Influence of CAV clustering strategies on mixed traffic flow characteristics: An analysis of vehicle trajectory data

Zhong, Zijia; Lee, Earl E.; Nejad, Mark; Lee, Joyoung · 2020 · OpenAlex-citations

DOI: 10.1016/j.trc.2020.102611

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study investigates the impact of Connected and Automated Vehicle (CAV) clustering strategies on human-driven vehicles (HVs) within mixed traffic conditions. While prior research has focused on network-level benefits of Cooperative Adaptive Cruise Control (CACC), this paper addresses the gap in understanding how CACC deployment affects HV safety and behavior. The authors evaluate two platoon formation strategies: ad hoc coordination, where CAVs form platoons randomly based on arrival, and local coordination, where CAVs actively maneuver to join existing platoons. The research aims to quantify impacts on HV lane changes, braking events, and traffic flow characteristics using high-resolution trajectory data. The methodology employs microscopic traffic simulation using the VISSIM software on a calibrated 8-km segment of Interstate 66 near Washington, D.C. Human driving behavior is modeled using the Wiedemann car-following model, calibrated with real-world data from INRIX, remote traffic microwave sensors, and video cameras. CAV longitudinal control utilizes the Enhanced Intelligent Driver Model (E-IDM), while lateral control for local coordination follows a gap-acceptance algorithm. The study tests market penetration rates (MPR) of 10%, 20%, 30%, and 40%, assuming a 30% increase in baseline traffic demand. Five simulation replications were conducted for each scenario to account for stochasticity, generating approximately seven million vehicle trajectory records per replication. Results indicate that local coordination consistently outperforms ad hoc coordination in terms of network throughput and productivity across all tested MPRs. However, local coordination significantly alters HV safety metrics. Two-sample Kolmogorov-Smirnov tests reveal that the distribution of hard braking events for HVs changes significantly under local coordination, with a drastic increase in braking probability between -6.5 m/s² and -3.5 m/s² even at 10% MPR. Both strategies increase the average lane change frequency for HVs, with a break-even point observed at 30% MPR. Lane change frequency follows a monotonically increasing pattern, reaching a peak of 5.48 per vehicle at 40% MPR. Additionally, analysis of HV interaction states shows that the time spent in the "approaching" state declines as CAV presence increases. The findings suggest that while local coordination enhances overall network efficiency, it induces greater turbulence for human drivers through increased lane changes and hard braking events compared to ad hoc coordination. The study highlights that CACC deployment, particularly with active clustering strategies, creates significant lateral interactions that affect HV driving behavior. These insights are critical for anticipating the near-term deployment of CACC in mixed traffic environments, indicating that operational strategies must balance network throughput gains against potential safety risks and behavioral disruptions for non-equipped vehicles.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• traffic density
• following distance