Traffic Urban Control Using an Intelligent PSO Algorithm Based on Integrated Approach

Abstract

In this paper, we intend to contribute to the improvement of urban traffic mobility using a learning method of traffic lights controllers. We proposed a Particle Swarm Optimization (PSO) method in which the intelligent swarm acts as the cycle time of the traffic signal. The best swarm (solution found) meets the evaluation criteria selected to describe desired objectives. The main measures of traffic lights efficiency are to maximize flow-rate at which vehicles can cross a road junction and minimize the additional travel time of the driver called vehicle delay. Particle Swarm Optimizer was coupled with the traffic flow model based on Continuous Petri nets (PN). One potential advantage of CPN model is to provide insights regarding a behavior of the platoon of vehicles on the target road network. The result obtained from this study has tested with various scenarios related to intersections in different situations. The developed self-scheduling of the optimal signal timing ensures safety and continuous traffic flow, thus increasing the mobility and reducing fuel consumption and pollutant emissions.