Quantum Use Case #17: Traffic signal optimisation in cities

Short answer: This use case involves formulating the global optimization of traffic signal timing on urban road networks as a Quadratic Unconstrained Binary Optimization (QUBO) problem, which is then mapped to an Ising Hamiltonian for execution on quantum annealing hardware. Researchers have proposed and tested in simulation and limited hardware tests that this approach can minimize traffic flow imbalance more effectively than conventional local control methods, though the technology remains at the research stage and faces significant scalability challenges for real-world deployment.

Theoretical Foundation and Problem Formulation

The core challenge addressed by this quantum use case is the heavy computational load imposed by intelligent transportation systems in urban environments. As cities grow, the complexity of managing large-scale traffic increases, requiring novel architectures for control. Traditional methods often rely on local control, where each intersection operates independently or with limited coordination. However, global control, which considers the entire network simultaneously, offers the potential for superior efficiency but is computationally intensive.

To address this, researchers have developed methods for globally controlling traffic signals arranged on a square lattice. The fundamental step is the reformulation of the traffic signal optimization problem into a mathematical format compatible with quantum hardware. Specifically, the problem is cast as a Quadratic Unconstrained Binary Optimization (QUBO) problem. QUBO is a mathematical formalism used to phrase and solve a class of combinatorial optimization problems. Due to the natural equivalence between QUBO and the two-dimensional Ising model for ferromagnetism in statistical mechanics, these problems can be mapped to the energy landscape of a quantum annealer.

In the context of traffic, the objective function typically minimizes the imbalance of traffic flows in two orthogonal directions. By converting the traffic control variables into binary spins, the optimization problem becomes an Ising Hamiltonian. This Hamiltonian is then compatible with quantum annealers, such as the D-Wave system, which seeks the ground state of the Hamiltonian to find the optimal signal configuration. This transformation allows the complex, multi-variable traffic problem to be handled by hardware designed to explore energy landscapes efficiently.

Research Evidence and Simulation Results

Several studies have investigated the viability of this approach. One prominent line of research focuses on the application of quantum annealing to traffic signal optimization on a square lattice. In this work, the researchers formulated a signal optimization problem aimed at minimizing traffic flow imbalance. They then compared the new quantum-based control method against a conventional local control method. The comparison was conducted for a large-scale scenario, specifically a 50-by-50 city grid. The results indicated that the quantum annealing approach could outperform the conventional local control in managing the flow of traffic within this simulated environment.

Another study reported on the optimal control of time-dependent traffic signals on an artificial grid-structured road network. This work explicitly detailed the QUBO formatting of the problem and the subsequent use of D-Wave Systems’ quantum annealer. The goal was to ease the flow of traffic by finding optimal timing configurations. These studies collectively suggest that quantum annealing can provide solutions to traffic optimization problems that are difficult for classical algorithms to solve efficiently, particularly when the problem size grows.

Practical Usability and Real-Time Considerations

Beyond academic simulations, there is interest in the practical usability of quantum computing for real-time applications in traffic optimization and control. The DLR Quantum Computing Initiative has developed customized compilation strategies and tailor-made quantum algorithms for difficult, industry-relevant computational problems. It describes this work as showing practical usability for real-time applications, which is its own claim.

The efficient control of traffic lights has a significant influence on the quality of traffic flow. In large cities with a large number of such systems, this becomes a critical challenge, especially when coordination is required. Quantum computing is posited to make an important contribution to this optimization by handling the combinatorial complexity of the network. The focus here is not merely on theoretical superiority but on the ability to compile and execute these algorithms in a manner that could support real-time decision-making, although current implementations are largely confined to research and demonstration phases.

Limitations and Research-Stage Constraints

Despite the promising results in simulation and limited hardware tests, several limitations remain. First, the scale of the problem is a critical factor. While a 50-by-50 grid has been simulated, real-world cities are vastly larger and more complex, with irregular road networks that do not conform to simple square lattices. Mapping such irregular topologies to the hardware constraints of a quantum annealer is a non-trivial task that may require significant approximation or decomposition.

Second, the current research is primarily based on simulations and controlled experiments with specific hardware configurations. There is no evidence of commercial deployment or operational adoption in live city traffic systems. The claims of "practical usability" refer to the technical feasibility of running the algorithms on existing quantum hardware in a research context, not to widespread integration into municipal infrastructure.

Third, the comparison with conventional methods is often limited to specific metrics, such as flow imbalance, and specific grid structures. Real-world traffic optimization involves multiple conflicting objectives, including pedestrian safety, emergency vehicle priority, and environmental impact, which may not be fully captured by the simplified Ising models used in these studies.

Finally, the speed and reliability of quantum annealing for real-time control are still under investigation. While the algorithms are "tailor-made" for industry-relevant problems, the latency and error rates of current quantum hardware must be rigorously evaluated against the strict time constraints of traffic signal control. Until these challenges are resolved, the use case remains a research-stage opportunity rather than an operational reality.

FAQ

Is this technology currently deployed in cities?

No. The research indicates that this is a research-stage application. Studies have shown results in simulation and on specific quantum annealing hardware, but there are no claims of commercial deployment or operational adoption in live urban traffic systems.

How does the quantum approach differ from traditional control?

Traditional methods often use local control, where intersections act independently. The quantum approach formulates the problem as a global optimization task, mapping it to an Ising Hamiltonian to be solved by a quantum annealer. This allows for coordinated control across the entire network, aiming to minimize flow imbalance globally rather than locally.

What hardware is used for these experiments?

The research excerpts specifically mention the use of D-Wave quantum annealers. These devices are designed to solve QUBO and Ising model problems by exploring the energy landscape of the system. Customized compilation strategies are developed to map traffic problems to this hardware.

Can this handle real-time traffic demands?

Researchers have described the "practical usability" of quantum computing for real-time applications in a research context. However, this refers to the technical capability of running the algorithms on existing hardware, not to proven reliability in live, high-stakes urban environments. Real-time latency and error rates remain areas of active investigation.

What are the main obstacles to wider adoption?

The primary obstacles include the scalability of the problem to real-world city sizes, the complexity of mapping irregular road networks to the hardware's constraints, and the lack of operational deployment. Additionally, the current models may not fully capture the multi-objective nature of real-world traffic management, such as safety and environmental concerns.