<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>Influence of ship manoeuvres on collision damage</title>
      <link>https://trid.trb.org/View/2633436</link>
      <description><![CDATA[Ship collisions, though infrequent, can result in severe consequences, including the loss of human life, ships, or cargo, as well as substantial environmental damage. This paper presents a comprehensive study of ship collision phenomena by analysing two key aspects: pre-collision manoeuvres under different rudder angles and comparative collision scenarios involving varying parameters such as impact angle and ship velocity. The study begins by examining various manoeuvres a ship can perform to minimise the risk of collision, taking into account positional and hydrodynamic factors. The second part of the study evaluates collision scenarios through a risk assessment framework, contrasting a high-risk collision scenario with a low-risk scenario, both modelled under the assumption of human error. Both analyses use rotational velocity and acceleration metrics derived from a three-degrees-of-freedom (3DOF) manoeuvrability model established in the first part. Simulations are conducted using LS-DYNA. The model incorporates hydrodynamic forces generated during the collision, as well as forces resulting from ship manoeuvring. The KVLCC2 serves as a case study, with its hydrodynamic properties - such as added mass and viscous damping, determined using Hydrostar software for integration into the Mitsubishi Collision Code (MCOL) boundary condition. The results indicate that large rudder angles (±40°) significantly reduce collision energy and penetration depth compared to no manoeuvre, while higher collision angles also mitigate structural deformation. The findings confirm that incorporating manoeuvrability into collision simulations improves predictive accuracy and provides valuable insight for assessing ship safety and collision prevention strategies.]]></description>
      <pubDate>Fri, 20 Feb 2026 09:04:07 GMT</pubDate>
      <guid>https://trid.trb.org/View/2633436</guid>
    </item>
    <item>
      <title>A note on observed injury bias in police-reported pre-crash travel speed estimates</title>
      <link>https://trid.trb.org/View/2617181</link>
      <description><![CDATA[Vehicle pre-crash travel speed is one of the most important determinants of driver injury severity. However, pre-crash travel speed estimates made by police officers, especially those in crashes with less severe injuries (where there is less of a need for high levels of accuracy due to potential litigation), can be susceptible to biases because of the tendency to associate less severe driver injuries with lower pre-crash travel speeds. This potential bias makes the use of pre-crash travel speeds in injury-severity modeling highly problematic due to its endogeneity with injury severity. To detect the presence and extent of this problem, a bias correction term for pre-crash travel speed estimation equations is applied by treating injury-severity level (discrete) and pre-crash travel speed (continuous) as a discrete/continuous econometric model. The findings show that for severe injury crashes, the bias correction is statistically insignificant, reflecting the increased accuracy required of police officers in severe crashes. However, for crashes resulting in less severe occupant injuries, there is a significant bias resulting from observed injury levels, which distorts the effects of explanatory variables on pre-crash travel speed estimates. The results of this paper not only provide empirical evidence of potential endogeneity problems in models of crash injury severity but also underscore the need to more fully consider potential endogeneity issues and their associated consequences in statistical models and machine learning models.]]></description>
      <pubDate>Mon, 09 Feb 2026 08:53:26 GMT</pubDate>
      <guid>https://trid.trb.org/View/2617181</guid>
    </item>
    <item>
      <title>Analyzing Pedestrian–Automated Vehicle Crash Dynamics: A Comparative Study of Autonomous and Conventional Precrash Mode</title>
      <link>https://trid.trb.org/View/2665686</link>
      <description><![CDATA[Pedestrian safety is a critical concern in urban environments, particularly with the increasing presence of automated vehicles (AVs) on the roads. Because of the unpredictable movement of pedestrians, a significant challenge lies in the limited understanding of factors contributing to pedestrian–AV collisions. This study addresses this gap by analyzing pedestrian crashes involving AVs using association rules mining (ARM). Data, including crash reports from the California Department of Motor Vehicles, comprised 46 pedestrian crashes involving AVs, categorized by precrash mode: autonomous mode (24 crashes) and conventional mode (22 crashes). The ARM algorithm was employed to uncover significant relationships and patterns in the crash data. A total of 67 association rules were generated across three distinct scenarios—intersections, within 150?ft of intersections, and midblock locations—revealing key associations between factors such as time of day, location, vehicle and pedestrian behavior, and environmental conditions. The study’s findings offer valuable insights into pedestrian safety in the context of precrash modes of AVs and provide important guidance for developing targeted safety measures and policies to reduce pedestrian–AV collisions.]]></description>
      <pubDate>Mon, 09 Feb 2026 08:43:03 GMT</pubDate>
      <guid>https://trid.trb.org/View/2665686</guid>
    </item>
    <item>
      <title>Uncertainty-aware spatiotemporal interaction learning for pre-conflict risk evolution with a risk-increase prior</title>
      <link>https://trid.trb.org/View/2652425</link>
      <description><![CDATA[Quantifying real-time conflict risk and revealing its evolution are of great importance for enhancing vehicle active safety. Recent studies estimate dynamic risk via conflict probability, yet annotation still relies on threshold based static views and uncertainty is only partially modeled, which limits the assessment of a model’s ability to learn conflict evolution. Addressing this gap, the authors posit a prior hypothesis of increasing pre-conflict risk and develop a risk quantification model that integrates driver control inputs and multi vehicle spatiotemporal interactions with explicit uncertainty outputs. The model is evaluated for accuracy and stability of risk perception, parameter sensitivity, and capacity for pattern learning. Experiments show that, relative to Time To Collision (TTC), Deceleration Rate to Avoid a Crash (DRAC), Proportion of Stopping Distance (PSD), Anticipated Collision Time (ACT), and Emergency Index (EI), the proposed model achieves stronger risk discrimination. On the test set of 15 conflict events used in this study, the proposed model detects elevated conflict risk on average 1.15 s before the conflict point. In four representative scenarios, including car following, ego lane change, unobstructed cut in and cut in under occluded view, the proposed model yields a lower false alarm rate than TTC and, on average, perceives rising conflict risk 1.44 s before the conflict point. Uncertainty analysis indicates lower uncertainty during the rising risk phase, enabling reliable capture of risk evolution. Sensitivity results support the expressiveness of the proposed hypothesis and reveal a common regularity across scenarios, where risk begins to increase approximately 4–6 s before conflict. The results establish a pre conflict risk modeling paradigm that jointly estimates risk and its confidence, supports calibration and transfer across scenarios, and provides an operational basis for proactive safety assessment.]]></description>
      <pubDate>Fri, 06 Feb 2026 08:45:32 GMT</pubDate>
      <guid>https://trid.trb.org/View/2652425</guid>
    </item>
    <item>
      <title>ROAR: Robust accident recognition and anticipation for autonomous driving</title>
      <link>https://trid.trb.org/View/2655846</link>
      <description><![CDATA[Accurate accident anticipation is essential for enhancing the safety of autonomous vehicles (AVs). However, existing methods often assume ideal conditions, overlooking challenges such as sensor failures, environmental disturbances, and data imperfections, which can significantly degrade prediction accuracy. Additionally, previous models have not adequately addressed the considerable variability in driver behavior and accident rates across different vehicle types. To overcome these limitations, this study introduces ROAR, a novel approach for accident detection and prediction. ROAR combines Discrete Wavelet Transform (DWT), a self-adaptive object-aware module, and dynamic focal loss to tackle these challenges. The DWT effectively extracts features from noisy and incomplete data, while the object-aware module improves accident prediction by focusing on high-risk vehicles and modeling the spatial–temporal relationships among traffic agents. Moreover, dynamic focal loss mitigates the impact of class imbalance between positive and negative samples. Evaluated on three widely used datasets — Dashcam Accident Dataset (DAD), Car Crash Dataset (CCD), and AnAn Accident Detection (A3D) — the model consistently outperforms existing baselines in key metrics such as Average Precision (AP) and mean Time-to-Accident (mTTA). These results demonstrate the model’s robustness in real-world conditions, particularly in handling sensor degradation, environmental noise, and imbalanced data distributions. This work offers a promising solution for reliable and accurate accident anticipation in complex traffic environments.]]></description>
      <pubDate>Wed, 04 Feb 2026 17:05:19 GMT</pubDate>
      <guid>https://trid.trb.org/View/2655846</guid>
    </item>
    <item>
      <title>Collision risk identification and prediction considering heterogeneous braking patterns using large-scale pre-collision trajectories</title>
      <link>https://trid.trb.org/View/2630627</link>
      <description><![CDATA[Rear-end collisions often occur in vehicles when successive braking events lead to insufficient deceleration by one or more vehicles. This paper proposed a novel method for collision risk identification and prediction by integrating the braking dynamics of both leading and following vehicles in pre-crash scenarios. Using a piecewise linear model of deceleration profiles, 45 collision risk moments were identified across 10 scenarios based on ten kinematic parameters. A novel critical time-to-collision metric was proposed to integrate both the timing and execution of braking behaviors into collision risk prediction. To account for driver heterogeneity in deceleration, deceleration rate, and reaction time, Gaussian mixture regression was used to perform conditional inference to estimate braking pattern parameters and generate interval-valued crash risk predictions. The performance and optimal threshold were validated using large-scale vehicle trajectories from collision and non-collision events. The results demonstrate that the collision risk moments vary with both the braking timing and execution of the leading and following vehicles. The proposed metric outperformed traditional surrogate safety measures in predicting collision risk, consistently yielding higher accuracy with less variability across pre-crash time intervals. These findings indicate that the estimated critical time-to-collision is a reliable and effective measure for collision avoidance systems and advanced driver assistance technologies.]]></description>
      <pubDate>Tue, 23 Dec 2025 09:29:17 GMT</pubDate>
      <guid>https://trid.trb.org/View/2630627</guid>
    </item>
    <item>
      <title>In-depth investigation into the hierarchical causal chain of fatal crashes between vulnerable road users and single motor vehicle</title>
      <link>https://trid.trb.org/View/2625336</link>
      <description><![CDATA[Crash pattern recognition and characterization are essential for reducing the damage vulnerable road users (VRUs) suffer in motor vehicle crashes. However, traditional methods provide an incomprehensive understanding of crash causality and the impacts of VRU-vehicle interactions. Therefore, this study aims to provide a reasonable causality for various types of crashes. To achieve this goal, a three-layer causal analysis framework was developed. The layers consist of physical states (mainly environmental and human factors), interactions (pre-crash behaviors of drivers and VRUs), and crashes. First, latent class cluster analysis and sequence analysis were used to identify the interactive behavior patterns and crash patterns of VRU-vehicle pairs, respectively. Besides, an oversampling algorithm was proposed to assist the Granger causality test in uncovering the latent causal relationships between pre-crash behavior patterns and crash patterns. Finally, Sankey diagrams were utilized to compare and analyze the crash path. The results show that single and consecutive crashes have nine and eleven typical scenarios, respectively, excluding considering the potential causal chains. These potential causal chains provide nine new scenarios. It was found that personal subjective factors primarily influence pre-crash behavior of drivers, while for VRUs, the traffic environment plays a crucial role. Noteworthily, the highest crash risk was only associated with the causal chain where vehicles are unable to brake in time. Clarifying the causal relationships between VRU-vehicle interaction and crash is essential, which can help finding the critical causes of fatal crashes. The analysis identified VRU violations and the inability of vehicles to brake in time as critical determinants of crash severity in both single and consecutive crash scenarios. Accordingly, targeted safety interventions were proposed, including enhancements to pedestrian crossing infrastructure and improvements in vehicle braking systems to mitigate crash risk.]]></description>
      <pubDate>Thu, 18 Dec 2025 15:37:24 GMT</pubDate>
      <guid>https://trid.trb.org/View/2625336</guid>
    </item>
    <item>
      <title>Characteristics analysis of autonomous vehicle pre-crash scenarios</title>
      <link>https://trid.trb.org/View/2618082</link>
      <description><![CDATA[To date, hundreds of crashes have occurred in open-road testing of autonomous vehicles (AVs), highlighting the need for improving AV reliability and safety. However, current studies predominantly analyze crash data based on oversimplified classification schemes that lack clear scenario definitions. Consequently, they impede an in-depth investigation of crash characteristics. Pre-crash scenario typology classifies crashes based on vehicle dynamics and kinematics features. Building on this, characteristics analysis can identify similar features under comparable crashes, offering a more effective reflection of general crash patterns and providing more targeted recommendations for enhancing AV performance. In this paper, the authors initially collected the latest 774 California AV crash reports, then selected 384 autonomous mode crashes, and used the newly revised pre-crash scenario typology to identify AV pre-crash scenarios. To improve the efficiency of scenario identification and adaptability to future updates in scenario typology, the authors proposed a set of mapping rules to extract pre-crash scenarios automatically. The authors successfully identified 27 types of AV pre-crash scenarios with an accuracy of 98.1%. Through detailed analysis, the authors obtained two key groups of AV pre-crash scenarios: rear-end scenarios and intersection scenarios. Based on the abundance of crash data, the authors adopted different analysis methods to analyze the features of key scenarios. Association analysis of rear-end scenarios showed that the significant environmental influencing factors were traffic control type, location type, light, etc. For intersection scenarios prone to severe crashes with detailed descriptions, the authors employed causal analysis to obtain the significant causal factors: habitual violations and temporary obstruction of view. The extracted scenarios in this paper and their features can assist in constructing the AV simulation test with precise environmental parameters and realistic interactions with other traffic parties. The resulting optimization recommendations can inform regulators and reveal control-algorithm weaknesses across diverse real-world conditions, thereby enhancing the AV safety.]]></description>
      <pubDate>Mon, 24 Nov 2025 10:19:56 GMT</pubDate>
      <guid>https://trid.trb.org/View/2618082</guid>
    </item>
    <item>
      <title>Evaluation of Comfort Zone Boundary Based Automated Emergency Braking Algorithms for Car-to-Powered-Two-Wheeler Crashes in China</title>
      <link>https://trid.trb.org/View/2448775</link>
      <description><![CDATA[Crashes between cars and powered two-wheelers (PTWs: motorcycles, scooters, and e-bikes) are a safety concern; as a result, developing car safety systems that protect PTW riders is essential. While the pre-crash protection system automated emergency braking (AEB) has been shown to avoid and mitigate injuries for car-to-car, car-to-cyclist, and car-to-pedestrian crashes, much is still unknown about its effectiveness in car-to-PTW crashes. Further, the characteristics of the crashes that remain after the introduction of such systems in traffic are also largely unknown. This study estimates the crash avoidance and injury risk reduction performance of six different PTW-AEB algorithms that were virtually applied to reconstructed car-to-PTW pre-crash kinematics extracted from a Chinese in-depth crash database. Five of the algorithms include combinations of drivers? and PTW riders? comfort zone boundaries for braking and steering, while the sixth is a traditional AEB. Results show that the average safety performance of the algorithms using only the driver's comfort zone boundaries is higher than that of the traditional AEB algorithm. All algorithms resulted in similar distributions of impact speed and impact locations, which means that in-crash protection systems likely can be made less complex, not having to consider differences in AEB algorithm design among car manufacturers.]]></description>
      <pubDate>Fri, 21 Nov 2025 08:44:21 GMT</pubDate>
      <guid>https://trid.trb.org/View/2448775</guid>
    </item>
    <item>
      <title>Parametric Study of Pre-Crash Vehicle Maneuvers and Occupant Safety Performance Response</title>
      <link>https://trid.trb.org/View/2625398</link>
      <description><![CDATA[This report addresses the influence of pre-crash vehicle maneuvers on the injury risks of front passengers during a frontal crash. Seat position and occupant characteristics including anthropometry, sex, and age were also included in the developed design of experiments, which consisted of 55 finite element (FE) crash simulations. A generic buck vehicle model was developed based on a publicly available FE model of a 2014 Honda Accord, which included the vehicle interior and the front passenger air bag (PAB). The PAB model was validated in a drop- tower test, which was designed by the manufacturer, and a reasonable comparison was obtained for oblique side impacts using a Test Device for Human Occupant Restraint-M50, or THOR-M50, in both driver and passenger seats and for frontal impacts using a Hybrid-, or H-, III M50 in the driver seat and H-III F05 in the passenger seat. A generic 1 g braking and turning-and-braking pulse were used as two different pre-crash maneuvers using specific rigid-body human models (Global Human Body Models Consortium [GHBMC] models) with active joints, called the GHBMCsi-pre models. Then, the kinematics data was transferred using a developed switch algorithm to simplified deformable human models (the GHBMCsi models), and a Federal Motor Vehicle Safety Standard No. 208 pulse was applied to simulate the in-crash phase. Injury metrics were recorded for all the belted GHBMCsi models to evaluate injury risks. Seat recline angle and seat track position showed the highest influence on the head injury criteria, or HIC. The brain injury criteria, or BrIC, and neck injury criteria, or Nij, were most sensitive to pre-crash maneuver type, seat recline angle, and occupant model size.]]></description>
      <pubDate>Thu, 20 Nov 2025 16:48:31 GMT</pubDate>
      <guid>https://trid.trb.org/View/2625398</guid>
    </item>
    <item>
      <title>The effect of time-extended evasive swerving maneuvers on occupants’ bracing strategies</title>
      <link>https://trid.trb.org/View/2617060</link>
      <description><![CDATA[Occupant bracing behavior in pre-crash maneuvers has been previously investigated but the effect of the duration of the pre-crash maneuver on bracing is unknown. This is critical to understand as time-extended pre-crash maneuvers may emerge in cases where drivers lose control of a vehicle and in autonomous vehicles as they may take different approaches to avoid crashes than the current vehicles. Therefore, the aim of this study was to understand the effect of pre-crash maneuver duration on child and adult occupants’ bracing behavior and resulting kinematics. Forty seatbelt restrained subjects (9–40 years old) experienced sled-simulated time-extended lateral swerving maneuvers (8 s, 4 cycles, peak acceleration 0.7 g) producing an alternating motion initially out-of-the-belt, followed by into-the-belt for each cycle. In a braced condition, subjects were instructed to hold on to a laterally placed handle with their right hand before the maneuver onset, while in an unbraced condition no instructions were given. A 3D-motion capture system, electromyography (EMG), and seatbelt load cells captured head and trunk kinematics (normalized to seated height), muscle activation (normalized to maximum voluntary isometric contraction, MVIC), and seatbelt reaction forces (normalized to body weight), respectively. The effects of cycles and interaction with bracing and age on peak lateral head and trunk displacement into- and out-of-the belt were examined with Mixed-Effects Models and Tukey’s post-hoc tests (p ≤ 0.05). Out-of-the-belt peak lateral head and trunk displacements were the greatest in the first cycle and the smallest in the second cycle (p<0.01). The third and four cycles were not significantly different from one another (p>0.8). Into-the-belt peak lateral head and trunk displacements were smaller in the first cycle than the remaining cycles (p<0.001) and were not significantly different across the remaining cycles (p>0.8). No interactions between cycle, age and bracing were found (p>0.3). Right bicep, trapezius and rectus femoris activations slightly increased with increasing cycles in the unbraced condition and in the into-the-belt direction for the 9–11 year-old group. Out-of-belt seat belt loads increased with increasing cycles in the unbraced condition for all age groups. Occupant kinematics as a result of their bracing behavior changed across cycles of swerving maneuvers from an exaggerated displacement in cycle 1 to an overcompensation due to bracing in cycle 2, ending with a plateau of a moderate displacement in cycle 3 and 4. Younger children (age 9–11) took longer to adapt to the oscillatory motion as they increased their muscle activation over time unlike the other age groups. These findings suggest that it may take time for occupants to find the optimal bracing strategy in time-extended maneuvers. Furthermore, children may find challenging to calibrate their bracing response overtime from a neuromotor perspective.]]></description>
      <pubDate>Wed, 19 Nov 2025 17:09:39 GMT</pubDate>
      <guid>https://trid.trb.org/View/2617060</guid>
    </item>
    <item>
      <title>How do motorists’ pre-crash behaviors contribute to the injury severity of police officers? Using interpretable machine learning to untangle the behavioral pathways in crashes involving police vehicles</title>
      <link>https://trid.trb.org/View/2569856</link>
      <description><![CDATA[Police officers are integral to enforcing traffic laws and providing assistance to motorists. While performing their duties on the road or roadside, they encounter significant hazards, many of which arise from the negligent or inappropriate behaviors of drivers. Despite the prevalence of these risks, there is a paucity of research specifically examining the outcomes of traffic crashes involving police officers or police vehicles, particularly in relation to the injury severity sustained by police officers in such incidents. The objective of this study is to explore the roles of motorist behaviors in police-involved crashes. Specifically, using police-involved crashes data from 2017 to 2021 in Alabama, this study employs a path analysis framework integrated with interpretable machine learning to examine the behavioral pathways from contributing factors to motorists’ pre-crash behaviors, ultimately leading to the injury severity of police officers. The path analysis allows the identification of various factors that directly contribute to injury severity and those that indirectly contribute to injury severity through their relationships with motorists’ pre-crash behaviors. The results indicate that several factors are directly associated with policy injury severities, including motorists’ pre-crash behaviors (such as failure to yield, driving under the influence (DUI), and traffic signal violations) that are significantly correlated with increased injury severity for police officers involved in crashes. Further, these pre-crash behaviors are associated with contributing factors including vehicle speeds, lighting conditions, seatbelt usage, and weather conditions. Thus, these contributing factors serve as significant predictors of motorists’ pre-crash behaviors, indirectly impacting police injury severity through their influence on unsafe driving practices. This study provides valuable insights into the behavioral pathway leading to police injury severities. By understanding the direct and indirect contributors to police injury severity, policymakers can develop targeted interventions to address these key risk factors and improve overall road safety for police officers.]]></description>
      <pubDate>Wed, 20 Aug 2025 11:57:17 GMT</pubDate>
      <guid>https://trid.trb.org/View/2569856</guid>
    </item>
    <item>
      <title>Effects of early notifications on driver responses to lateral collision warnings</title>
      <link>https://trid.trb.org/View/2567226</link>
      <description><![CDATA[Despite the demonstrated effectiveness of advanced driver assistance systems, lateral collisions still occur because of incorrect and deferred responses of drivers to potential hazards attributed to approaching vehicles from the sides. In such a situation, this study introduces early notifications prior to lateral collision warnings, displayed in either a visual-only format or a visual-auditory format. These notifications inform drivers about the presence of approaching vehicles from behind in adjacent lanes, aiming to improve their performance after receiving lateral collision warnings. The improvement is measured by analyzing the differences in drivers’ reaction times and exhibition of appropriate reactions when early notifications are present and absent. Additionally, the effects of confounding factors including display formats (visual-only versus visual-auditory), drivers’ socio-demographics, and driving experience on drivers’ reaction times and exhibition of appropriate reactions are considered in the binary logit and Tobit models. Furthermore, the effects of individual heterogeneity are accounted using the random parameters approach. Results indicate that there is significant heterogeneity in drivers’ reaction time when early notifications is present, prior to lateral collision warnings. Even that drivers’ reaction time may increase in some circumstances, their likelihood of exhibiting appropriate reactions increases. However, it is interesting to find that drivers’ reaction time increases and their likelihood of exhibiting appropriate reactions decreases when early notifications are presented in a visual-auditory format, compared to a visual-only format.]]></description>
      <pubDate>Tue, 22 Jul 2025 14:39:51 GMT</pubDate>
      <guid>https://trid.trb.org/View/2567226</guid>
    </item>
    <item>
      <title>Research on drivers’ hazard perception and visual characteristics before vehicle-to-powered two-wheeler collisions</title>
      <link>https://trid.trb.org/View/2569161</link>
      <description><![CDATA[Understanding drivers’ hazard perception levels and visual behavior in conflict scenarios is crucial for improving traffic safety and advancing intelligent driving systems, especially given the growing complexity of traffic conditions and the rapid evolution of intelligent driving technologies. This study examines typical near-collision scenarios involving vehicles and powered two-wheelers, focusing on the effects of collision scenarios, driving states, and risk conditions on drivers’ hazard perception and visual characteristics. Using quantile regression and generalized linear mixed models, the study quantitatively assesses how these factors influence hazard perception and visual behavior, uncovering the visual response mechanisms underlying hazard perception. The results reveal that different vehicle-to-powered two-wheeler collision scenarios significantly affect drivers’ hazard perception and visual behavior. Drivers exhibited higher hazard perception levels and collision avoidance success rates in “Crossing from Right” and “Cut-in from Right” scenarios, whereas lower hazard perception abilities were observed in “Crossing from Left” and “Oncoming” scenarios. Fatigue was shown to severely impair drivers’ alertness and visual search abilities, resulting in diminished hazard perception levels. Under high-risk conditions, while drivers exhibited reduced collision avoidance success rates, their heightened attention and vigilance toward powered two-wheeler enhanced hazard perception. Besides, the study also highlights a strong correlation between visual characteristics and drivers’ hazard perception. These findings are significant for understanding the mechanisms underlying drivers’ hazard perception in intersection scenarios and may provide a scientific basis for future developments in human–machine collaborative monitoring and intelligent traffic safety strategies.]]></description>
      <pubDate>Tue, 22 Jul 2025 14:39:51 GMT</pubDate>
      <guid>https://trid.trb.org/View/2569161</guid>
    </item>
    <item>
      <title>Using near misses, artificial intelligence, and machine learning to predict maritime incidents: A U.S. Coast Guard case study</title>
      <link>https://trid.trb.org/View/2543883</link>
      <description><![CDATA[This article reports on a U.S. Coast Guard case study that used data from near misses, artificial intelligence (AI), and machine learning to predict maritime incidents.  The authors note that it has become accepted that near misses can be predictors of future negative events.  The U.S. Coast Guard (USCG) uses a reporting system called the Marine Information for Safety and Law Enforcement (MISLE) database.  Many of the events that are included in this database are minor but might prove useful as predictors of future events.  Using the analysis that showed this was indeed the case, the authors built a predictive AI model that included data on waterway type by vessel combination.  The R-based predictive model was designed to be run monthly, using data from prior months to make future predictions. The authors used data from 2007 to 2022 to train the predictive models.  They determined that the overall accuracy of the predictions generated by this model was 92%–99.9%.   The authors conclude that AI predictive models may potentially be trained on data from incident reports in real time, with the aim of increasing safety by predicting where future serious incidents are likely to occur.]]></description>
      <pubDate>Fri, 11 Jul 2025 10:00:27 GMT</pubDate>
      <guid>https://trid.trb.org/View/2543883</guid>
    </item>
  </channel>
</rss>