Deep learning cuts travel times and improves urban traffic control
Learning to Move Cities: Deep Meta-Models and Reinforcement Policies for Calibration and Control in Urban Networks
Machine Learning
Summary
City traffic is really complicated, making it hard to adjust systems quickly or set up accurate simulations. The authors created a way to simplify the complex traffic data into a smaller, useful set that helps both tune simulations and make better traffic control decisions with AI. Their method reduced travel times by almost half in tests on standard traffic networks. This approach links understanding city traffic patterns with controlling them in real time, making urban travel smoother.
What this means in practice
- •For urban traffic planners: Improve traffic simulation calibration and reduce city-wide travel times by integrating compressed traffic models into operational control.
- •For intelligent transportation system developers: Deploy AI agents using learned traffic representations for real-time routing and scheduling adjustments in large-scale urban networks.
Authors
Adewumi Augustine Adepitan, Christopher J. Haruna, Oluwasegun Adegoke, Ayooluwatomiwa Ajiboye, Oluwatobi Oluwasakin
Abstract
Urban transportation networks present complex optimization challenges spanning calibration of high-fidelity simulators and real-time operational control. This paper presents a shared latent-space framework that connects simulator calibration and reinforcement learning control through a common learned representation of urban traffic dynamics. First, we develop a combinatorial MLP-autoencoder architecture that learns low-dimensional manifolds linking simulator inputs (origin-destination demand, network parameters) to outputs (travel times, congestion patterns), enabling efficient Bayesian optimization for calibration. This approach demonstrates superior sample efficiency compared to traditional dimension reduction methods, achieving better fit to observational data within fixed computational budgets. Second, we implement a deep Q-learning agent with experience replay and target networks to optimize dynamic traffic assignment through scheduling and routing adjustments. In empirical evaluations on benchmark networks, our approach reduces system-wide travel times by up to 51% compared to baseline operations. The learned latent representation is not only used to reduce the dimensionality of Bayesian calibration, but is also incorporated into the reinforcement learning state representation, allowing the control policy to operate on compressed and calibrated traffic dynamics. This shared latent-space formulation provides a unified pathway from simulator calibration to adaptive operational control within intelligent transportation systems. Our results highlight the transformative potential of deep learning methods in urban mobility planning and management, particularly for large-scale networks where traditional optimization approaches face computational bottlenecks.