Information sharing among autonomous vehicles crossing an intersection

In this paper we compare the performance of autonomous vehicles at intersections with respect to the type of information shared. In addition to the position of other vehicles, in each simulated scenario one type of information is available. These types of information include the intended directions at intersection and the inertias of the other vehicles. An existing control method based on navigation functions is modified in order to benefit from the availability of such information. The results show that if autonomous vehicles know each other´s inertia they achieve significantly smoother paths, use less fuel and avoid full stops.