Longitudinal Influence of Autonomous Vehicles and Vehicular Communication on Post-Accident Traffic

Abstract

The primary objective of this study is to quantify the role of autonomous vehicles (AVs) and vehicular communication on post-accident traffic flow. We propose an analytical framework embedding car-following models for safety analysis for mixed traffic flow of human-driven vehicles and AVs, especially under vehicular communication. We then calibrate car-following models with real-world data. Through the model-based simulation experiments, we examine and discuss in details traffic flow stability, throughput, and post-accident safety. The results show that the traffic congestion resulting from the primary accident lasted shorter with the growing penetration of AVs. It is appealing that relaxing the safety grade of AVs can increase the throughput and this enables us to balance the safety level of AV operation and traffic throughput. In addition, incorporating vehicular communication can alleviate the risk of relaxing safety requirement and strikingly reduce the number and the severity of the secondary accidents.