Connected Vehicle-Based Traffic Signal Coordination

Li, Wan; Ban, Xuegang · 2020 · Crossref

DOI: 10.1016/j.eng.2020.10.009

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 addresses the challenge of optimizing traffic signal coordination for connected vehicle (CV) environments on arterial corridors. Traditional signal coordination methods often rely on fixed parameters or bandwidth maximization, which may not fully leverage real-time data from CVs. The authors aim to minimize total system costs, defined as a combination of fuel consumption and travel time for individual vehicles, by simultaneously optimizing phase durations and offsets. To address the computational complexity of solving this problem as a centralized mixed-integer nonlinear program (MINLP), the researchers decompose the optimization into a two-level framework: an intersection level for phase durations and a corridor level for offsets. The methodology formulates the signal coordination problem as an MINLP that accounts for individual vehicle trajectories, including second-by-second speed and location data obtained via vehicle-to-infrastructure (V2I) and vehicle-to-vehicle (V2V) communications. The objective function minimizes the sum of monetary values assigned to fuel consumption and travel time across all vehicles in the corridor. To solve this large-scale problem, the authors propose a prediction-based iterative solution. At the intersection level, dynamic programming (DP) is used to determine optimal phase durations for each intersection, ensuring they sum to a fixed common cycle length. At the corridor level, the model iteratively calculates "temporary" offsets to coordinate traffic flow between adjacent intersections, acknowledging the signal status and vehicle states of neighboring intersections. Vehicle trajectories are predicted using an Intelligent Driving Model (IDM), where traffic signals are modeled as "virtual vehicles" to influence driver behavior. The proposed models were tested using traffic simulations under various scenarios, comparing them against traditional actuated signal timing plans. The results indicate that both the centralized MINLP and the decomposed two-level model improve signal control performance. Specifically, under high demand levels with varied vehicle types, the two-level model reduced total system cost by 3.8% compared to the baseline actuated plan, while the centralized MINLP achieved a 5.9% reduction. The study also found that coordination schemes are particularly beneficial for corridors with relatively high traffic demand. Additionally, the analysis revealed that coordinating signals for major streets had minimal impact on vehicles traveling on minor streets. The significance of this work lies in its demonstration that CV-based data can be effectively utilized to enhance traffic signal coordination through a computationally tractable two-level optimization approach. By decomposing the problem, the method offers a viable alternative to complex centralized models for real-time applications. The findings suggest that integrating individual vehicle trajectory data into signal control can yield measurable improvements in fuel efficiency and travel time, particularly in high-demand urban corridors. This approach advances the field by providing a structured framework for leveraging CV technologies to optimize multi-intersection 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.

• v2x connected vehicle