Bi-level Ramp Merging Coordination for Dense Mixed Traffic Conditions

Connected and Autonomous Vehicles (CAVs) hold great potential to improve traffic efficiency, emissions and safety in freeway on-ramp bottlenecks through coordination between mainstream and on-ramp vehicles. This study proposes a bi-level coordination strategy for freeway on-ramp merging of mixed traffic consisting of CAVs and human-driven vehicles (HDVs) to optimize the overall traffic efficiency and safety in congested traffic scenarios at the traffic flow level instead of platoon levels. The macro level employs an optimization model based on fundamental diagrams and shock wave theories to make optimal coordination decisions, including optimal minimum merging platoon size to trigger merging coordination and optimal coordination speed, based on macroscopic traffic state in mainline and ramp (i.e., traffic volume and penetration rates of CAVs). Furthermore, the micro level determines the real platoon size in each merging cycle as per random arrival patterns and designs the coordinated trajectories of the mainline facilitating vehicle and ramp platoon. A receding horizon scheme is implemented to accommodate human drivers’ stochastics as well. The developed bi-level strategy is tested in terms of improving efficiency and safety in a simulation-based case study under various traffic volumes and CAV penetration rates. The results show the proposed coordination addresses the uncertainties in mixed traffic as expected and substantially improves ramp merging operation in terms of merging efficiency and traffic robustness, and reducing collision risk and emissions, especially under high traffic volume conditions.