A pheromone-based traffic management model for vehicle re-routing and traffic light control

We propose a pheromone-based traffic management model, which simultaneously optimizes vehicle re-routing and traffic light control. Specifically, each car agent deposits multiple digital pheromone on its route. The road infrastructure agents fuse the pheromone to forecast traffic conditions. Once a congested road is predicted, we adopt a proactive vehicle re-routing algorithm for assigning alternative routes to cars before they enter the congested road. At the same time, the traffic light control agents use an online strategy for assigning long time duration of green traffic lights to the roads with a large amount of pheromone. Experimental results confirm its superiority in alleviating traffic congestion.