Analysis of Traffic Flow Characteristics Under Freeway Debris Events Considering Hierarchical Autonomous Driving

Freeway debris events are sudden, randomly distributed, and impose severe safety risks, threatening both the stability and efficiency of traffic systems. With the rapid advancement of connected and automated vehicle (CAV) technologies, understanding the role of hierarchical automation in such unexpected scenarios has become increasingly critical. This study develops a two-lane microscopic traffic flow model that incorporates multiple automation levels by integrating the Intelligent Driver Model (IDM), Adaptive Cruise Control (ACC), Cooperative Adaptive Cruise Control (CACC), and a CAR-ToC-based takeover framework. A debris event occurring in the left lane is simulated to evaluate the combined effects of autonomous vehicle penetration rate and lane-closure duration on traffic flow dynamics. Simulation results demonstrate that higher penetration of advanced autonomous vehicles significantly improves traffic resilience by facilitating earlier and smoother lane-changing maneuvers, reducing congestion propagation, and mitigating instability. By contrast, flows dominated by lower-level vehicles exhibit stronger congestion waves and weaker recovery ability. Moreover, lane-closure duration proves to be a decisive factor for system performance, as prolonged closures greatly intensify congestion and reduce overall roadway capacity.