Road Side Perception Systems for Safer Intersections: Field Test

Intersections have been known as hazardous points of the road networks for over a century. A diversity of solutions have been developed to reduce the number of accidents at intersections, such as traffic lights, pedestrian crossings, as well as roundabouts. In the past decade, many cities have had to deal with a growing number of alternate means of transport such as bicycles, powered two-wheelers and electric personal transporters, which are both more mobile and more vulnerable. As for the announced automated vehicles, they might be less adaptive that human drivers. These changes again raise the question of how accidents can be prevented at intersections of roads, bus lanes, cycling lanes and walkways. In order to improve the awareness of road users, we propose a cooperative information system based on computer vision to monitor moving obstacles and on communication to warn road users facing imminent danger of collision. We have implemented and tested this system on a roundabout with a high traffic volume, and we report the difficulties that we experienced.