Cooperative Schedule-Driven Intersection Control with Connected and Autonomous Vehicles (1907.01984v1)

Abstract: Recent work in decentralized, schedule-driven traffic control has demonstrated the ability to improve the efficiency of traffic flow in complex urban road networks. In this approach, a scheduling agent is associated with each intersection. Each agent senses the traffic approaching its intersection and in real-time constructs a schedule that minimizes the cumulative wait time of vehicles approaching the intersection over the current look-ahead horizon. In this paper, we propose a cooperative algorithm that utilizes both connected and autonomous vehicles (CAV) and schedule-driven traffic control to create better traffic flow in the city. The algorithm enables an intersection scheduling agent to adjust the arrival time of an approaching platoon through use of wireless communication to control the velocity of vehicles. The sequence of approaching platoons is thus shifted toward a new shape that has smaller cumulative delay. We demonstrate how this algorithm outperforms the original approach in a real-time traffic signal control problem.

• The paper introduces a cooperative algorithm that integrates schedule-driven control with CAV-based velocity adjustments to achieve up to 19% delay reduction.
• It uses real-time V2I communication to dynamically re-sequence and adjust vehicle platoons, optimizing traffic flow at intersections.
• Simulations with SUMO validate the approach, demonstrating robust performance across various traffic demands and partial CAV deployments.

Summary of "Cooperative Schedule-Driven Intersection Control with Connected and Autonomous Vehicles"

This paper explores a novel integration of connected and autonomous vehicle (CAV) technologies with schedule-driven intersection control to optimize urban traffic flow. The paper emphasizes this integration's efficiency in reducing vehicular delays by collaboratively manipulating vehicle velocities using real-time scheduling information obtained via vehicle-to-infrastructure (V2I) communication. The proposed cooperative algorithm significantly outperforms traditional and existing schedule-driven methods in varied traffic conditions, particularly under high congestion scenarios, facilitating substantial improvements in traffic management and control.

Introduction

Traffic congestion remains a critical challenge, intensifying due to urbanization and increased vehicle usage. Existing centralized and decentralized approaches to traffic signal optimization have demonstrated improvements by leveraging online and distributed planning systems. A schedule-driven approach, which efficiently sequences spatially proximate vehicle clusters through intersections, organizes traffic flows into clustered sequences based on arrival predictions.

Recent advancements in CAV technologies, including V2I communication, offer enhancements in managing traffic flows through responsive vehicle control. The paper proposes utilizing these technologies to enable scheduling agents at intersections to directly influence vehicle velocities, thereby reshaping upcoming vehicle platoons to minimize overall delay. This cooperative approach bridges the gap between traditional signal-based control architectures and direct vehicle interaction, offering refined traffic solutions adaptable to real-time conditions.

Algorithmic Approach

The cooperative algorithm builds upon existing schedule-driven traffic control infrastructures, extending their capabilities via CAV interaction. Scheduling agents generate real-time signal timing plans, using cluster sequence information to compute optimal vehicular velocities. Vehicles, upon receiving scheduled timing information, adjust speeds accordingly, thus modifying platoons to reduce cumulative traffic delay.

Key operations include:

1. Re-sequencing Vehicles: Adjust vehicle sequences to optimize cumulative delay.
2. Velocity Adjustment: Control approach vehicle speeds based on real-time traffic conditions, facilitated by V2I mechanisms.

The paper exemplifies scenarios where the cooperative algorithm substantially reduces delays, particularly under heavy congestion, by reshaping vehicular platoons to bridge existing scheduling optimization and direct vehicular control.

Simulation and Results

Extensive simulations conducted via the Simulation of Urban MObility (SUMO) validate the effectiveness of the cooperative algorithm across diverse traffic conditions. The paper measures average delays in various scenarios, comparing them against schedule-driven and fixed-timing control methodologies.

Key findings include:

• Significant delay reductions under high traffic demand (up to 19% improvement over existing methods).
• Robust performance under varied penetration rates of CAV technologies, retaining substantial improvements even at partial deployment levels.

Implications and Future Research

The cooperative algorithm exemplifies a novel intersection of CAV technologies and traffic management systems, demonstrating practical benefits in congestion reduction. The implications extend to enhancing traffic safety, optimizing throughput, and fostering the deployment of smart transportation infrastructures.

Future advancements aim to refine global traffic coordination strategies, integrating network-wide scheduling and control to further optimize urban traffic scenarios. Continued research will explore interactions between mixed vehicle environments (autonomous and non-autonomous) and further quantification of algorithm scalability in larger urban settings.

Conclusion

The paper successfully demonstrates a cooperative framework that leverages schedule-driven traffic control in tandem with CAVs to optimize urban intersections. By facilitating direct vehicular control via real-time communication, the approach achieves notable improvements in traffic efficiency across various demand levels. The paper calls for further exploration into comprehensive network-wide applications, harnessing the burgeoning capabilities of connected and autonomous vehicle technologies.