2026-09-08

7 arXiv categories· 96 journals· 2032 candidates — 8 worth your time· 283 without an open abstract

Headline

A multi-level safety-augmented calibration framework for car-following models

Today, together

tag shift
no tag ran above its 30-day average
shares references
no items share references today counted over the 8 papers published above
on the same shoulders
6 of the day's papers cite Travel demand and the 3Ds: Density, diversity, and design (Robert Cervero, Kara M. Kockelman 1997); 5 cite Accessibility evaluation of land-use and transport strategies: review and research directions (Karst Geurs, Bert van Wee 2003); 5 cite Travel and the Built Environment (Reid Ewing, Robert Cervero 2010) counted over the 128 of this day's 291 named papers whose reference lists we have — standing on the same older work is not the same as being about the same thing
first appearance
Hebei University of Engineering; Institutes of Science and Development; Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application; and 3 more none of them on a paper we published in the 88 days before this issue

Two counts, not one counted twice: every row here but “on the same shoulders” counts the papers published above, and that one counts every paper this day named — those and the ones under Also published today, whose abstracts we could not read.

The first paper developed a multi-level safety-augmented calibration framework for car-following models to integrate kinematic alignment with regime and risk-level constraints. A second study presented a reproducible multi-sensor workflow integrating photogrammetry, SLAM-based mapping, and geodetic control to document complex cultural heritage monuments. Finally, researchers used clustering algorithms on multi-source geographical big data to delineate and classify 67 commercial districts within Beijing's 5th Ring Road.

codedatapublished: Transportation Research Part C Emerging Technologies

Pengcheng Fan, Zuduo Zheng, Tong Chen
Transportation Research Part C Emerging Technologies

The University of Queensland

Traffic control and management · Traffic and Road Safety · Autonomous Vehicle Technology and Safety · The University of Queensland

Beyond kinematic error: a multi-level safety-augmented calibration framework for car-following models

The researchers developed the Multi-level Safety-augmented Car-Following Calibration (MS-CFC) framework, which integrates kinematic alignment with regime and risk-level constraints to calibrate car-following models. The framework was validated using synthetic trajectories with ground-truth parameters, high-fidelity driving simulator data, and real-world vehicle trajectories.

Why it matters — It prevents calibration algorithms from selecting physically unrealistic parameter values that happen to minimize kinematic error through error compensation, making car-following models reliable for safety-critical traffic safety simulations even when using short or incomplete trajectory data.

Transportation Research Part C Emerging Technologies · doi · code

datapublished: ISPRS International Journal of Geo-Information

Styliani Verykokou, Konstantinos Nikolitsas, George Piniotis, et al.
ISPRS International Journal of Geo-Information

National Technical University of Athens

3D Surveying and Cultural Heritage · Remote Sensing and LiDAR Applications · Robotics and Sensor-Based Localization · National Technical University of Athens

Integrating Photogrammetry and SLAM for the 3D Geometric Documentation of Cultural Heritage Monuments: A Reproducible Multi-Sensor Workflow Supported by an Open Dataset

This study presents a reproducible multi-sensor workflow that integrates UAV and close-range photogrammetry, SLAM-based mapping, and geodetic control within a single reference system to document complex cultural heritage monuments. Tested on a historic monastery, both photogrammetry and SLAM datasets achieved centimeter-level agreement with geodetic observations, with photogrammetry providing superior exterior coverage and textures, while SLAM captured interior spaces rapidly.

Why it matters — It establishes a reliable, integrated methodology that overcomes the individual limitations of photogrammetry and SLAM, providing a complete and metrically accurate 3D representation of complex structures for conservation and monitoring.

Caveat: The workflow was demonstrated and validated on a single historic monastery, which may not capture the full range of environmental and structural challenges found in other heritage sites.

ISPRS International Journal of Geo-Information · doi

published: Urban Informatics

Zonghan Yang, Ci Song, Peiqi Wang, et al.
Urban Informatics

Institute of Geographic Sciences and Natural Resources Research

Human Mobility and Location-Based Analysis · Smart Cities and Technologies · Global Urban Networks and Dynamics · Beijing · Institute of Geographic Sciences and Natural Resources Research · University of Chinese Academy of Sciences

The characteristic analysis and pattern classification of Beijing’s commercial districts based on multi-source geographical big data

Using clustering algorithms on multi-source geographical big data, this study delineated 67 commercial districts within Beijing's 5th Ring Road. The districts were evaluated across four dimensions—scale and composition, vitality and temporal heterogeneity, radiation, and location and environment—and categorized into seven distinct operational patterns, such as City-level Core, Suburban Hub, and Weekday-oriented Local.

Why it matters — It replaces simplified classification schemes with a multi-dimensional, data-driven taxonomy that captures how commercial districts actually function across different times of the week, enabling planners to tailor revitalization strategies to specific district profiles.

Caveat: The classification system and empirical findings are restricted to the area within Beijing's 5th Ring Road and may not directly generalize to other metropolitan contexts.

Urban Informatics · doi

preprint

Om Chiddarwar, Priyanka Mandal, Praveen Kumar Chandaliya, et al.
arXiv cs.AI

Air Quality Monitoring and Forecasting · Traffic Prediction and Management Techniques · Air Quality and Health Impacts · graph neural network

Predicting Spatiotemporal Mobile Sensing-Based PM2.5 Concentrations Using Low-Rank Adapted Spatially Attentive Graph Neural Network

The researchers developed a Spatially Attentive Graph Neural Network (SA-GNN) to forecast short-term PM2.5 concentrations using a new mobile-sensing dataset from Surat, India. The model processes meteorological and land-use data mapped to spatial nodes defined via uniform 200-400 meter segments and DBSCAN clustering. The SA-GNN achieved an R2 of 0.95, an RMSE of 6.8, and an MAE of 4.2 micrograms per cubic meter, outperforming LSTM, RNN, GRU, and ANN baselines.

Why it matters — It provides a method to capture rapid, localized air quality fluctuations along transit corridors that standard deep learning models fail to resolve, enabling real-time hotspot identification and personalized exposure tracking.

Caveat: The model's high performance was demonstrated using a dataset from a single city, which may limit immediate generalizability to urban environments with different spatial layouts or meteorological patterns.

arXiv · pdf

preprint

Giulia Zarpellon, Joseph Aylett-Bullock, Frank Krauss, et al.
arXiv physics.soc-ph

Evacuation and Crowd Dynamics · COVID-19 epidemiological studies · Facility Location and Emergency Management · simulation

Mitigating Disease Spread by Design in Refugee and IDP Camps

This study presents a methodology to evaluate how different refugee and internally displaced person camp layouts influence human movement and disease transmission. The approach is demonstrated through proof-of-concept experiments using the JUNE agent-based epidemic model to simulate interaction patterns and disease spread within camp boundaries.

Why it matters — It establishes a framework for camp planners to use spatial design and layout configuration as active tools for mitigating epidemic outbreaks in highly vulnerable settlements.

Caveat: The methodology is presented only as a proof-of-concept simulation, and the abstract does not validate the model against real-world outbreak data from an active camp.

arXiv · pdf

preprint

Siddharth Patwardhan, Şirag Erkol, Filippo Radicchi, et al.
arXiv physics.soc-ph

Boston · Paris

Crowding controls the scaling of bus frequency with demand

By analyzing approximately 3,000 bus routes carrying over 4 billion annual riders across 19 global metropolitan areas, this study identifies a universal scaling law where bus service frequency scales with passenger demand and route duration to an exponent between 1/2 and 2/3. The research demonstrates that this relationship is driven by cities minimizing passenger wait times under fixed budgets, transitioning from an exponent of 1/2 in uncongested networks to 2/3 when routes are highly crowded.

Why it matters — It establishes a predictive, optimization-based framework for transit resource allocation, demonstrating that the return on budget increases is highly unequal across cities; for example, a 20% budget expansion reduces daily passenger wait times by nearly 5 minutes in Boston but only about 1 minute in Paris.

arXiv · pdf · doi

preprint

Laura Fraade-Blanar, Jackson Graves, Aaron Clark-Ginsberg, et al.
arXiv cs.CY

Waymo, LLC

Human-Automation Interaction and Safety · Occupational Health and Safety Research · Risk and Safety Analysis · LEK Consulting (United States)

On Being Prepared: Automated Vehicle Incident Management Exercise Practices

This paper adapts established incident management practices into a series of company-neutral frameworks specifically tailored for automated vehicle (AV) emergency exercises. It details how to design, facilitate, and assess different exercise formats—including workshops, tabletop simulations, drills, and full-scale exercises—to address AV-specific socio-technical challenges ranging from collisions to natural hazards.

Why it matters — It provides the first structured guidance for AV incident response training, establishing a standardized methodology that allows different AV companies and public safety organizations to collaborate on emergency readiness.

Caveat: The paper presents conceptual frameworks and best practices rather than empirical data or case studies evaluating the real-world performance of these exercises.

arXiv · pdf

preprint

Astrid N. Hoefler, Robert B. Lund, J. Elliott Campbell
arXiv stat.AP

Quantifying Impacts of Wildfire Smoke on Crop Yields in California

This study analyzed the impact of wildfire smoke on wine grape yields at three sites in Northern California using statistical regression models. Incorporating particulate matter alongside temperature as predictor variables yielded models with R-squared values of 0.79 for Sonoma and 0.66 for Napa, demonstrating a significant association between smoke-induced air quality degradation and reduced crop yields.

Why it matters — It establishes that agricultural productivity models in wildfire-prone regions can no longer rely solely on climate variables, proving that particulate matter is now a critical factor in predicting crop yields.

Caveat: The findings are based on only three study sites in Northern California, and the statistical association was only significant for two of those three locations.

arXiv · pdf

Still cited

Martin Treiber, Ansgar Hennecke (2000), Congested traffic states in empirical observations and microscopic simulations 1 of today's items cite it · 22 of 3240 in the archive stand on it

Also published today

These appeared today in journals we track. Their abstracts are not openly available, so we cannot summarise them.