Traffic flow optimization using a quantum annealer (1708.01625v2)

Abstract: Quantum annealing algorithms belong to the class of meta-heuristic tools, applicable for solving binary optimization problems. Hardware implementations of quantum annealing, such as the quantum processing units (QPUs) produced by D-Wave Systems, have been subject to multiple analyses in research, with the aim of characterizing the technology's usefulness for optimization and sampling tasks. In this paper, we present a real-world application that uses quantum technologies. Specifically, we show how to map certain parts of the real-world traffic flow optimization problem to be suitable for quantum annealing. We show that time-critical optimization tasks, such as continuous redistribution of position data for cars in dense road networks, are suitable candidates for quantum applications. Due to the limited size and connectivity of current-generation D-Wave QPUs, we use a hybrid quantum and classical approach to solve the traffic flow problem.

• The paper presents a novel hybrid approach that integrates quantum annealing with classical computing to optimize traffic routing in urban environments.
• It formulates the traffic congestion problem as a QUBO using real GPS data and leverages D-Wave’s QPU with qbsolv to address routing challenges.
• The hybrid method achieves significant reductions in computation time and congestion compared to random route assignments.

Traffic Flow Optimization Using a Quantum Annealer

The paper presents a novel hybrid approach utilizing quantum annealing technologies for optimizing traffic flow in dense urban environments. The paper employs D-Wave Systems' Quantum Processing Units (QPUs), integrating them with classical computing resources to tackle real-world traffic flow challenges. The research aims to minimize time delays by alleviating congestion, a classic combinatorial optimization problem, using both quantum and classical computational techniques.

Problem Formulation and Methodology

The paper's core objective is to minimize congestion in a road network, particularly around Beijing, by redirecting vehicle routes. Leveraging the publicly available T-Drive dataset, which details GPS trajectories for over ten thousand taxis, the paper presents an approach that transforms this traffic optimization task into a Quadratic Unconstrained Binary Optimization (QUBO) problem.

To achieve this, the traffic flow problem is decomposed into several stages:

1. Data Preparation: Pre-processing of the map and GPS data to establish routes and areas of congestion.
2. Congestion Identification: Identifying traffic bottlenecks through classical analysis.
3. Alternative Routing: Developing and evaluating spatially and temporally feasible alternative paths for vehicles.
4. QUBO Formulation: Transforming the alternative route problem into a QUBO to be processed by the quantum computer.
5. Hybrid Optimization: Using D-Wave's hybrid classical/quantum software tools to mitigate congestion effectively.
6. Route Redistribution: Reassigning vehicles based on optimized results to minimize congestion iteratively.

The use of the OSMnx library enabled the integration and analysis of street network data, facilitating the segmentation and alternation routing processes.

Implementation and Results

The research employed the D-Wave 2X QPU, noted for its Chimera topology with 1152 qubits, using this quantum hardware in conjunction with classical computation to address traffic optimization for 418 selected vehicles. Due to connectivity constraints on the QPU, the paper utilizes the qbsolv software to decompose and address larger QUBO problems iteratively. The hybrid approach exhibits the potential to circumvent latency issues associated with public access to shared QPU resources, achieving computation times as low as 22 seconds in optimal conditions.

Quantitative evaluation of this hybrid approach indicates a significant improvement over random route assignments, with all experiments achieving congestion resolution. The paper underscores that future enhancements in QPU capabilities should further improve solution quality and computational efficiency for complex, real-world applications.

Implications and Future Work

The practical implications of this paper lie in its feasibility to deliver near-instantaneous adaptive traffic management solutions, potentially reducing urban congestion and enhancing vehicular flow efficiency. Theoretically, the research supports the proposition that real-world optimization problems can effectively leverage quantum annealing, given appropriate problem formulation and hybrid computational methodologies.

Future work aims to enrich the problem complexity by incorporating additional real-world parameters such as other traffic participants and infrastructure interactions. There is also an expressed intent to generalize and extend optimization queries to other domains where similar combinatorial challenges are prevalent.

Overall, this research demonstrates a substantial stride in applying emerging quantum computation technologies with classical computing paradigms, providing a promising outlook for quantum-based solutions in expansive urban traffic systems.