Cooperative Platooning in Mixed Traffic of Connected, Automated, and Human-Driven Vehicles

University Of Virginia Main Campus, Charlottesville VA

Investigators

Abstract

This research project will develop cooperative platooning algorithms for mixed traffic of connected automated vehicles and conventional human-driven vehicles. Recent development of connected and automated vehicle technology allows a group of such vehicles to travel closely one after another in a safe manner (known as, cooperative platooning), which greatly improves mobility and energy efficiency. However, when a human-driven vehicle exists within the group, the cohesion of vehicle platoon is not possible, due to uncertain human driver behavior. The novel cooperative platooning algorithms to be developed in this project will enable connected and automated vehicles to safely follow human-driven vehicles at shorter headway while mitigating traffic disturbances. These algorithms will also be able to assist a human driver in connected-but-not-automated vehicle by complementing human’s imperfect behaviors. As such, the cooperative platooning can be efficiently operated at low market penetration of connected automated vehicles. This research will benefit national economic welfare and public health with improved surface transportation mobility and reduced greenhouse gas emissions. This research will enable multi-disciplinary education and collaboration in transportation engineering, cyber-physical systems, control theory, and mechanical engineering. The research team will encourage participation from diverse and underrepresented groups in the education and research. Connected automated vehicle (CAV) has been enabled to stably travel as a platoon with short headway, which leads to improvements in mobility and energy efficiency. However, it fails to work effectively in mixed traffic where CAVs are interacting with non-CAVs. The goal of this research is to develop and validate Cooperative Adaptive Cruise Control in mixed traffic (CACC-MT) that can safely and efficiently stabilize the mixed traffic including CAV, traditional vehicles and connected human-driven vehicles. CACC-MT makes the CAV capable of performing feed-forward control or model predictive control using the received information from a further preceding vehicle, when the immediately preceding vehicle is unconnected. This allows CAV to closely follow an unconnected vehicle. CACC-MT also includes a human-in-the-loop CACC algorithm that enables co-piloting the human driver based on received information from preceding connected vehicle and help the vehicle behave more smoothly and safely in the traffic turbulence. As CACC-MT adopts robust control strategies to handle uncertainties of human driver’s behavior, it ensures cooperative platooning for mixed traffic safely and efficiently without requiring any prior knowledge of the human drivers, even at the early stage of CAV deployment. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.