Optimal Control of Traffic Signals using Quantum Annealing

TL;DR

This paper formulates the optimal control of traffic signals as a QUBO problem and employs a hybrid quantum-classical approach using D-Wave's quantum annealer to improve traffic flow on a grid network.

Contribution

It presents a novel QUBO formulation for traffic signal control and demonstrates a hybrid quantum-classical solution approach with periodic problem solving.

Findings

Successful formulation of traffic control as QUBO

Implementation of hybrid quantum-classical solution workflow

Potential for improved traffic flow management

Abstract

Quadratic unconstrained binary optimization (QUBO) is the mathematical formalism for phrasing and solving a class of optimization problems that are combinatorial in nature. Due to their natural equivalence with the two dimensional Ising model for ferromagnetism in statistical mechanics, problems from the QUBO class can be solved on quantum annealing hardware. In this paper, we report a QUBO formatting of the problem of optimal control of time-dependent traffic signals on an artificial grid-structured road network so as to ease the flow of traffic, and the use of D-Wave Systems' quantum annealer to solve it. Since current-generation D-Wave annealers have a limited number of qubits and limited inter-qubit connectivity, we adopt a hybrid (classical/quantum) approach to this problem. As traffic flow is a continuous and evolving phenomenon, we address this time-dependent problem by adopting a workflow to generate and solve multiple problem instances periodically.