Maneuver coordination for connected automatic vehicles in mixed traffic

Automated vehicles can increase their awareness radius beyond their sensor range using Ad-hoc vehicular communication networks (VANET). Vehicles send messages that can propagate for hundreds of meters and get interpreted by other traffic participants. Currently, message types to exchange vehicle status like position and speed have been standardized. Researchers have proposed a Maneuver Coordination Service for Connected Automated Vehicles (CAV) to negotiate driving maneuvers such as lane changes using VANET. This doctoral research project addresses some of the weak points of available maneuver coordination concepts: Concurrent negotations between three or more vehicles can lead to conflicts, zip mergers can trigger a cascade of new maneuvers, and vehicles might take other counterproductive decisions.

A new coordination service is developed to be used together with vehicle platoons and implemented in an Artery-Veins simulation environment (including SUMO traffic simulation). A motorway merging scenario allows to compare the coordination service with previously published concepts and evaluate its sensitivity to traffic density and to the CAV share.