marte
    Claudio Marte
  • MZMarcelo K Zuffo
    Marcelo K Zuffo

Abstract
Uncontrolled urban expansion has exacerbated traffic congestion and increased the number of accidents at signalized intersections, negatively impacting the quality of life in metropolitan areas. Despite advances in urban mobility technologies, there remains a gap in the adoption of adaptive traffic light systems that prioritize pedestrian safety and accessibility. This paper presents an autonomous and contextaware traffic light control logic focused on preserving human integrity while maintaining traffic flow, that are named MP-SAFE. The proposed solution models urban traffic as a network of interconnected queues and introduces a phase-based control algorithm that makes decisions based on the mathematical concept of pressure applied to traffic control. Each phase represents a time period during which, one or more traffic flows are granted to flow. The algorithm was implemented and tested in a microscopic traffic simulation environment representing real urban intersection. Simulation results demonstrate substantial improvements over fixed-time control: a 63% reduction in average delay, a 69% reduction in stop delay, and a 42% decrease in the total accumulated delay. Additionally, average vehicle speed increased by 116%, and distance traveled improved by 65%. This algorithm, associated with edge computing and computer vision has a great potential to contribute with the development of smarter, more efficient, and pedestrian-friendly cities.