Decentralized Spatial Decomposition for Traffic Signal Synchronization

The Traffic Signal Synchronization is a traffic engineering technique of matching the green light times for a series of intersections to enable the maximum number of vehicles to pass through, thereby reducing stops and delays experienced by motorists. Synchronizing Traffic Signals ensures a better flow of traffic and minimizes gas consumption and pollutant emissions. In this paper we provide a solution to the the traffic signals problem via simulation. The objective function used in this work is a weighted sum of the delays caused by the signalized intersections, and it is calculated by platoon model. Urban signal timing is a non-convex problem and finding an optimal solution for not very small and simple networks may take long time, wherever possible. For this reason we proposed a spatial decomposition of the network, it is obtained by the distributed consensus algorithms. In this paper we provide a distributed communication architecture for a network of smart traffic lights. Each semaphore shares information in a fully distributed way, only with its neighbors according to the topology of the communication network, hence avoiding to resort to a central authority. Given the subnetwork a surrogate method is applied to solve the Traffic Signal Synchronization problem.