<rss version="2.0" xmlns:atom="https://www.w3.org/2005/Atom">
  <channel>
    <title>Transport Research International Documentation (TRID)</title>
    <link>https://trid.trb.org/</link>
    <atom:link href="https://trid.trb.org/Record/RSS?s=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" rel="self" type="application/rss+xml" />
    <description></description>
    <language>en-us</language>
    <copyright>Copyright © 2026. National Academy of Sciences. All rights reserved.</copyright>
    <docs>http://blogs.law.harvard.edu/tech/rss</docs>
    <managingEditor>tris-trb@nas.edu (Bill McLeod)</managingEditor>
    <webMaster>tris-trb@nas.edu (Bill McLeod)</webMaster>
    <image>
      <title>Transport Research International Documentation (TRID)</title>
      <url>https://trid.trb.org/Images/PageHeader-wTitle.jpg</url>
      <link>https://trid.trb.org/</link>
    </image>
    <item>
      <title>Automated knowledge elicitation for rail vehicle identification using vision-language models</title>
      <link>https://trid.trb.org/View/2701467</link>
      <description><![CDATA[Rail transportation accounts for a significant portion of long-distance freight movement. Existing datasets typically rely on aggregated reports. Recent studies have explored vision-based deep learning models to capture details of locomotives and railcars. However, training an accurate supervised deep learning model requires a labor-intensive data annotation. Vision-language models (VLMs) have shown strong potential in general-purpose visual recognition tasks by aligning image regions with textual descriptions. However, their performance in specialized transportation domains often declines without domain-specific knowledge refinement. Current methods typically rely on labor-intensive crafted prompts or manual incorporation of expert knowledge. To address this limitation, this study proposes Knowledge Auto-Elicitation for System Prompts (KAESP), a novel framework that systematically extracts domain knowledge from identification errors and iteratively refines system prompts, which enables adaptation without model retraining or parameter updates. KAESP integrates automatically elicited textual knowledge with few-shot visual examples to create a system prompt that bridges visual grounding and domain expertise to enhance fine-grained railway vehicle identification. Experimental evaluations were conducted on a real-world dataset with 24 locomotive and railcar classes collected in Barstow, California. The results demonstrate that KAESP improved the average F1 score from 0.77 with only few-shot in-context learning to 0.82, with 14 classes achieving F1 scores above 0.80. Significant improvements were observed for several railcar categories, such as Container on Flatcar Double-Stacked with one 53-foot and two Twenty-foot Equivalent Unit containers F1 score improved from 0.41 to 0.61, and Gondola increased from 0.66 to 0.80. The results show that auto-elicitation knowledge significantly enhances VLM performance on rail train understanding.]]></description>
      <pubDate>Thu, 27 Aug 2026 16:32:56 GMT</pubDate>
      <guid>https://trid.trb.org/View/2701467</guid>
    </item>
    <item>
      <title>Fine-Grained Vehicle Classification Using Loop Detectors: A Wireless Fingerprinting Approach</title>
      <link>https://trid.trb.org/View/2734986</link>
      <description><![CDATA[The lack of secure identification of vehicles poses multiple threats to a rapidly growing intelligent transportation system. One notable threat is the impersonation of an authorized vehicle to deceive the automated vehicle access control (VAC) system, designed to prevent unauthorized vehicles from entering restricted areas or accessing special privileges. Another potential attack involves battery electric vehicles (BEVs), wherein an adversary that has compromised the BEV's software system deceives a charging station (plug-in or wireless) into overcharging the vehicle to produce catastrophic failure of the battery pack, including combustion or explosion. To address these threats we propose to identify vehicles, and hence their privileges and/or charging capabilities, using device fingerprinting. In particular, we leverage inductive loop detectors (ILD) to determine the make, model, and year of vehicles. A wide-band signal is used to capture unique frequency-dependent features of a vehicle resulting from its size, shape, metal structure, and content. An ILD-based approach is cost-effective to implement as it utilizes an already widely deployed infrastructure, in contrast to approaches that use surveillance cameras or other sensors. A circuit-level, low-cost, drop-in replacement to enable existing ILD deployments to fingerprint vehicles is proposed and realized as a custom, open-source PCB. A comprehensive evaluation against a dataset acquired from 32 vehicles of different make, model, and year, over nine months, shows that the proposed approach is effective even under temporal, environmental, and measurement variation without retraining. Overall multi-class classification accuracy of up to 93% is achieved.]]></description>
      <pubDate>Wed, 26 Aug 2026 14:11:01 GMT</pubDate>
      <guid>https://trid.trb.org/View/2734986</guid>
    </item>
    <item>
      <title>Using Artificial Intelligence-Based Computer Vision for Traffic Monitoring </title>
      <link>https://trid.trb.org/View/2736590</link>
      <description><![CDATA[The proposed research project would employ Artificial Intelligence (AI)-based Computer Vision to collect traffic data including vehicle count (traffic volume), Federal Highway Administration (FHWA) vehicle classification, and turning movements. Using video footage taken by existing roadside cameras (such as the Ohio Department of Transportation (ODOT)'s Milestone cameras) or other temporary cameras to collect traffic data allows for a safer and less expensive option compared with other methods. Computer Vision is an important AI application and using it to automatically collect traffic data will save significant amount of time and cost. This project epitomizes innovation as it will use cutting edge technologies to perform practical tasks while improving ODOT workers' and driving public's safety, reducing cost, and enhancing operational efficiency. 

A previous project conducted by the University of Toledo for ODOT Office of Technical Services developed a prototype tool that used AI-Computer Vision to collect vehicle count and classification data from recorded roadway videos. This research would investigate ways to more efficiently access videos from Milestone cameras and improve the previous work by developing a web-based traffic monitoring application, so that it can be used for routine traffic flow data collection. The outcome of this research will provide ODOT a safe, efficient, accurate, and economical alternative for collecting traffic data.

The overall goal of the project is to develop a safer method to collect accurate traffic data, while reducing costs and increasing operational efficiency.
                  ]]></description>
      <pubDate>Mon, 27 Jul 2026 13:57:59 GMT</pubDate>
      <guid>https://trid.trb.org/View/2736590</guid>
    </item>
    <item>
      <title>Boosting Classification of Electric Vehicles from Charging Patterns for Smart Mobility and Sustainability</title>
      <link>https://trid.trb.org/View/2713919</link>
      <description><![CDATA[While electric vehicles (EVs) offer an eco-friendly transportation option, uncoordinated residential charging may lead to peak demand, asset stress, and inefficient capacity utilization, which adversely affects smart grid operations. Addressing these challenges requires timely and accurate identification of EV charging behavior at fine temporal resolution. Accordingly, this paper addresses the practical task of classifying EVs directly from statistics over 1-minute charging profiles using a two-stage pipeline suitable for real-time deployment. In the first stage, segmentation is performed by applying K-means clustering with hysteresis to the average real power signal, allowing the classifier to be trained on charging-only intervals. In the second stage, additional features are extracted to summarize short-horizon dynamics including level, slope, and stability over two rolling windows. We then train three gradient-boosting tree classifiers and analyze their performance using different metrics. The best performing model is LGBM with an accuracy of 0.9951, macro-F1 of 0.9936, MCC of 0.9945, top-3 accuracy and AUC (macro/micro) of approximately 1. The inference latency is approximately 0.044 ms/sample and the training time is approximately 3.13s. Treating segmentation as preprocessing sharpens class signatures, and causal short-horizon statistics disambiguate overlapping plateaus to produce strong, interpretable decision boundaries—enabling type-aware power budgeting, safer concurrent charging, and standardized interoperability tests using only commodity smart meter signals.]]></description>
      <pubDate>Mon, 27 Jul 2026 11:16:37 GMT</pubDate>
      <guid>https://trid.trb.org/View/2713919</guid>
    </item>
    <item>
      <title>A lightweight hybrid machine and deep learning framework via contrastive learning for audio-based multiclass vehicle counting</title>
      <link>https://trid.trb.org/View/2682024</link>
      <description><![CDATA[Traditional multiclass vehicle counting relies on inductive loop detectors and camera-based systems, which present limitations including invasive installation, high maintenance costs, privacy concerns, weather vulnerability, and computational demands that hinder edge deployment. Audio-based methods offer cost-effective, privacy-preserving, all-weather and computationally lightweight alternative. However, existing approaches face significant challenges in handling class imbalance, resulting in poor performance on minority classes, particularly for heavy vehicles, which have a substantial impact on traffic management due to their disproportionate road impact. This study proposes a lightweight machine-learning-based pairwise classification framework that employs the CatBoost algorithm with optional diffusion-based augmentation to address class imbalance. The framework trains multiple binary classifiers with adaptive feature selection across six domains, incorporating synthetic samples for class pairs with insufficient or imbalanced data. This enables the training of high-performance binary classifiers, which are then aggregated into the final multiclass framework. Evaluation on three benchmark datasets, each with manually annotated ground truth labels, demonstrates consistent improvements over state-of-the-art methods under the same evaluation metrics: IDMT achieves an overall F1-score of 89.58% (+10.2 percentage points (pp) from a Transformer), with bus class improving from 42.86% to 77.78%; MELAUDIS achieves +4.5pp recall over VGGNet, and MAVD achieves an F1-score of +7.3pp over Random Forest.The proposed framework deploys only lightweight CatBoost classifiers, with the smallest model at 51 KB and ensemble inference latency of 26–127 ms on Raspberry Pi 4, enabling cost-effective and privacy-preserving vehicle monitoring across diverse road environments from highways to local roads that have lacked systematic surveillance. This provides the granular traffic data needed for road safety planning, infrastructure maintenance, and both short and long-term transportation management.]]></description>
      <pubDate>Mon, 29 Jun 2026 09:12:10 GMT</pubDate>
      <guid>https://trid.trb.org/View/2682024</guid>
    </item>
    <item>
      <title>Neural Networks-Assisted PCF Interferometers for Smart Road Traffic Surveillance</title>
      <link>https://trid.trb.org/View/2617763</link>
      <description><![CDATA[Efficient road traffic monitoring has a significant impact on the social and economic welfare of modern cities. Hence, developing a sensitive detection system that identifies various features of moving vehicles in real-time could suffice the demand for an integrable and competent traffic surveillance network. In this work, we propose and demonstrate a combination of in-line fiber interferometer-based sensing probes to act as weigh-in-motion systems (WIMs) for active detection of vehicular flow. The interferometers are fabricated using solid core photonic crystal fiber (SCPCF) sections, whose sensitivity is tailored by varying the dimensions of the section and installation depth of the interferometers. The sensitive interferometry-based probes operate by wavelength interrogation and enable the detection of a wide range of vehicles, from bicycles to buses. To augment the functionality of the proposed WIMs, artificial neural network (ANN) models are developed that enable the determination of vehicle parameters such as average speed, weight, wheelbase, and class from the recorded temporal response of the sensing probes. The models perform accurately in determining the vehicle features, with the classification model having an average accuracy of 96%. The algorithms can be integrated with the Internet of Things (IoT) to establish a cloud-based traffic monitoring interface for civic departments and public utilities. The interferometry-enabled WIMs are tailorable and compatible with varied road conditions. Besides road traffic monitoring, the proposed system can also be used for security surveillance, structural assessment, and industrial processes.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:36 GMT</pubDate>
      <guid>https://trid.trb.org/View/2617763</guid>
    </item>
    <item>
      <title>Weakly Supervised Bilinear Convolutional Neural Network for Fine-Grained Vehicle Classification</title>
      <link>https://trid.trb.org/View/2659021</link>
      <description><![CDATA[Fine-grained vehicle classification, which is a key technology within intelligent transportation systems, has been gaining increasing importance with the burgeoning growing number of vehicles. Previous studies have predominantly focused on intricate and distinctive local features. However, in various tasks, it has been proven that global features are of significant importance when they can be effectively integrated with local features in a harmonious manner. So, we consider that a comprehensive consideration of both local and global features is crucial for enhancing classification decisions. Consequently, the paper designs a novel architecture for the task, which combines global and local features to improve classification performance. The architecture consists of two components: the local-feature net and the global-feature net. Specially, for the local feature, we propose an Essential Part Locator module that uses global feature-weighted attention masks to obtain local features, and a Cross-Part Feature Transformer that boosts interactions between local features. Meanwhile, our architecture processes the entire image through an encoder to capture global features and then integrates both global and local features. Experimental results on the Stanford Cars, CompCars, and BoxCars116K datasets demonstrate that the proposed approach surpasses state-of-the-art methods, achieving accuracies of 97.5%, 96.4%, and 92.1%, respectively.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:32 GMT</pubDate>
      <guid>https://trid.trb.org/View/2659021</guid>
    </item>
    <item>
      <title>Traffic Management and Control Strategies for Reducing Air and Noise Pollution in Private Road Transport</title>
      <link>https://trid.trb.org/View/2579525</link>
      <description><![CDATA[Nowadays, the environmental sustainability of human activities is a widely debated topic in the context of social and economic development. Especially in the transport sector, a major contributor to greenhouse gas emissions and noise pollution, several efforts have been adopted aimed at addressing these environmental challenges, such as new technological solutions in the automotive sector, the implementation of the EURO standards and improvements in road surfaces and pavements. This study provides a contribution to the enhancement of air quality in the areas close to the road infrastructures, with the aim of reducing both air and noise pollution due to private road transport, by managing and controlling traffic flow through different actions and specific policies. These strategies are designed to influence users towards adopting more sustainable driving behaviours. In this paper, several traffic management solutions will be analyzed, combining different policies that act on traffic dynamics, involving the project’s variables in different ways. These policies will include progressive reduction of speed limits, lane openings and closures based on vehicle classes, and changes in the composition and volume of vehicle fleets. Results obtained from an extensive laboratory test campaign conducted on a real network will be discussed to identify the effects of actions and policies on air pollution, noise, and travel time. Finally, the study will propose a procedure to optimize the environmental performance of the system. This optimization will consider not only the environmental aspects but also the overall performance and level of service of the infrastructure.]]></description>
      <pubDate>Mon, 27 Apr 2026 15:01:25 GMT</pubDate>
      <guid>https://trid.trb.org/View/2579525</guid>
    </item>
    <item>
      <title>A Data Quality Framework on Urban Vehicle Access Regulations</title>
      <link>https://trid.trb.org/View/2682777</link>
      <description><![CDATA[Urban Vehicle Access Regulations (UVARs) are policy measures aimed at regulating vehicle access in urban areas, considering criteria such as emission standards, vehicle classifications, and time-based restrictions. UVARs are crucial when pursuing environmental and traffic management objectives. While there have been relevant endeavours in digitising those regulations, the lack of harmonised quality definitions and assessment methods hinders the effective reuse of UVAR data by service providers. This paper addresses this gap by developing a Data Quality Framework (DQF) for UVARs, under the umbrella of the NAPCORE project. The framework was built by tailoring general data quality dimensions to UVAR-specific needs, defining a set of criteria and metrics, and introducing a method to compute quality indicators. It comprises eight quality dimensions, 12 criteria, and 15 metrics, and was tested through a case study using a real-world Low Emission Zone (LEZ) dataset from the city of Lisbon (Portugal). Despite some limitations, the framework offers a flexible approach for assessing and improving the quality of UVAR datasets, setting the ground for higher availability of UVAR data and facilitating its reuse and delivery to commuters, thereby contributing to urban mobility goals.]]></description>
      <pubDate>Mon, 27 Apr 2026 15:01:17 GMT</pubDate>
      <guid>https://trid.trb.org/View/2682777</guid>
    </item>
    <item>
      <title>LSTM-based Vehicle Type Classification using Real-time Trajectory Data in Automated Driving</title>
      <link>https://trid.trb.org/View/2682142</link>
      <description><![CDATA[In the era of automated driving, accurately understanding the behavior of surrounding vehicles is crucial for safe and efficient driving actions. Since each vehicle type exhibits a unique driving behavior, understanding other vehicles’ types and intentions enables AVs to make more informed decisions regarding predicting nearby vehicles’ future positions and actions. In this research, we implement a multi-layer LSTM model for vehicle type classification task using the trajectories of nearby vehicles in a traffic scene. The model takes the positions and actions of a vehicle over several time steps and outputs the probability distribution of vehicle classes. We test the model with HighD and NGSIM datasets and conduct performance evaluations. The findings reveal that the proposed model could accurately capture the vehicle classes from the vehicle’s driving behavior.]]></description>
      <pubDate>Mon, 27 Apr 2026 15:01:17 GMT</pubDate>
      <guid>https://trid.trb.org/View/2682142</guid>
    </item>
    <item>
      <title>Machine Vision Toolkit for Automated Fleet Composition Assessment and Reporting</title>
      <link>https://trid.trb.org/View/2691665</link>
      <description><![CDATA[State Departments of Transportation (DOTs) and Metropolitan Planning Organizations (MPOs) employ fleet composition data (e.g., passenger vehicles, single-unit trucks, and combination trucks) in a variety of planning, economic, roadway performance, and safety applications. Accurate fleet composition data is essential for pavement management, safety analysis, and fuel consumption modeling. However, traditional methods are labor-intensive, costly, and often lack the temporal or spatial resolution required to capture variations between freeways, arterials, and managed lanes vs. general-purpose lanes. Using machine vision tools to quickly, efficiently, and accurately capture on-road percentages of light-duty vehicle, light-duty truck, medium-duty truck, and a variety of heavy-duty truck classifications will enhance analytical and modeling accuracy and reduce state DOT data management costs. Building upon prior National Center for Sustainable Transportation (NCST) research that developed machine vision algorithms for vehicle identification, this project will package those research findings into a deployable, open-source Automated Fleet Classification Toolkit for practitioners and researchers. The research team will develop and release comprehensive Standard Operating Procedures (SOPs) and software tools allowing agencies to convert standard roadside or overpass video feeds into high-resolution fleet composition data. The toolkit will utilize advanced object detection (e.g., YOLO architectures) to automate the identification of vehicle classes (aligning with FHWA 13-category schemes where possible) and propulsion types based on visual vehicle features. The system is designed to distinguish traffic conditions on complex roadway geometries, allowing users to generate separate classification profiles for managed lanes vs. general-purpose lanes, and separating freeway mainlines from adjacent arterial service roads. The project focuses on technology transfer: providing the "how-to" manuals, open-source code, and data processing protocols so that State DOTs, consultants, university partners and research institutes can replicate the data collection and extraction without relying on proprietary "black box" services.]]></description>
      <pubDate>Sun, 12 Apr 2026 23:29:28 GMT</pubDate>
      <guid>https://trid.trb.org/View/2691665</guid>
    </item>
    <item>
      <title>Automated FHWA Vehicle Classification System via LiDAR and Camera Data Fusion</title>
      <link>https://trid.trb.org/View/2686268</link>
      <description><![CDATA[The FHWA classification scheme provides a standardized method for vehicle classification based on the number of axles and axle configuration, making it essential for effective traffic management, toll collection, and transportation planning. Although image-based classification methods have advanced significantly, they often struggle to accurately distinguish between certain FHWA vehicle classes using visual data alone. This challenge has led researchers to combine classes that are difficult to differentiate visually, thereby reducing the granularity and effectiveness of classification systems. In this research, we present a novel approach that enhances vehicle classification by leveraging the complementary strengths of both camera and LiDAR data. Because of the lack of existing fine-grained camera-LiDAR data sets for vehicle classification, we collected our own comprehensive dataset for the FHWA 13 classes. We employ a pretrained YOLOv8 model to detect and localize vehicles. Features from these localized images are extracted using a ResNet model. Simultaneously, the localized LiDAR point clouds are used to generate distance maps and calculate vehicle features such as length and height. The combined features and dimensions are input into a camera-LiDAR fusion multilayer perceptron classifier for final classification. To evaluate the performance of our proposed models, we employed metrics including recall, precision, and F1 score. Our comparative analysis demonstrates that the model fusing images and the distance maps yielded the highest performance, achieving a precision of 0.958, a recall of 0.932, and an F1 score of 0.938 across all classes.]]></description>
      <pubDate>Thu, 02 Apr 2026 15:23:59 GMT</pubDate>
      <guid>https://trid.trb.org/View/2686268</guid>
    </item>
    <item>
      <title>Intelligent Vehicle Automatic Identification and Classification With Distributed Acoustic Sensing</title>
      <link>https://trid.trb.org/View/2610672</link>
      <description><![CDATA[Distributed Acoustic Sensing (DAS) is an emerging vibration collection technology with advantages such as low cost, high-density sampling, and high sensitivity. It utilizes regional unlit fiber-optic telecommunication infrastructure (dark fiber) in the urban underground to record real-time environmental signals. How to identify vehicle signals, classify vehicle types, and estimate vehicle speeds from DAS signals has significant potential for the development of intelligent urban transportation. Addressing the challenges of high-noise environments and dense traffic, we develop an end-to-end two-stage deep learning process to identify and classify various vehicle signals in urban traffic rapidly. First, we propose the CarDenoiseNet network, based on Generative Adversarial Networks (GAN) and contrastive learning, to denoise and enhance the weak signal of DAS data. Then, the YOLOv8 segmentation model is employed to segment and classify the vehicle signal. Finally, the vehicle speeds are estimated using the segmented vehicle trajectory time and location information. We use the urban DAS field data recorded in the downtown area of Changchun to test the proposed workflow. The test result has good generalization and over 90% accuracy in identifying different vehicle types and speeds in high-density traffic environments. Moreover, transfer learning successfully applies the model to other datasets, proving its excellent generalization ability. Additionally, statistical analysis of traffic flow and speed trends provides technical references for alleviating urban traffic congestion, reducing traffic accidents, and enhancing the intelligence level of urban traffic management.]]></description>
      <pubDate>Thu, 26 Mar 2026 17:02:26 GMT</pubDate>
      <guid>https://trid.trb.org/View/2610672</guid>
    </item>
    <item>
      <title>Statistical analysis of vehicle usage intensity in a transport company</title>
      <link>https://trid.trb.org/View/2666015</link>
      <description><![CDATA[The article presents the application of selected statistical analyses to monitor the degree of efficiency of the use of a fleet of vehicles in a transport company. The main objective of the work was to present selected statistical tests for a given number of vehicles. 179 vehicles of various types and brands in use. Three groups of vehicles were distinguished in the analysis in terms of the load capacity of the cargo space: small cars, delivery vans and trucks. One of the factors differentiating vehicles within the distinguished groups was their mileage at the beginning of the observation period. Data on the vehicle usage intensity during one year of operation were analyzed. The single-factor statistical analyses used are of a preliminary nature, being an introduction to the issues of multifactor analysis. On the other hand, single-factor analyses can be used in the issues of classification of a heterogeneous the fleet of vehicles in road transport companies. The presented procedure showed the possibility of adapting statistical analyses to the management and forecasting of vehicle use in a transport company from the B2B and B2C sectors.]]></description>
      <pubDate>Mon, 23 Mar 2026 15:15:34 GMT</pubDate>
      <guid>https://trid.trb.org/View/2666015</guid>
    </item>
    <item>
      <title>Optimizing winter traffic forecasting through spatially transferable models in cold regions: Insights for infrastructure management</title>
      <link>https://trid.trb.org/View/2672072</link>
      <description><![CDATA[Transportation agencies in cold regions need winter-robust traffic forecasts that can transfer across road types when monitoring sites are sparse. Using five Alberta WIM sites (FMD for model development; LED, VID, RDD, LVD for transfer testing) and winter seasons (Nov–Mar, 2005–2009), we predict the day-ahead daily volume factor (DVF) for three vehicle classes (total, passenger cars, trucks) with four model structures: Winter-weather (Ww), Naïve (Na), Base (Ba), and Para (Pa). The Ww model combines an expected daily volume factor (EDVF), continuous snowfall, and temperature-category dummies. Spatial transfer tests show high accuracy across functionally similar and distinct highways. Representative gains versus baseline structures include LVD-Ww (total traffic) MAPE 6.23 % vs Ba 7.77 % (19.8 % reduction), and LED-Ww (total traffic) 5.46 % vs Ba 6.19 % (11.8 % reduction), with R²≈ 0.994–0.996. Class-specific patterns emerge: trucks often favor simpler structures (e.g., LVD-Na 4.88 % vs Ww 5.24 %-6.9 % reduction). Results indicate that careful model choice by vehicle class and road function enables spatially transferable winter forecasts without deploying additional WIM sites, supporting resource-efficient maintenance, operations, and traveler information.]]></description>
      <pubDate>Mon, 23 Mar 2026 15:15:32 GMT</pubDate>
      <guid>https://trid.trb.org/View/2672072</guid>
    </item>
  </channel>
</rss>