Safe Merging Control in Mixed Vehicular Traffic

Abstract

Despite the potential benefits of a traffic system with only Coordinated and Automated Vehicles (CAVs), it is expected that Human Driven Vehicles (HDVs) and CAVs co-exist for the foreseeable future. Due to uncertainty and the unpredictability of human drivers, developing a control framework with safety guarantees, especially in the traffic bottlenecks such as merging points is a challenging problem. Motivated by this fact, in this paper we study a merging problem in mixed vehicular traffic and we develop a safety-critical real-time decentralized control of CAVs in the presence of HDVs. We use Control Lyapunov Functions (CLFs) to attain the desired control objectives, and Control Barrier Functions (CBFs) to guarantee the safety of the merging operation. It is assumed that a high level coordinator determines the sequence of vehicles pass the merging area and forms a triplet of vehicles in the main lane and the merging lane. Then, three different combinations of CAVs and HDVs are considered and for each one, required CLFs and CBFs to safely accomplish the merging operation are developed. Simulation results are provided to demonstrate the efficacy of the proposed schemes.