Platoon-based multi-agent intersection management for connected vehicle

Abstract

As wireless communication advances, multi-agent system (MAS) approaches to intersection traffic management have received increased attention. In this paper, we propose an improved multi-agent intersection management system where vehicle agents may form platoons using connected vehicles technologies. Traffic performance measures in terms of mobility and environmental sustainability were evaluated through a series of simulation studies, along with an analysis of the communication load of the system. Compared to the conventional traffic signal control system, the proposed platoon-based multi-agent intersection management system can shorten the average travel time by as much as 30% and reduce the fuel consumption and carbon dioxide emissions by around 23%, when the traffic volume is high. In comparison with a non-platoon-based multi-agent intersection management system, the proposed system can reduce the communication loads of an intersection management agent by up to 90% and exhibits strong robustness against the variation of traffic volumes.