<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>How to improve the AV safety: From the understanding of AV crashes to safety enhancement strategy</title>
      <link>https://trid.trb.org/View/2661857</link>
      <description><![CDATA[Autonomous vehicles (AVs) are increasingly prioritized in national transport strategies for their potential to improve safety and sustainability. Yet real-world crashes show that AVs remain vulnerable to safety risks, highlighting the need to understand their crash patterns and develop evidence-based countermeasures. The objective of this study is to investigate the factors associated with AV crashes and develop safety countermeasure to improve AV safety. 335 AV crash data from 2021 to 2022 was collected from collision reports from San Francisco at Traffic Analysis Zones (TAZ) level. Sociodemographic, built environment, land-use, and exposure variables were incorporated into Ordinary Least Squares (OLS) and Geographically Weighted Regression (GWR) models. Compared with OLS, the GWR model provides a superior fit, as reflected by its lower AIC (535.17) and residual sum of squares (580.66). Model results revealed diverse effects of build environment on AV crashes across TAZs. Specifically, bus stop and transit lane densities exhibit strong negative associations with crash frequency, particularly in the southwestern regions. Bicycle parking density is negatively correlated with crashes. In contrast, wider sidewalks and higher proportions of speed limit zones are positively associated with AV crashes in certain urban areas. The impact of traffic signal density is spatially inconsistent—showing a crash-reducing effect in northeastern urban areas but a positive association in southwestern regions. Safety countermeasures were proposed from the perspective of understanding the AV crash influencing factors. The study underscores the significance of well-planned transportation facilities in enhancing AV safety.]]></description>
      <pubDate>Thu, 30 Apr 2026 16:38:38 GMT</pubDate>
      <guid>https://trid.trb.org/View/2661857</guid>
    </item>
    <item>
      <title>Management measures to improve safety of partially automated vehicles on rural roads using artificial neural networks</title>
      <link>https://trid.trb.org/View/2661845</link>
      <description><![CDATA[Concerns regarding the safety of partial automated vehicles (AVs) remain prevalent, especially in complex roadway environments. AVs incorporate Advanced Driving Assistance Systems to perform the dynamic driving task; however, unexpected disengagements of the systems still occur due to various contextual and infrastructural factors. Among these, two-lane rural roads—with their geometric design and operational challenges—represent a critical setting. Notably, a high number of disengagements are concentrated on horizontal curves.This study analyzes naturalistic disengagement data from two SAE Level 2 AVs operating on different segments of two-lane rural roads. The horizontal curves were characterized across geometrical and operational variables (radii, curvature change rate (CCR), curve direction, speed or visibility). These variables served as inputs to an artificial neural network (ANN) model designed to predict disengagement occurrences.The ANN, a feedforward multilayer perceptron with one hidden layer, was trained using backpropagation. Performance was validated with K-fold cross-validation, and accuracy assessed via cross-entropy loss and confusion matrices. A Monte Carlo-style simulation tested robustness by generating multiple confusion matrices from randomized data partitions to evaluate classification stability. The results highlight CCR and lane width as key predictive factors. The calibrated ANN demonstrated robust classification (accuracy = 87.8 %, sensitivity = 92.7 %, specificity = 85.9 %) in identifying curve segments with a higher likelihood of disengagement.This study provides road administrations a new neural network derived empirical formula to identify potential AV disengagement zones. By identifying risk-prone areas, authorities can consider targeted measures—such as enhanced signage or driver alerts— to support safer and more efficient automated driving in rural settings.]]></description>
      <pubDate>Thu, 30 Apr 2026 16:38:38 GMT</pubDate>
      <guid>https://trid.trb.org/View/2661845</guid>
    </item>
    <item>
      <title>An Interpretable Physics-Constrained Deep Operator Network for Battery Internal Temperature Estimation With Limited Data</title>
      <link>https://trid.trb.org/View/2659194</link>
      <description><![CDATA[Accurate internal temperature estimation is crucial for developing effective safety management strategies in battery energy storage systems (BESSs) and electric vehicles (EVs). Traditional model-based methods often suffer from complexities in model parameters, while data-driven approaches heavily rely on large amounts of training data and lack theoretical interpretability. To address these challenges, this article proposes an interpretable physics-constrained deep operator network (PC-DeepONet) for internal temperature estimation of batteries. By integrating data loss from DeepONet with physical loss from a lumped-thermal model, the network offers strong theoretical interpretability, aligning better with the underlying thermal dynamics and significantly reducing data dependence. Additionally, battery thermal model parameters are innovatively embedded into the network as trainable components and optimized via backpropagation (BP), enhancing the network’s adherence to physical constraints. To address the imbalance between the losses, a log dynamic weight averaging (LDWA) method is employed to reduce the scale disparity and dynamically balance the weights. The experimental results demonstrate that, despite using limited training data at 25 °C, the proposed method exhibits enhanced stability and higher accuracy under various operating conditions and different temperatures. Compared to traditional methods, the estimated root-mean-square error (RMSE) can be reduced by 59.5% at 0 °C and 35.9% at 40 °C, respectively.]]></description>
      <pubDate>Thu, 30 Apr 2026 11:28:21 GMT</pubDate>
      <guid>https://trid.trb.org/View/2659194</guid>
    </item>
    <item>
      <title>TraCR Foundational Project: TraCR Collective Transportation Cybersecurity Testbeds</title>
      <link>https://trid.trb.org/View/2697454</link>
      <description><![CDATA[The National Center for Transportation Cybersecurity and Resiliency's (TraCR's)
foundational project aims to develop technological tools, prototypes, testing platforms, and facilities to ensure the cybersecurity and cyber-resilience of multimodal transportation systems and related infrastructure. The project is led by Clemson University (Clemson) under the strategic direction of Dr. Ronnie Chowdhury (Lead PI), with coordination support from Dr. Sabbir Salek (Co-PI), and involves all eight other TraCR partner institutions organized into four subgroups. A structured project governance framework, including biweekly subgroup meetings, monthly full-team coordination meetings, quarterly progress reporting and advisory board engagement, ensures alignment with project milestones, integration across teams, and effective monitoring of technical progress and deliverables. 

Clemson collaborates with Benedict College (Benedict), South Carolina State University (SCSU), and the University of Texas at Dallas (UTD) to advance a comprehensive, automated threat modeling capability for multimodal transportation systems. Building on the Transportation Cybersecurity and Resiliency Threat Modeling Framework (TraCR-TMF), the team conducts testbed-in-the-loop evaluations within Clemson’s real-world cybersecurity testbed, implementing digital-twin-based cybersecurity analysis of in-vehicle networks, and engaging state transportation agencies to assess operational transferability. Additionally, the team will work to integrate graph-based reasoning models into threat modeling, deploy supervised ModernBERT classifiers, and align with the MITRE Embedded Systems Threat Matrix to strengthen structured system-to-vulnerability mapping and improve threat coverage across transportation cyber-physical systems.

The other partner institutions will develop additional real-world and virtual testing platforms to support cybersecurity experimentation for multimodal transportation. Florida International University (FIU) and the University of Alabama at Tuscaloosa (UA) are jointly advancing the Open-Source Connected and Automated Mobility Co-Simulation (OpenCAMS) environment and related simulation platforms, integrating SUMO, CARLA, and network simulation tools, to evaluate privacy-aware multimodal large language models and post-quantum-secure C-V2X communications. Their efforts further include the development and validation of spoofing attack models targeting Basic Safety Message transmissions and multi-frequency GPS receivers, as well as investigations into backdoor-resilient perception systems and the security of vision-language models for intelligent transportation applications.

Purdue University (Purdue) and the University of California, Santa Cruz (UCSC) are advancing adversarial testing methodologies through integrated physical-virtual experimentation frameworks that combine miniature autonomous vehicle testbeds, CARLA/METS-R simulation coupling, and scenario-based vulnerability discovery. These activities include simulation-to-real validation of perception and traffic signal spoofing attacks, evaluation of V2X safety message vulnerabilities, cybersecurity analysis of shared micromobility Bluetooth pairing protocols, implementation of lightweight post-quantum cryptographic protections for vulnerable road user beacons, and closed-loop security assessments of traffic signal controller infrastructures, along with investigations of secure multimodal AI agents and memory-augmented reasoning architectures for autonomous robotic transportation systems.

In addition, Morgan State University (MSU) is enhancing its connected vehicle cybersecurity experimentation capabilities by developing replay-attack models targeting C-V2X onboard units and evaluating mitigation strategies in its real-world testbed environment, in collaboration with Clemson. These efforts quantify communication-level impacts on safety-critical applications and support the development of deployable countermeasures to strengthen resilience against wireless attack vectors affecting connected transportation infrastructure.
]]></description>
      <pubDate>Thu, 30 Apr 2026 12:19:22 GMT</pubDate>
      <guid>https://trid.trb.org/View/2697454</guid>
    </item>
    <item>
      <title>A Novel Hybrid Attack Model and A Quantum-Infused Hybrid Defense Method for Resilient Perception of Autonomous Vehicles</title>
      <link>https://trid.trb.org/View/2697419</link>
      <description><![CDATA[This project strengthens the cybersecurity and resiliency of camera-based perception for autonomous vehicles by addressing two fast-growing attack classes: universal adversarial perturbations (UAPs) and generative/deepfake-style scene manipulations that can add, alter, or remove objects in the camera feed. The team will first build and validate a novel hybrid attack that combines image-agnostic UAP noise with generative object-disappearance attacks (ODAs) using real-time inpainting to create “hallucination” driving scenes where objects are misclassified or vanish entirely. The project will also develop a quantum-enhanced hybrid defense that fuses parameterized quantum circuits with classical deepfake/manipulation detection, leveraging quantum–classical disagreement and out-of-distribution signals to robustly detect both pixel-level perturbations and semantic object edits. The project will produce deployable prototypes: (1) a real-time hybrid “malware” attack pipeline and (2) a quantum-infused hybrid detector, which will be evaluated in realistic AV scenarios and deployed for testing on connected-vehicle testbeds (e.g., Clemson University Connected Vehicle Testbed or CU-CVT and Morgan State).

]]></description>
      <pubDate>Thu, 30 Apr 2026 12:17:12 GMT</pubDate>
      <guid>https://trid.trb.org/View/2697419</guid>
    </item>
    <item>
      <title>Multi-Sensor Fusion-Based Highway Vehicle Behavior Prediction and Its Application in Adaptive Cruise Control</title>
      <link>https://trid.trb.org/View/2659100</link>
      <description><![CDATA[We propose a vehicle behavior prediction approach using multi-sensor data fusion to address the challenges of high-level autonomous driving on expressways and deploying on vehicles with limited computation hardware capabilities. The proposed approach analyzes the interactions between the ego-vehicle and background vehicles to extract critical features such as congestion levels and occupancy states. After that, these features are fed into a Long Short-Term Memory (LSTM) model to predict the behaviors of key targets within the ego-vehicle's vicinity. This sequential fusion framework incorporating the hierarchical target association significantly reduces the computational intensity and memory requirements. Experimental results demonstrate that the proposed vehicle behavior prediction approach reduces the time required for background vehicle behavior analysis and enhances reliability and safety by providing valuable insights for defensive driving and planning strategies. Additionally, the ACC real-vehicle test incorporating the proposed behavior prediction reveals that the system can anticipate and respond appropriately to potential hazards, maintaining a safe distance between the ego-vehicle and surrounding traffic, thereby contributing to safer and more efficient autonomous driving.]]></description>
      <pubDate>Wed, 29 Apr 2026 09:10:08 GMT</pubDate>
      <guid>https://trid.trb.org/View/2659100</guid>
    </item>
    <item>
      <title>AI Safety Assurance for Automated Vehicles: A Survey on Research, Standardization, Regulation</title>
      <link>https://trid.trb.org/View/2659120</link>
      <description><![CDATA[Assuring safety of artificial intelligence (AI) applied to safety-critical systems is of paramount importance. Especially since research in the field of automated driving shows that AI is able to outperform classical approaches, to handle higher complexities, and to reach new levels of autonomy. At the same time, the safety assurance required for the use of AI in such safety-critical systems is still not in place. Due to the dynamic and far-reaching nature of the technology, research on safeguarding AI is being conducted in parallel to AI standardization and regulation. The parallel progress necessitates simultaneous consideration in order to carry out targeted research and development of AI systems in the context of automated driving. Therefore, in contrast to existing surveys that focus primarily on research aspects, this paper considers research, standardization and regulation in a concise way. Accordingly, the survey takes into account the interdependencies arising from the triplet of research, standardization and regulation in a forward-looking perspective and anticipates and discusses open questions and possible future directions. In this way, the survey ultimately serves to provide researchers and safety experts with a compact, holistic perspective that discusses the current status, emerging trends, and possible future developments.]]></description>
      <pubDate>Wed, 29 Apr 2026 09:10:07 GMT</pubDate>
      <guid>https://trid.trb.org/View/2659120</guid>
    </item>
    <item>
      <title>KoMA: Knowledge-Driven Multi-Agent Framework for Autonomous Driving With Large Language Models</title>
      <link>https://trid.trb.org/View/2659111</link>
      <description><![CDATA[Large language models (LLMs) as autonomous agents offer a novel avenue for tackling real-world challenges through a knowledge-driven manner. These LLM-enhanced methodologies excel in generalization and interpretability. However, the complexity of driving tasks often necessitates the collaboration of multiple, heterogeneous agents, underscoring the need for such LLM-driven agents to engage in cooperative knowledge sharing and cognitive synergy. Despite the promise of LLMs, current applications predominantly center around single-agent scenarios. To broaden the horizons of knowledge-driven strategies and bolster the generalization capabilities of autonomous agents, we propose the KoMA framework consisting of multi-agent interaction, multi-step planning, shared-memory, and ranking-based reflection modules to enhance multi-agents' decision-making in complex driving scenarios. Based on the framework's generated text descriptions of driving scenarios, the multi-agent interaction module enables LLM agents to analyze and infer the intentions of surrounding vehicles based on scene information, akin to human cognition. The multi-step planning module enables LLM agents to analyze and obtain final action decisions layer by layer to ensure consistent goals for short-term action decisions. The shared memory module can accumulate collective experience to make superior decisions, and the ranking-based reflection module can evaluate and improve agent behavior with the aim of enhancing driving safety and efficiency. The KoMA framework not only enhances the robustness and adaptability of autonomous driving agents but also significantly elevates their generalization capabilities across diverse scenarios. Empirical results demonstrate the superiority of our approach over traditional methods, particularly in its ability to handle complex, unpredictable driving environments without extensive retraining.]]></description>
      <pubDate>Wed, 29 Apr 2026 09:10:07 GMT</pubDate>
      <guid>https://trid.trb.org/View/2659111</guid>
    </item>
    <item>
      <title>Identifying Research Gaps Through Self-Driving Car Data Analysis</title>
      <link>https://trid.trb.org/View/2659129</link>
      <description><![CDATA[There are currently around thirty companies testing self-driving cars in San Francisco, CA, effectively creating a living laboratory. Of these companies, only Waymo is engaged in commercial operations, while Zoox conducts routine driverless testing operations in San Francisco. Despite these successes, federal investigations have been opened into both companies for safety concerns, and Cruise is attempting to reinstate its permit after a near-fatal pedestrian crash. An analysis of these three companies’ crash data from required reporting illustrates that many areas of self-driving need improvement. The most significant crash type for Waymo and Zoox are struck-from-behind events, while Cruise struggled most with unexpected actions by others. Computer vision systems are very brittle and likely play an outsized role in crashes. Self-driving cars also struggle to reason under uncertainty, and simulations are not effectively bridging the physical-to-real-world testing gap. This analysis underscores that research is lacking, especially for artificial intelligence involving computer vision and reasoning under uncertainty.]]></description>
      <pubDate>Wed, 29 Apr 2026 09:10:07 GMT</pubDate>
      <guid>https://trid.trb.org/View/2659129</guid>
    </item>
    <item>
      <title>Closing the Calibration Gap: A Real-Time Multi-Modal Fusion Framework for 3D Semantic Segmentation</title>
      <link>https://trid.trb.org/View/2659127</link>
      <description><![CDATA[LiDAR and camera are two critical sensors for multi-modal 3D semantic segmentation and are supposed to be fused efficiently and robustly to promise safety in various real-world scenarios. However, existing multi-modal methods face two key challenges: 1) difficulty with efficient deployment and real-time execution; and 2) drastic performance degradation under weak calibration between LiDAR and cameras. To address these challenges, we propose CPGNet-LCF, a new multi-modal fusion framework extending the LiDAR-only CPGNet. CPGNet-LCF solves the first challenge by inheriting the easy deployment and real-time capabilities of CPGNet. For the second challenge, we introduce a novel weak calibration knowledge distillation strategy during training to improve the robustness against the weak calibration. CPGNet-LCF achieves state-of-the-art performance on the nuScenes and SemanticKITTI benchmarks. Remarkably, it can be easily deployed to run in 20 ms per frame on a single Tesla V100 GPU using TensorRT TF16 mode. Furthermore, we benchmark performance over four weak calibration levels, demonstrating the robustness of our proposed approach.]]></description>
      <pubDate>Wed, 29 Apr 2026 09:10:07 GMT</pubDate>
      <guid>https://trid.trb.org/View/2659127</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>Context-Aware and Reliable Long-Term Decision-Making for Safe Intelligent Vehicles: A Survey</title>
      <link>https://trid.trb.org/View/2659156</link>
      <description><![CDATA[In complex environments, decision-making for an Intelligent Vehicle (IV) requires reliability to ensure safe and efficient navigation. During vehicle operation, context and vehicle's capabilities influence the feasibility of its possible decisions and actions functions. Context-awareness allows to enhance the reliability of the decision-making process, ensuring that the vehicle has the capabilities to effectively perform the desired action. To warrant the safe operation of a vehicle, the Operational Design Domain (ODD) concept has been introduced. In the literature, it defines the conditions under which the vehicle is designed to operate safely. In this survey the ODD concept serves as a formalism to describe the context according to an established taxonomy. This survey focuses on how the operational context and the vehicle's capabilities determine the manner of how decisions are taken to ensure driving safety, comfort, and reliability. This is a multidimensional problem as the vehicle's capabilities, the road, the road users, and other elements of the context need to be considered. The different approaches and methods used in decision-making for IVs which take into account contextual information are identified as well as the research gaps that still need to be addressed in order to ensure reliable decision-making. Further, recent approaches that consider the ODD framework are presented to highlight the importance of this formalism. Conclusions underscore the importance of this integration for IVs and offer key insights for future research, emphasizing the crucial synergy between reliable long-term decision-making and the ODD as a contextual-awareness formalism.]]></description>
      <pubDate>Wed, 29 Apr 2026 09:10:07 GMT</pubDate>
      <guid>https://trid.trb.org/View/2659156</guid>
    </item>
    <item>
      <title>Autonomous Emergency Collision Avoidance and Collaborative Stability Control Technologies for Intelligent Vehicles: A Survey</title>
      <link>https://trid.trb.org/View/2659152</link>
      <description><![CDATA[This paper presents a comprehensive literature review on intelligent driving technologies, with a special emphasis on Automatic Emergency Collision Avoidance Technology (AECA) and Collaborative Stability Control (CSC). These technologies play a crucial role in the active safety of vehicles. AECA proactively detects and responds to potential collisions, and CSC enhances vehicle stability by integrating various systems across multiple driving scenarios. The synergy between AECA and CSC is essential for improving passenger safety and the overall efficiency of traffic systems. This review delves into the application of AECA and CSC, particularly under conditions that might compromise vehicle stability, emphasizing the crucial balance between safety and stability in collision avoidance scenarios. The paper discusses the challenges faced by intelligent vehicles, such as the strong coupling nonlinearity in vehicle dynamics, unpredictable environmental conditions, and the increasing complexity of control systems. It examines strategies in braking, steering, and the coordination of multiple systems to achieve effective collision avoidance and stability control. Additionally, the review provides a forward-looking perspective on potential developments and insights for ongoing research in domains of AECA and CSC within intelligent technologies. The goal is to present a structured overview of the current state of research, highlight significant findings, and identify critical areas where future research could significantly advance the field of intelligent driving systems.]]></description>
      <pubDate>Wed, 29 Apr 2026 09:10:07 GMT</pubDate>
      <guid>https://trid.trb.org/View/2659152</guid>
    </item>
    <item>
      <title>Bootstrapped Neural Models for Predicting Self-Driving Vehicle Collisions With Quantified Confidence: Offline and Online Applications</title>
      <link>https://trid.trb.org/View/2659142</link>
      <description><![CDATA[Highly automated vehicles are complex systems, and ensuring their safe operation within their Operational Design Domain (ODD) presents significant challenges. Diagnosing failure modes and updating these systems are even more demanding tasks. This paper introduces a method to assist with assessing, diagnosing, and updating these systems by developing a stochastic model that predicts safety outcomes (collision, near-miss, or safe state) with quantified uncertainty in any parametrized scenario. The approach uses bootstrapping aggregation to create an ensemble of predictive models, leveraging fully connected feed-forward neural networks. These networks are designed with a flexible number of trainable parameters and hidden layers, requiring minimal computational resources. The model is trained on a small set of examples obtained through direct simulations that randomly sample the parametric scenario, bypassing the traditional test matrix definition. Once trained, the bootstrapped model serves as an identity card for the system under test, allowing continuous performance evaluation across the parametric scenario. The paper demonstrates applications, including safety assessment, failure mode identification, and developing a safe speed recommendation function. The model's compact size ensures rapid execution, facilitating extensive post-analysis for safety argumentation and diagnosis and real-time online use to extend the system's abilities.]]></description>
      <pubDate>Wed, 29 Apr 2026 09:10:07 GMT</pubDate>
      <guid>https://trid.trb.org/View/2659142</guid>
    </item>
    <item>
      <title>Developing geometric criteria to ensure Minimal Risk Conditions for highly automated vehicles</title>
      <link>https://trid.trb.org/View/2692654</link>
      <description><![CDATA[Connected and Automated Vehicles (CAVs) represent a transformative technology with the potential to significantly enhance road safety and improve mobility for all users. However, it is important to recognize that CAVs are not infallible; there will inevitably be situations in which these vehicles must come to a stop to ensure safety. This designated stopping location is called the Minimal Risk Condition (MRC). Achieving MRC should be a standard operational capability for SAE Level 4 + vehicles, which necessitates that the automated system performs a safe Dynamic Driving Task (DDT) fallback when required, especially in instances where the human driver may not be prepared to take control of the vehicle. This study proposes various solutions to facilitate MRC, including the use of hard shoulder, Emergency Refuge Lane (ERL), and Safe Harbor (SH). Initially, the authors examine the advantages and disadvantages of each solution, followed by an assessment of their respective capacities − experimental for ERLs and analytical for SHs. Based on these evaluations, some geometric design criteria and diverse solutions applicable to various highway types and interchanges are proposed. This work represents an important first step in addressing a critical topic that will receive further attention in the coming years, as the requirements for these zones are defined more precisely in relation to CAV penetration rates and Operational Design Domain (ODD) limitations. Consequently, it is essential that road and interchange design guidelines should be updated and adapted to incorporate these new facilities effectively.]]></description>
      <pubDate>Tue, 28 Apr 2026 17:05:55 GMT</pubDate>
      <guid>https://trid.trb.org/View/2692654</guid>
    </item>
  </channel>
</rss>