Evolutionary swarm traffic: If ant roads had traffic lights

Traffic congestion has become a major concern for many cities throughout the world. Consequently, simulations can provide helpful tools for engineers to plan traffic systems. In this paper we present SuRJE, a swarm-based traffic simulation system. All interactions among individual drivers are implemented through the dropping and sniffing of pheromones on a 2-dimensional map. Furthermore, SuRJE optimizes traffic light turn and time sequences using an evolutionary algorithm and swarm voting, which helps to speed up the adaptation process for the traffic light settings.