Ultra-fast Traffic Nowcasting and Control via Differentiable Agent-based Simulation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Makinoshima, Fumiyasu, Yamaguchi, Yuya, Segawa, Eigo, Niinuma, Koichiro, Qian, Sean
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915891830063104
author Makinoshima, Fumiyasu
Yamaguchi, Yuya
Segawa, Eigo
Niinuma, Koichiro
Qian, Sean
author_facet Makinoshima, Fumiyasu
Yamaguchi, Yuya
Segawa, Eigo
Niinuma, Koichiro
Qian, Sean
contents Traffic digital twins, which inform policymakers of effective interventions based on large-scale, high-fidelity computational models calibrated to real-world traffic, hold promise for addressing societal challenges in our rapidly urbanizing world. However, conventional fine-grained traffic simulations are non-differentiable and typically rely on inefficient gradient-free optimization, making calibration for real-world applications computationally infeasible. Here we present a differentiable agent-based traffic simulator that enables ultra-fast model calibration, traffic nowcasting, and control on large-scale networks. We develop several differentiable computing techniques for simulating individual vehicle movements, including stochastic decision-making and inter-agent interactions, while ensuring that entire simulation trajectories remain end-to-end differentiable for efficient gradient-based optimization. On the large-scale Chicago road network, with over 10,000 calibration parameters, our model simulates more than one million vehicles at 173 times real-time speed. This ultra-fast simulation, together with efficient gradient-based optimization, enables us to complete model calibration using the previous 30 minutes of traffic data in 455 s, provide a one-hour-ahead traffic nowcast in 21 s, and solve the resulting traffic control problem in 728 s. This yields a full calibration--nowcast--control loop in under 20 minutes, leaving about 40 minutes of lead time for implementing interventions. Our work thus provides a practical computational basis for realizing traffic digital twins.
format Preprint
id arxiv_https___arxiv_org_abs_2603_25068
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Ultra-fast Traffic Nowcasting and Control via Differentiable Agent-based Simulation
Makinoshima, Fumiyasu
Yamaguchi, Yuya
Segawa, Eigo
Niinuma, Koichiro
Qian, Sean
Multiagent Systems
Machine Learning
Traffic digital twins, which inform policymakers of effective interventions based on large-scale, high-fidelity computational models calibrated to real-world traffic, hold promise for addressing societal challenges in our rapidly urbanizing world. However, conventional fine-grained traffic simulations are non-differentiable and typically rely on inefficient gradient-free optimization, making calibration for real-world applications computationally infeasible. Here we present a differentiable agent-based traffic simulator that enables ultra-fast model calibration, traffic nowcasting, and control on large-scale networks. We develop several differentiable computing techniques for simulating individual vehicle movements, including stochastic decision-making and inter-agent interactions, while ensuring that entire simulation trajectories remain end-to-end differentiable for efficient gradient-based optimization. On the large-scale Chicago road network, with over 10,000 calibration parameters, our model simulates more than one million vehicles at 173 times real-time speed. This ultra-fast simulation, together with efficient gradient-based optimization, enables us to complete model calibration using the previous 30 minutes of traffic data in 455 s, provide a one-hour-ahead traffic nowcast in 21 s, and solve the resulting traffic control problem in 728 s. This yields a full calibration--nowcast--control loop in under 20 minutes, leaving about 40 minutes of lead time for implementing interventions. Our work thus provides a practical computational basis for realizing traffic digital twins.
title Ultra-fast Traffic Nowcasting and Control via Differentiable Agent-based Simulation
topic Multiagent Systems
Machine Learning
url https://arxiv.org/abs/2603.25068