Physics of automated driving in framework of three-phase traffic theory

Kerner, Boris S. · 2018 · Crossref

DOI: 10.1103/physreve.97.042303

archive: archived pipeline: cataloged verified

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

Summary

This paper investigates the physical dynamics of autonomous driving vehicles (ADV) within the framework of the three-phase traffic theory, contrasting it with classical adaptive cruise control (ACC) models. The research is motivated by the need to enhance traffic capacity and stability in future mixed traffic flows, where human-driven and autonomous vehicles coexist. Classical ACC models rely on maintaining a fixed desired time headway to preceding vehicles, a behavior the authors argue is inconsistent with empirical driver data and prone to causing traffic breakdowns at bottlenecks. In contrast, the three-phase theory posits that drivers adapt their speed to preceding vehicles within a synchronization space gap without targeting a specific time headway, provided the gap remains above a safe minimum. The study employs microscopic stochastic simulations on a single-lane road with an on-ramp bottleneck to compare the classical ACC model against a new three-phase ACC (TPACC) model. The TPACC model defines acceleration based solely on relative speed when the space gap is within the synchronization range, rather than adjusting for a fixed time headway. The simulations utilize discrete time steps and identical parameters for both models, including inflow rates, maximum acceleration/deceleration limits, and vehicle lengths, to ensure a direct comparison of their dynamic behaviors. The findings reveal three distinct advantages of TPACC over classical ACC. First, TPACC eliminates string instability; while classical ACC platoons can amplify speed disturbances upstream of a bottleneck, TPACC vehicles ensure these disturbances decay. Second, TPACC generates considerably smaller speed disturbances at bottlenecks compared to classical ACC, even when the latter is tuned to be string-stable. Third, and most significantly, the presence of TPACC vehicles in mixed traffic reduces the probability of traffic breakdown at bottlenecks. Conversely, the simulations demonstrate that even a single classical ACC vehicle in a mixed flow can provoke traffic breakdown by forcing following vehicles to decelerate sharply, whereas TPACC vehicles maintain smoother flow dynamics. The significance of this work lies in its challenge to the standard assumption that autonomous vehicles should target a fixed time headway. By aligning autonomous driving logic with the empirical behavior of human drivers described in three-phase traffic theory, the TPACC model offers a robust solution for stabilizing mixed traffic flows. The results imply that adopting TPACC dynamics can prevent the degradation of traffic capacity often associated with the introduction of autonomous vehicles, thereby enhancing the overall efficiency and reliability of transportation networks.

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.

• following distance