A Cooperative Control Framework for CAV Lane Change in a Mixed Traffic Environment

Abstract

In preparing for connected and autonomous vehicles (CAVs), a worrisome aspect is the transition era which will be characterized by mixed traffic (where CAVs and human-driven vehicles (HDVs) share the roadway). Consistent with expectations that CAVs will improve road safety, on-road CAVs may adopt rather conservative control policies, and this will likely cause HDVs to unduly exploit CAV conservativeness by driving in ways that imperil safety. A context of this situation is lane-changing by the CAV. Without cooperation from other vehicles in the traffic stream, it can be extremely unsafe for the CAV to change lanes under dense, high-speed traffic conditions. The cooperation of neighboring vehicles is indispensable. To address this issue, this paper develops a control framework where connected HDVs and CAV can cooperate to facilitate safe and efficient lane changing by the CAV. Throughout the lane-change process, the safety of not only the CAV but also of all neighboring vehicles, is ensured through a collision avoidance mechanism in the control framework. The overall traffic flow efficiency is analyzed in terms of the ambient level of CHDV-CAV cooperation. The analysis outcomes are including the CAVs lane-change feasibility, the overall duration of the lane change. Lane change is a major source of traffic disturbance at multi-lane highways that impair their traffic flow efficiency. In providing a control framework for lane change in mixed traffic, this study shows how CHDV-CAV cooperation could help enhancing system efficiency.

arXiv commentTRB 2021 Annual Meeting

References in corpus (3)

• A DRL-based Multiagent Cooperative Control Framework for CAV Networks: a Graphic Convolution Q Network
• Leveraging the Capabilities of Connected and Autonomous Vehicles and Multi-Agent Reinforcement Learning to Mitigate Highway Bottleneck Congestion
• Using Empirical Trajectory Data to Design Connected Autonomous Vehicle Controllers for Traffic Stabilization

Cited by in corpus (5)

• Multi-agent Reinforcement Learning for Cooperative Lane Changing of Connected and Autonomous Vehicles in Mixed Traffic
• Leveraging the Capabilities of Connected and Autonomous Vehicles and Multi-Agent Reinforcement Learning to Mitigate Highway Bottleneck Congestion
• Facilitating Connected Autonomous Vehicle Operations Using Space-weighted Information Fusion and Deep Reinforcement Learning Based Control
• Using Empirical Trajectory Data to Design Connected Autonomous Vehicle Controllers for Traffic Stabilization
• Urban traffic dynamic rerouting framework: A DRL-based model with fog-cloud architecture