<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>From emotion to safety-critical events: The temporal causal pathways in naturalistic driving</title>
      <link>https://trid.trb.org/View/2700596</link>
      <description><![CDATA[Driving is a complex task demanding cognitive readiness and attentional resources. High emotion states, such as anger, can impair driving abilities and increase crash risk. There is a need to investigate how emotion affects driving behavior that leads to safety-critical events (SCEs), which include crashes and near-crashes. This study investigated the temporal sequence from emotion to SCEs using a naturalistic driving dataset, focusing on 193 emotion-involved driving events spanning a total of 7.8 h and their corresponding trips. Detailed data annotation of the selected trips examined associations between emotion, driving errors, and SCEs. The Kaplan-Meier method was employed to investigate the temporal sequence of critical episodes (emotion expression, driving errors, and SCEs). Results showed the risk of committing driving errors was higher when drivers experienced emotions, especially for SCEs. Emotional drivers tended to make more judgment errors than performance errors. Drivers commonly expressed emotion before driving errors, and driving errors frequently preceded SCEs. Eighty percent of crashes and 67% of near-crashes began with emotion followed by driving errors before the SCE. About 20% of near-crashes involved emotion expression or driving errors after the near-crash occurred. With integral emotion triggers (i.e., triggers associated with the driving task), angry drivers were more likely to commit driving errors before expressing emotion. These results highlight the role of emotional regulation in driver safety and provide critical information for developing safety countermeasures to mitigate the hazardous effects of emotion on driving risk.]]></description>
      <pubDate>Thu, 28 May 2026 09:03:40 GMT</pubDate>
      <guid>https://trid.trb.org/View/2700596</guid>
    </item>
    <item>
      <title>Assessing the Influence of Managed Lane Separation Type on Lane Deviation</title>
      <link>https://trid.trb.org/View/2701112</link>
      <description><![CDATA[Managed lanes (MLs) are increasingly implemented to enhance mobility and safety on congested freeways, but researchers have rarely explored how the type of separation between MLs and general-purpose lanes (GPLs) affects lane deviation, a key factor linked to crash risk, particularly sideswipe collisions. This study examines the relationship between ML separation type and lane deviation using naturalistic driving data from the Strategic Highway Research Program 2 in Washington State. The dataset includes 5,976 trips made by 310 drivers traveling along I-5 and I-90, focusing on three separation types: buffer, median, and concrete barrier. An Analysis of Variance was conducted to determine whether there were significant differences in mean lane deviation across separation types. A random forest model was then applied to assess the importance of explanatory variables, followed by a mixed-effects multinomial model to identify factors influencing lane deviation for drivers on ML facilities. Results show that, on average, drivers tend to steer slightly away from the separator across all designs. Median-separated facilities resulted in the highest mean lane deviation compared with concrete barriers and buffer-separated MLs. Separation type, driver age, driver gender, total miles driven, vision acuity, lane width, shoulder width, driver risk-taking behavior, and the number of MLs and GPLs were all significant factors influencing lane positioning behavior at the 95% confidence level. These findings provide valuable guidance for transportation agencies when selecting separation types for ML facilities, helping them improve safety, reduce crash risk, and improve the overall safety performance of MLs.]]></description>
      <pubDate>Mon, 11 May 2026 08:51:42 GMT</pubDate>
      <guid>https://trid.trb.org/View/2701112</guid>
    </item>
    <item>
      <title>Multi-head attention-based intelligent vehicle lane change decision and trajectory prediction model in highways</title>
      <link>https://trid.trb.org/View/2596547</link>
      <description><![CDATA[With the aim to improve the interaction between intelligent vehicles and human drivers, this article proposes the MCLG (multi-head attention + convolutional social pooling + long short-term memory + Gaussian mixture model) lane change decision and trajectory prediction model, which includes a lane-changing intention decision module. The model comprises a lane change decision module responsible for determining three lane change intentions: left lane change, right lane change, and car-following. Subsequently, a multi-head attention mechanism processes complex vehicle interaction information to enhance modeling accuracy and intelligence. In addition, uncertainty in trajectory prediction is considered by using multimodal trajectory prediction and Gaussian mixture model, and diversity and uncertainty are combined by combining trajectory prediction from several different modalities through probabilistic combinatorial sampling patterns. Test results indicate that the MCLG model, based on the multi-head attention module, outperforms existing methods in trajectory prediction. The decision module, which takes interactive information into account, exhibits higher predictability and accuracy. Furthermore, the MCLG model, considering the lane-changing decision module, significantly enhances trajectory prediction accuracy, providing robust decision-making support for autonomous driving systems.]]></description>
      <pubDate>Wed, 06 May 2026 15:21:26 GMT</pubDate>
      <guid>https://trid.trb.org/View/2596547</guid>
    </item>
    <item>
      <title>Measuring Cognitive Workload of Novice Law Enforcement Officers in a Naturalistic Driving Study</title>
      <link>https://trid.trb.org/View/2689056</link>
      <description><![CDATA[There is a large amount of variation between novices and experts in their cognitive workload when performing tasks. A naturalistic pilot study was conducted with nine novice law enforcement officers (nLEOs) to determine how their use of in-vehicle technology affected their cognitive workload during their normal patrols. Physiological data were collected using a novel synchronization process for naturalistic driving studies, allowing heart rate variability and eye tracking measurements to be synchronized together and directly compared to subjective workload levels. It was found that nLEOs have average or higher workload compared to experienced officers and the general population when they are on duty. Future studies can utilize the approaches and findings of this pilot study for conducting naturalistic driving studies and developing cognitive performance models for novice users.]]></description>
      <pubDate>Sun, 03 May 2026 18:19:54 GMT</pubDate>
      <guid>https://trid.trb.org/View/2689056</guid>
    </item>
    <item>
      <title>Enhancing Accessibility for Individuals With Limited Mobility: Leveraging AI and Cycling Data for an Inclusive Active Transportation Network</title>
      <link>https://trid.trb.org/View/2690989</link>
      <description><![CDATA[This study introduces a low-cost, scalable framework for monitoring and assessing the condition of paved trail infrastructure to enhance accessibility for all users, particularly those with limited mobility. The research utilizes a custom-built "Data Bike" with GPS and accelerometers embedded with high-resolution cameras to collect multimodal data along Utah's trail networks. A simple and intuitive data processing pipeline was developed that uses accelerometer-derived "jerk" events to automatically identify potential surface anomalies, triggering the extraction of corresponding image frames. A deep-learning object detection model (YOLOv8) was fine-tuned on a custom dataset and augmented with public imagery to detect and classify ten surface defects and obstructions, including various cracks, potholes, and vegetation obstacles. The model achieved promising results, with detections geolocated to enable spatial analysis and hotspot identification. This data-driven approach provides transportation agencies an objective and efficient tool to move from reactive to proactive maintenance, allowing for better resource allocation and systematic improvements to trail safety and accessibility. The project lays the groundwork for an integrated asset management system, including an interactive dashboard for visualizing trail conditions and prioritizing repairs.]]></description>
      <pubDate>Thu, 30 Apr 2026 09:11:06 GMT</pubDate>
      <guid>https://trid.trb.org/View/2690989</guid>
    </item>
    <item>
      <title>Human-Like Implicit Intention Expression for Autonomous Driving Motion Planning Based on Learning Human Intention Priors</title>
      <link>https://trid.trb.org/View/2659105</link>
      <description><![CDATA[One of the key factors determining whether autonomous vehicles (AVs) can be seamlessly integrated into existing traffic systems is their ability to interact smoothly and efficiently with human drivers and communicate their intentions. While many studies have focused on enhancing AVs' human-like interaction and communication capabilities at the behavioral decision-making level, a significant gap remains between the actual motion trajectories of AVs and the psychological expectations of human drivers. This discrepancy can seriously affect the safety and efficiency of AV-HV (Autonomous Vehicle-Human Vehicle) interactions. To address these challenges, we propose a motion planning method for AVs that incorporates implicit intention expression. First, we construct a trajectory space constraint based on human implicit intention priors, compressing and pruning the trajectory space to generate candidate motion trajectories that consider intention expression. We then apply maximum entropy inverse reinforcement learning to learn and estimate human trajectory preferences, constructing a reward function that represents the cognitive characteristics of drivers. Finally, using a Boltzmann distribution, we establish a probabilistic distribution of candidate trajectories based on the reward obtained, selecting human-like trajectory actions. We validated our approach on a real trajectory dataset and compared it with several baseline methods. The results demonstrate that our method excels in human-likeness, intention expression capability, and computational efficiency.]]></description>
      <pubDate>Wed, 29 Apr 2026 09:10:08 GMT</pubDate>
      <guid>https://trid.trb.org/View/2659105</guid>
    </item>
    <item>
      <title>Biosignals Based Automated Driver Cognitive Load Assessment Using a Pre-Trained Transformer</title>
      <link>https://trid.trb.org/View/2659123</link>
      <description><![CDATA[Assessing driver cognitive load (DCL) is essential for enhancing driving safety and performance. This study introduces a novel method that leverages a pre-trained biosignal transformer (BIOT) to classify DCL using unimodal and multimodal multivariate biosignals. The proposed transformer-based model features an encoder-only architecture with efficient multi-head self-attention to process input tokens derived from various biosignals. Our approach includes Fourier-based biosignal tokenization techniques and a perturbation module as a data augmentation layer. Additionally, we employed a hierarchical spatiotemporal embedding framework, using additive absolute position embeddings for temporal encoding and additive learnable tokens for each channel for spatial encoding. Experimental results demonstrate that our model, trained on raw biosignals, outperforms previous methods, highlighting its potential for monitoring DCL using wearable devices. It achieved error rates that were at least 2.4% and up to 20% lower in ternary classification tests when compared to the baseline models. The findings suggest that integrating a pre-trained transformer-based architecture with perturbation-based augmentation can significantly enhance the model's accuracy and robustness. Our results offer a promising direction for future research and development in intelligent transportation systems.]]></description>
      <pubDate>Wed, 29 Apr 2026 09:10:07 GMT</pubDate>
      <guid>https://trid.trb.org/View/2659123</guid>
    </item>
    <item>
      <title>Predicting Crash Severity using Naturalistic Driving Data and Neural Networks</title>
      <link>https://trid.trb.org/View/2658008</link>
      <description><![CDATA[This study leverages artificial intelligence (AI) to predict crash severity using naturalistic driving data (NDD) from the SHRP-2 dataset. It employs logistic regression and Recursive Feature Elimination (RFE) for feature selection, SHapley Additive exPlanations (SHAP) for model interpretation, and Feedforward Neural Networks (FNN) for prediction. The FNN models achieved high accuracy, correctly classifying severe crashes at 93.20% and moderate crashes at 94.98%. The Synthetic Minority Oversampling Technique (SMOTE) improved predictive performance for severe crashes, which were less frequent in the dataset. SHAP analysis identified near-miss events and driver responsibility indicators as key predictors of crash severity. The study underscores the importance of AI-driven predictive analytics in traffic safety, emphasizing the need for targeted interventions to mitigate crash risks. These insights can guide policymakers in developing effective strategies to enhance road safety and integrate AI technologies into traffic management systems.]]></description>
      <pubDate>Tue, 21 Apr 2026 14:30:17 GMT</pubDate>
      <guid>https://trid.trb.org/View/2658008</guid>
    </item>
    <item>
      <title>Wayside Detector System Study</title>
      <link>https://trid.trb.org/View/2689420</link>
      <description><![CDATA[This report evaluates both existing and emerging wayside detector systems, assessing their effectiveness, life-cycle costs, and broader industry impacts, to satisfy Minnesota’s legislative mandate for a comprehensive analysis of these safety systems. By providing a data-driven assessment of safety benefits, operational impacts, and implementation pathways, the report will help public agencies, railroads, and technology vendors align on good practice. The report is organized as follows: (Chapter 2) introduces the types of wayside detectors currently in service in Minnesota; (Chapter 3) presents additional detector technologies implemented in North America and internationally to determine good practice and innovation trends; (Chapter 4) assesses accuracy, reliability, detection rates, siting considerations, data management, and maintenance requirements for wayside detectors; (Chapter 5) discusses safety benefits through statistical analysis of national and Minnesota-specific incident data; (Chapter 6) details capital, operating, and maintenance cost ranges and benefits of wayside detectors, discusses industry impacts of wayside detector installation scenarios, and outlines federal and state funding or financing options; and (Chapter 7) examines federal preemption and Federal Railroad Administration (FRA) guidance on statewide wayside detector system deployment.]]></description>
      <pubDate>Thu, 16 Apr 2026 16:54:37 GMT</pubDate>
      <guid>https://trid.trb.org/View/2689420</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>Supervised visual docking network for unmanned surface vehicles using auto-labeling in real-world water environments</title>
      <link>https://trid.trb.org/View/2641351</link>
      <description><![CDATA[Unmanned Surface Vehicles (USVs) are increasingly applied to water operations such as environmental monitoring and river-map modeling. However, precise autonomous docking at ports or stations remains a significant challenge, often relying on manual control or external positioning systems, which severely limits fully autonomous deployments. In this paper, we propose a novel supervised learning framework featuring an auto-labeling pipeline to enable USVs autonomous visual docking. The primary innovation lies in our automated data collection pipeline, which directly provides paired relative pose data and corresponding images, eliminating the conventional need for manual labeling, such as tagging bounding boxes. We introduce the Neural Dock Pose Estimator (NDPE), capable of accurately predicting the relative dock pose without relying on traditional methods such as handcrafted feature extraction, camera calibration, or peripheral markers. Unlike common bounding-box-based detection algorithms (e.g., Yolo-like methods), our NDPE explicitly predicts the relative pose transformation between the camera frame and USV body frame, significantly simplifying the data annotation and training process. Additionally, the generality of our data collection pipeline allows integration with various neural network architectures, ensuring broad applicability beyond the specific architecture demonstrated here. Experimental validation in real-world water environments demonstrates that NDPE robustly handles variations in docking distances and USV velocities, ensuring accurate and stable autonomous docking performance. The effectiveness and practicality of our approach are clearly verified through extensive experiments. The dataset, tutorial and experimental videos for this project are publicly available at: https://sites.google.com/view/usv-docking/home.]]></description>
      <pubDate>Fri, 20 Mar 2026 14:47:19 GMT</pubDate>
      <guid>https://trid.trb.org/View/2641351</guid>
    </item>
    <item>
      <title>Understanding driver visual distraction and its relationship with speeding: Insights from naturalistic driving data</title>
      <link>https://trid.trb.org/View/2668526</link>
      <description><![CDATA[Urban environments involve complex traffic, roadside features, and varying speed limits that shape drivers' visual attention. Driving speed influences how attention allocation and surrounding stimuli are perceived. Notably, visual distraction, defined as gaze diverted from driving, and speeding co-occur. Yet previous studies rarely addressed heterogeneity by relative speed ratio based on speed limit and condition under which these behaviors coincide. This study employs naturalistic driving study data and applies a Bayesian hierarchical probit model to assess heterogeneity in visual distraction across speed ratio levels, followed by a conditional Bayesian regularized horseshoe (RHS) probit model to identify the conditions under which visual distraction and speeding co-occur. The results show that visual distraction varies across speed ratio levels, with the effects of traffic density and rainfall differing, and visual distraction becoming less likely as the speed ratio increases. The conditional Bayesian RHS probit model indicates that, under speeding conditions, visual distraction is less likely on multilane roadways with high traffic density and in areas characterized by complex land-use combined with higher posted speed limits. Under conditions of visual distraction, speeding is more likely with longer headways and less likely when driving in the rightmost lane adjacent to sidewalks where pedestrians are present. Speeding under visual distraction is more predictable than visual distraction under speeding, indicating that the lower predictability of the latter limits context-based explanations of visual distraction under speeding. This study reveals visual distraction across relative speeds and provides evidence on its co-occurrence with speeding, offering new insights into their relationship.]]></description>
      <pubDate>Wed, 18 Mar 2026 09:00:15 GMT</pubDate>
      <guid>https://trid.trb.org/View/2668526</guid>
    </item>
    <item>
      <title>Subjective driving experience recorded live: A naturalistic driving approach</title>
      <link>https://trid.trb.org/View/2665972</link>
      <description><![CDATA[For the development of human-like automated driving functions, recent years have increasingly highlighted the importance of understanding drivers' everyday experiences, which have received less attention compared to the well-established research on human driving behavior. Drivers' subjective perspectives on daily driving events have so far been based primarily on retrospective data collected through online surveys, interviews or diary studies – resulting in potentially biased post-trip assessments. Surprisingly, this has not been amended with the recent increase of Natural Driving Studies (NDS) which capture real-time driving behaviors, offering genuine insights into drivers' daily driving events. The research fills this gap and takes a step towards collecting subjective assessments during a daily drive. The authors designed, developed and tested a smartphone-based qualitative approach to assess drivers' subjective impressions of their driving experiences, in particular their perceptions of challenging situations. Drawing on results from two studies with N = 110 participants who produced 2514 voice recordings, the authors present nuances, advantages, and limitations of this novel approach in comparison to existing methodologies. The authors show that this approach can capture in real time the wide range of events that drivers are confronted with in daily driving and their assessment thereof. With this approach, the authors further advance the understanding of drivers' subjective assessment during a drive and build a foundation for future research.]]></description>
      <pubDate>Wed, 18 Mar 2026 09:00:01 GMT</pubDate>
      <guid>https://trid.trb.org/View/2665972</guid>
    </item>
    <item>
      <title>Technical Factors Matter for Driving Behavior Analysis: Experiment and Validation</title>
      <link>https://trid.trb.org/View/2617941</link>
      <description><![CDATA[Technological factors like driver assistance systems have the potential to influence human driving behavior, forming their driving habits. However, the precise mechanisms underlying the impact of these technological factors on behavior remain elusive. This paper specifically targets drive modes (i.e., eco and sport) as the technological factors and explores their influence on humans' driving behavior. First, we collect naturalistic and simulation driving data from 16 drivers who operated vehicles in eco and sport modes. We then employ eight hypothesis tests to analyze the changes in driving behavior with eco and sport modes. This study reveals that human drivers' behavior preferences are related to the level of urgency of the driving task that needs to be completed, with observable behavior characteristics being contingent upon both the drive mode and the complexity of traffic situations. We experimentally speculate that single-drive modes such as eco or sport mode cannot adapt to the full range of driving needs due to differences in the level of urgency of driving tasks to be completed, which is further validated with practical applications in simulation. These findings provide the basis for improving the acceptance of human-centric driver assistance systems for different vehicles.]]></description>
      <pubDate>Mon, 09 Feb 2026 08:53:26 GMT</pubDate>
      <guid>https://trid.trb.org/View/2617941</guid>
    </item>
    <item>
      <title>Development of a Trail Management Program</title>
      <link>https://trid.trb.org/View/2659347</link>
      <description><![CDATA[This report outlines the development of a comprehensive trail management program designed to improve the maintenance of paved trails in Central Iowa. Using the innovative Iowa Data Bike—a data collection tool developed and built by the Des Moines Area Metropolitan Planning Organization in 2017 that integrates an electric-assist bicycle with a smartphone and a GoPro camera—the program collected detailed data on trail surface roughness and overall conditions. The primary objectives were to establish a standardized metric for trail surface roughness, termed the Bike Roughness Index (BRI), and to develop an automated framework for surface distress evaluation using deep learning techniques. The BRI was calculated using accelerometer data, while trail conditions were assessed through double integration and wholebody vibration analyses. Additionally, a Mask R-CNN model was trained to detect, classify, and segment various types of trail surface distress, enabling accurate and efficient condition assessments. The results demonstrate that the Iowa Data Bike is effective for rapid data collection and that the BRI is a reliable measure of surface roughness. The deep learning model successfully identified and quantified different distress types, which can significantly aid in maintenance planning. This integrated approach supports proactive maintenance strategies, ensuring the safety and quality of recreational trails. Moreover, the report provides guidelines for optimal data collection, highlighting the potential for these methods to be refined and applied across diverse trail management contexts.]]></description>
      <pubDate>Fri, 06 Feb 2026 13:53:32 GMT</pubDate>
      <guid>https://trid.trb.org/View/2659347</guid>
    </item>
  </channel>
</rss>