The optimal coordination of connected and automated vehicles at roundabouts

Uncoordinated merging of vehicles at roundabouts is one of the main causes of traffic congestion, which directly affects travel delay, fuel consumption, and emissions of individual vehicles. Recent advances of automated vehicle (AV) and vehicle-to-vehicle (V2V) communication technologies has led to interests to implement optimal merging system to improve the traffic flow performance at roundabouts. This paper presents a novel cloud-based roundabout coordination system (RCS) for AVs at a four-leg roundabout to achieve smooth traffic flows near the capacity for a connected traffic environment. The coordination of vehicles is implemented in two levels. In the higher level, vehicles approaching the roundabout are coordinated to gradually form small clusters. In the lower level, the optimal sequences of vehicles and their merging times into the roundabout are computed by solving a combinatorial optimization problem, which is implemented in a receding horizon control (RHC) framework. The objective of the optimization considered is to minimize the total merging time of all vehicles with minimum impacts on the vehicles that are already flowing in the roundabout. The performance of the proposed RCS is evaluated through microscopic traffic simulation and benchmarked with the existing traditional roundabout system (TRS).