ITS Berkeley

DeepAir: Deep Learning and Satellite Imagery to Estimate High-resolution PM2.5 at Scale

Guo, Wenxuan
Hu, Zhaoping
Jin, Ling
Xu, Yanyan
González, Marta C.
2025

Air pollution, specifically PM2.5, has become a significant global concern owing to its detrimental impacts on public health. Even so, the high-resolution monitoring of air pollution is still a challenge on a global scale. To cope with this, machine learning (ML) techniques have been utilized to infer the concentration of air pollutants at a fine scale. In this study, we propose DeepAir, a learning framework for estimating PM2.5 concentrations at a fine scale with sparsely distributed observations. DeepAir integrates a pre-trained convolutional neural network with the LightGBM method. This...

Modeling Potential Fire Spread Polygons and Networks for Suppression Strategies

Kim, Minho
Castellnou, Marc
González, Marta C.
2025

Unprecedented fire seasons are overwhelming fire suppression capacity in Mediterranean Europe. Fire services respond to urgent risks, but are being outpaced by more complex wildfires. Fire suppression needs proactive and risk-informed strategies to avoid catastrophic wildfires. In this study, we present an automatic method for generating potential fire polygons by adapting hydrological basin delineation techniques to fire spread simulations. Using the time since ignition as input, we segment the landscape into polygons representing discrete spatial units of fire potential. These polygons...

A Computational Framework for Quantifying Route Diversification in Road Networks

Cornacchia, Giuliano
Pappalardo, Luca
Nanni, Micro
Pedreschi, Dino
González, Marta C.
2025

The structure of road networks impacts various urban dynamics, from traffic congestion to environmental sustainability and access to essential services. Recent studies reveal that most roads are underutilized, faster alternative routes are often overlooked, and traffic is typically concentrated on a few corridors. In this article, we examine how road network structure, and in particular the presence of mobility attractors (e.g., highways and ring roads), shapes the counterpart to traffic concentration: route diversification. To this end, we introduce DiverCity, a measure that quantifies...

Pipeline Distribuído para Análise Espacial em Larga Escala: Avaliação da Regra 3 do Índice 3-30-300 em Fortaleza com Apache Spark e Sedona

Silva, Lucas L.
González, Marta C.
Babadopulos, Lucas F. A.
Soares, Jorge B.
Furtado, Lara S.
2025

A aplicação de operações espaciais em grandes conjuntos de dados enfrenta limitações nas ferramentas tradicionais de geoprocessamento e em bibliotecas como Geopandas. Este trabalho apresenta um pipeline distribuído baseado em Apache Spark e Sedona para analisar dados geolocalizados de edificações e arborização, em Fortaleza, CE, identificando residências com visibilidade mínima de 3 árvores em 30 metros. O processamento em batches, a indexação espacial e a persistência estruturada permitiram superar gargalos do Python. Apresenta-se uma documentação detalhada de código modular que permite...

Optimal Control of ODE Car-Following Models: Applications to Mixed-Autonomy Platoon Control via Coupled Autonomous Vehicles

Alanqary, Arwa
Bayen, Alexandre M.
Gong, Xiaoqian
Gollakota, Anish
Keimer, Alexander
Pandian, Ashish
2025

In this paper, we study the optimal control of a mixed-autonomy platoon driving on a single lane to smooth traffic flow. The platoon consists of autonomous vehicles, whose acceleration is controlled, and human-driven vehicles, whose behavior is described using a microscopic car-following model. We formulate the optimal control problem where the dynamics of the platoon are describing through a system of non-linear ODEs, with explicit constraints on both the state and the control variables. Theoretically, we analyze the well-posedness of the system dynamics under a reasonable set of...

Operational Air Taxi Flight Routes in a Metropolitan Region

Kam, Jordan K.
Casanova, Matthias
Bulusu, Vishwanath
Bayen, Alexandre
Sengupta, Raja
2025

Aviation technology advances in Urban Air Mobility (UAM) will bring new aircraft to the National Airspace System (NAS) and with it near-term operational barriers: traffic density, Class B/C overlap, and pilot-controller workload. This article presents the design and evaluation of four operational air taxi flight routes for manned flight of electric vertical takeoff and landing (eVTOL) aircraft in the San Francisco Bay Area network (SF network). In the SF network, where many complex, congested, and controlled airspaces restrict flight, pilot-selected operational flight routes utilizing...

A Tutorial on Neural Network-Based Solvers for Hyperbolic Conservation Laws: Supervised vs. Unsupervised Learning, and Applications to Traffic Modeling

Canesse, Alexi
Fu, Zhe
Lichtle, Nathan
Matin, Hossein Nick Zinat
Liu, Zhe
Monache, Maria Laura Delle
Bayen, Alexandre M.
2025

Neural networks (NNs) are powerful tools for solving complex partial differential equations (PDEs) with high accuracy. However, many NN-based solvers are designed as general-purpose models or lack theoretical grounding, limiting their ability to capture essential solution properties such as regularity, conservation, and entropy conditions. This issue is especially critical for hyperbolic conservation laws, which govern wave propagation and shock formation, and are among the most challenging PDEs to solve accurately. This tutorial examines both supervised and unsupervised NN-based solvers...

(U)NFV: Supervised and Unsupervised Neural Finite Volume Methods for Solving Hyperbolic PDEs

Lichtle, Nathan
Canesse, Alexi
Fu, Zhe
Matin, Hossein Nick Zinat
Monache, Maria Laura Delle
Bayen, Alexandre M.
2025

We introduce (U)NFV, a modular neural network architecture that generalizes classical finite volume (FV) methods for solving hyperbolic conservation laws. Hyperbolic partial differential equations (PDEs) are challenging to solve, particularly conservation laws whose physically relevant solutions contain shocks and discontinuities. FV methods are widely used for their mathematical properties: convergence to entropy solutions, flow conservation, or total variation diminishing, but often lack accuracy and flexibility in complex settings. Neural Finite Volume addresses these limitations by...

Enabling Analysis and Visualization of Transportation Big Data

Rees, Stephen
Sprinkle, Jonathan
Wang, Xia
Bunting, Matthew
Work, Daniel B.
Lee, Jonathan W.
Monache, Maria Laura Delle
Bayen, Alexandre M.
Piccoli, Benedetto
2025

Transportation studies generate massive amounts of data that are difficult to store, process, query and visualize quickly and easily. Overcoming these challenges are an essential aspect of making the collected data useful to both the original study and other research that could build on the results. We explore the impact of database implementation, specifically IoTDB, on these aspects of data management with respect to transportation on existing datasets.

Human-In-The-Loop Classification of Adaptive Cruise Control at a Freeway Scale

Wang, Xia
Nice, Matthew
Bunting, Matt
Wu, Fangyu
Monache, Maria Laura Delle
Lee, Jonathan W.
Piccoli, Benedetto
Seibold, Benjamin
Bayen, Alexandre M.
Work, Daniel B.
Sprinkle, Jonathan
2025

The goal of this paper is to estimate whether a human or Adaptive Cruise Control (ACC) is managing a vehicle's speed control, based on observations by external sensors. The driving characteristics of individual vehicles---whether human-driven or ACC-controlled---play a crucial role in shaping overall traffic flow. To enable advanced traffic control strategies tailored to specific vehicle behaviors, this paper introduces a time-series deep learning classifier that leverages multiple models, including One-Dimensional Convolutional Neural Networks (1D-CNN), Recurrent Neural Networks (RNN),...