An approach to advisory-based traffic control

In this paper we propose an architecture integrating a multi-agent system and a centralised solution for intelligent traffic light control, aiming at the overall improvement of the network's performance under the cooperative and the adaptive assumptions. Knowing that most decentralised traffic control solutions do not offer direct guarantees on the overall network's performance, we propose the inclusion of a higher-level agent responsible for the analysis of the entire network performance from a macroscopic perspective. This agent will then communicate with traffic light controlling agents in order to provide them with useful suggestions based on the globally perceived information. We present a case study in order to evaluate the impact of our proposed architecture as an intelligent traffic management solution.