Optimal Traffic Flow in Quantum Annealing-Supported Virtual Traffic Lights

2024-12-22

arXiv.org

AI Summary

Traditional real-time traffic signal optimization struggles with stochastic vehicle arrivals at un-signalized intersections. Method: This paper proposes a Quantum Annealing–enabled Virtual Traffic Light (VTL) system that first formulates dynamic right-of-way allocation—under the NEMA standard—as a Quadratic Unconstrained Binary Optimization (QUBO) problem, solved optimally using a D-Wave quantum annealer to determine phase sequences and timing plans; it integrates SUMO-based simulation, the VTL protocol, and Signal Phase and Timing (SPaT) data generation to enable vehicle-infrastructure cooperative right-of-way assignment. Contribution/Results: Experiments across multi-level traffic volumes demonstrate significant reductions in vehicle stop delay and travel time versus classical optimization methods, without requiring physical traffic infrastructure. This work constitutes the first deployable validation of quantum computing in urban traffic control and establishes QUBO modeling coupled with quantum annealing as an effective paradigm for dynamic traffic scheduling.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Quantum Machine Learning
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Virtualization and resource management in Web systems and infrastructures
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Security and Privacy: Large-scale security measurements

Abstract

The Virtual Traffic Light (VTL) eliminates the need for physical traffic signal infrastructure at intersections, leveraging Connected Vehicles (CVs) to optimize traffic flow. VTL assigns right-of-way dynamically based on factors such as estimated times of arrival (ETAs), the number of CVs in various lanes, and emission rates. These factors are considered in line with the objectives of the VTL application. Aiming to optimize traffic flow and reduce delays, the VTL system generates Signal Phase and Timing (SPaT) data for CVs approaching an intersection, while considering the impact of each CV movement on others. However, the stochastic nature of vehicle arrivals at intersections complicates real-time optimization, challenging classical computing methods. To address this limitation, we develop a VTL method that leverages quantum computing to minimize stopped delays for CVs. The method formulates the VTL problem as a Quadratic Unconstrained Binary Optimization (QUBO) problem, a mathematical framework well-suited for quantum computing. Using D-Wave cloud-based quantum computer, our approach determines optimal solutions for right-of-way assignments under the standard National Electrical Manufacturers Association (NEMA) phasing system. The system was evaluated using the microscopic traffic simulator SUMO under varying traffic volumes. Our results demonstrate that the quantum-enabled VTL system reduces stopped delays and travel times compared to classical optimization-based systems. This approach not only enhances traffic management efficiency but also reduces the infrastructure costs associated with traditional traffic signals. The quantum computing-supported VTL system offers a transformative solution for large-scale traffic control, providing superior performance across diverse traffic scenarios and paving the way for advanced, cost-effective traffic management.

Problem

• Optimizing traffic flow by reducing delays at intersections
• Using quantum computing to solve complex VTL optimization problems
• Minimizing stopped delays for connected vehicles via QUBO

Innovation

• Quantum annealing optimizes traffic light timing
• QUBO framework minimizes vehicle stopped delays
• Quantum-in-the-loop simulation with SUMO and D-Wave