<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>The governance of AI-enabled transport: Bridging Kuwait’s tech-policy gap vis-à-vis the UAE and Singapore</title>
      <link>https://trid.trb.org/View/2728110</link>
      <description><![CDATA[Artificial intelligence (AI) for urban mobility, including adaptive signal control, incident management, predictive maintenance, Mobility as a Service, and Level 4 shuttles, is technically mature, yet scale depends on institutional readiness. The authors propose a Technology-Policy Readiness Matrix (TPRM) coupling Technology Readiness Levels (TRL, one to nine) with a custom nine-rung Policy Readiness Level (PRL). To evaluate deployment risk, the matrix plots initiatives into qualitative alignment zones based on their Ordinal Divergence (Dordinal): Synchronized (Green), Pacing Gap (Amber), and Institutional Void (Red). Using a mixed-methods design (39 documents, 20 key-informant interviews; 2025 cutoff), Kuwait serves as the primary case study, contextualized against documentary benchmarks from the United Arab Emirates (UAE) and Singapore. Findings reveal a structural “hardware-first” bias rooted in Rentier State incentives: while analytics sit at TRL six to nine, policy readiness lags at PRL two to four. This divergence creates an Institutional Void driven by institutional decoupling, where the state prioritizes allocative procurement over the regulatory enforcement required to routinize data sharing and application programming interfaces (APIs). The benchmarks demonstrate near parity for analytics, while autonomy remains in a pacing gap pending liability and insurance frameworks. A targeted sequence of governance mechanisms lifts PRL by one to two rungs: transport data and API circulars, single-window permitting, KPI-linked public–private partnerships, and standardized assurance with safety cases for autonomous vehicle pilots. A 2026 to 2035 roadmap models three pathways (Managed Transition, Accelerated Innovation, and Policy Lag) setting triggers for API enforcement, KPI disclosure, and sandbox licenses. Closing the readiness gap requires a shift from allocative capacity to regulatory autonomy; doing so yields earlier gains in travel time, ensures compliance with national decarbonization targets, and provides a portable diagnostic for derisking autonomy.]]></description>
      <pubDate>Fri, 14 Aug 2026 15:04:42 GMT</pubDate>
      <guid>https://trid.trb.org/View/2728110</guid>
    </item>
    <item>
      <title>AI’s role in efficient logistics management: Shaping the future of supply chain in India</title>
      <link>https://trid.trb.org/View/2728066</link>
      <description><![CDATA[The National Logistics Policy (NLP) (2022) of the Government of India aims to transform the national logistic landscape, facilitate multimodal transport and reduce logistics costs to enhance global competitiveness. The study conceptualizes and empirically analyses the impact of policy framework, infrastructure development, multimodal integration (MI) and artificial intelligence (AI) adoption on logistics performance (LP) and cost reduction in Indian supply chain. This research examines how the NLP with AI impacts these issues. Utilizing a systematic approach align with the NLP, this research examines primary data of 500 transport providers to assess improvement in transport efficiency, modal shifts and cost reductions across major transport corridors. This empirical analysis incorporated geographical information, freight volumes, vehicle turnaround time and cost indicators. The proposed model demonstrates a 30.5% reduction in the Average Transit Time (hrs.) following the implementation of NLP. Ultimately, this research highlights the transformative potential of combining national policy with AI-driven analysis offering actionable insights to improve policies formulation, infrastructure investments and digital connectivity in India’s logistics sector.]]></description>
      <pubDate>Fri, 14 Aug 2026 15:04:42 GMT</pubDate>
      <guid>https://trid.trb.org/View/2728066</guid>
    </item>
    <item>
      <title>Vision-language and generative models in traffic video safety analysis: A computational framework and research agenda</title>
      <link>https://trid.trb.org/View/2732616</link>
      <description><![CDATA[Vision–language and generative models have recently emerged as powerful tools for interpreting multimodal traffic–video data and advancing safety analysis. This paper reviews and integrates progress across foundation vision–language models, multimodal large language models, video-centric temporal reasoning frameworks, and diffusion-based world models, emphasizing how these paradigms enable richer semantic understanding, causal reasoning, and counterfactual safety assessment. The authors propose a unified taxonomy that maps model families to three application levels across diverse deployment environments—from cloud to onboard systems: network-scale monitoring, event-level crash and near-miss understanding, and generative or counterfactual scenario analysis. Key technical and methodological challenges are identified, including hallucination control, temporal consistency, sim–to–real transfer, and safety alignment with physical and rule-based constraints. The paper synthesizes open research problems and outlines a structured agenda toward grounded, interpretable, and computationally efficient multimodal cognition for real-world traffic safety applications.]]></description>
      <pubDate>Thu, 13 Aug 2026 17:01:45 GMT</pubDate>
      <guid>https://trid.trb.org/View/2732616</guid>
    </item>
    <item>
      <title>Enhancing semantic and risk controllability in safety-critical scenario generation: An LLM-guided conditional diffusion method</title>
      <link>https://trid.trb.org/View/2728312</link>
      <description><![CDATA[Safety-critical scenario generation is essential for autonomous vehicle validation. However, existing methods often struggle to simultaneously achieve semantic controllability, risk-level controllability, and trajectory realism. To address this limitation, the authors propose a Large Language Model (LLM)-guided conditional diffusion framework for controllable safety-critical scenario generation. Specifically, the authors first use an LLM to convert natural-language descriptions into structured generation conditions. These conditions are then incorporated into a conditional latent diffusion model to guide scenario generation with controllable semantic and risk features. The generated candidate scenarios are further filtered to retain those with high semantic consistency. Experiments are conducted on the highD dataset for car-following and cut-in scenarios. The results show that the proposed method can generate realistic and risk-controllable scenarios. Compared with representative baseline methods, it achieves a better balance among criticality, realism, and controllability. Ablation experiments further demonstrate that LLM-based semantic conditioning improves both scenario-type and risk-level accuracy. The proposed method provides a flexible and controllable tool to generate safety-critical scenarios for autonomous vehicle testing and safety evaluation.]]></description>
      <pubDate>Thu, 13 Aug 2026 17:01:44 GMT</pubDate>
      <guid>https://trid.trb.org/View/2728312</guid>
    </item>
    <item>
      <title>Agentic AI for autonomous condition monitoring and predictive maintenance of marine vessels</title>
      <link>https://trid.trb.org/View/2733013</link>
      <description><![CDATA[Modern marine vessels generate large volumes of high-frequency operational data, yet translating this data into reliable maintenance intelligence remains challenging. Existing solutions rely on threshold alarms, rule-based logic, or isolated machine learning models, which lack contextual reasoning, adaptability, and interpretability in safety-critical environments. This paper presents an Agentic AI architecture for condition monitoring and predictive maintenance of marine vessels. The proposed system integrates a Transformer-based Autoencoder for multivariate anomaly detection with a Large Language Model (LLM) that enables interactive, explainable analysis through structured tool-calling. The autoencoder learns normal operational patterns from high-dimensional time-series data and detects deviations associated with abnormal loading, inefficient generator utilisation, and potential equipment degradation. Rather than ingesting raw numerical streams, the LLM invokes domain-specific tools to retrieve structured measurements, compute derived metrics, and query anomaly outputs before generating responses. This mechanism mitigates hallucination risks and avoids the brittleness of fine-tuned models as system behaviour evolves. The approach is validated on 91 days of data from an offshore construction vessel with a hybrid diesel-electric propulsion system and six gensets under dynamic positioning constraints. Four fault scenarios (slow drift, load imbalance, temporary reduction, spikes) were all correctly detected, with prognostic lead times between 97 and 637 min.]]></description>
      <pubDate>Thu, 13 Aug 2026 17:00:45 GMT</pubDate>
      <guid>https://trid.trb.org/View/2733013</guid>
    </item>
    <item>
      <title>Towards Comprehensive Safety Assurance of a DAL A AI/ML-based Runway Alignment System using Overarching Properties</title>
      <link>https://trid.trb.org/View/2732486</link>
      <description><![CDATA[This report details an Overarching Properties (OPs)-based approach for assuring the safety of Artificial Intelligence/Machine Learning (AI/ML)-based digital aerospace systems. To rigorously evaluate and enhance this approach, an AI-Assisted Autonomous Runway Alignment (AARA) system is introduced as a motivating use case, allowing for the identification and mitigation of potential safety risks through premise-based arguments. The study demonstrates how multi-level safety assessments can be conducted for AI/ML systems to pinpoint failure conditions inherent to AI/ML's nature. It also provides examples of how requirements can be formulated to address risks posed by black-box AI/ML components with unpredictable or uncontrollable behaviors. Comprehensive discussions cover various aspects of the OPs-based approach, including foreseeable operating conditions, development and training activities, evidence generation for premises, hybrid certification, design assurance, necessary assumptions, and a plan for OPs compliance. While currently focused on Artificial Neural Networks (ANNs) developed using supervised learning, the arguments may be adaptable to other AI techniques. The findings establish a strong foundation for applying OPs to AI/ML assurance, acknowledging the need for further work to ensure robustness and practicality across diverse criticality and autonomy spectrums.]]></description>
      <pubDate>Thu, 13 Aug 2026 15:07:33 GMT</pubDate>
      <guid>https://trid.trb.org/View/2732486</guid>
    </item>
    <item>
      <title>Genai-Driven Quantum-Resilient Consensus Framework for Blockchain-Enabled Vehicular Digital Twins</title>
      <link>https://trid.trb.org/View/2732060</link>
      <description><![CDATA[Digital Twin (DT) technology is elevating the next-generation intelligent transportation systems industry to new heights, as it enables real-time monitoring, predictive maintenance, and adaptive control of connected and autonomous vehicles. However, the use of GenAI and DTs in interconnected vehicular technology ecosystems introduces new attack vectors, particularly from quantum computing, which can easily break classical encryption systems. This paper introduces Reputation-based Proof-of-Stake (R-PoS), a hybrid consensus mechanism tailored for lattice-based post-quantum cryptography (PQC) operations on vehicular edge devices. The core contribution is a lightweight hybrid consensus mechanism optimized for lattice-based PQC on edge devices, enabling secure and scalable synchronization between physical assets and their digital twins. Experimental results from a containerized IoT testbed using the Open Quantum Safe (OQS) library show that the proposed PQC-blockchain (PQC-BC) framework achieves an average throughput of 1178 transactions per second with latency of 0.78 second. These results affirm the framework’s efficacy in securing future interconnected vehicular environments and establishing a trust foundation for sustainable quantum-resistant digital twin applications.]]></description>
      <pubDate>Wed, 12 Aug 2026 15:14:56 GMT</pubDate>
      <guid>https://trid.trb.org/View/2732060</guid>
    </item>
    <item>
      <title>Efficient Driving Behavior Narration and Reasoning on Edge Device Using Large Language Models</title>
      <link>https://trid.trb.org/View/2731703</link>
      <description><![CDATA[Large language models (LLMs) with robust reasoning capabilities have significantly advanced the development of autonomous driving technologies, particularly in the narration and reasoning of driving behaviors, which hold substantial importance for accident analysis and traffic management. However, traditional deployment of these models relies on cloud servers, resulting in high latency and training costs, making it challenging to meet the stringent real-time requirements of autonomous driving scenarios. Recent studies suggest that edge computing, by deploying models closer to the data source, offers a promising solution to these issues. While existing general-purpose LLMs excel in video understanding and task reasoning, their generalization capabilities in rapidly changing traffic scenarios remain questionable. This paper provides a valuable reference for deploying LLMs at the edge in autonomous driving contexts. By leveraging real-world 5G networks for rapid deployment, we validate the performance and response speeds of various models in autonomous driving scenarios. Furthermore, we introduce an innovative prompt engineering strategy that enhances model performance by 25% without changing model parameters through minimal prompt tuning. Experimental results demonstrate that LLMs deployed on edge devices achieve satisfactory response times. Tests on the OpenDV-YouTube dataset further confirm that our prompt strategy significantly improves the performance of driving behavior narration and reasoning.]]></description>
      <pubDate>Wed, 12 Aug 2026 15:14:55 GMT</pubDate>
      <guid>https://trid.trb.org/View/2731703</guid>
    </item>
    <item>
      <title>Enhancing Universal Mixed-Autonomy Channel Modeling with Explainable Artificial Intelligence</title>
      <link>https://trid.trb.org/View/2731736</link>
      <description><![CDATA[In the forthcoming transportation landscape, the integration of Connected and Automated Vehicles (CAVs) with traditional human-driven traffic environments presents a multifaceted challenge. In this mixed-autonomy scenario, the coexistence of CAVs and Human-Driven Vehicles (HDVs) necessitates the sharing of road space and resources, all while striving to ensure safety and transportation efficiency. Within this dynamic context, the rapid and accurate prediction of channel quality becomes paramount for ensuring system stability and reliability. However, the inherent complexity and variability of such traffic environments introduce a multitude of interfering factors that conventional channel models struggle to address effectively. Thus, the development of a universal mixed-autonomy channel model that can adapt to diverse conditions and enhance communication quality is of paramount importance. Our proposed channel model, enhanced by Explainable Artificial Intelligence (XAI), encompasses system design of comprehensive mixed-autonomy environments, data collection, Machine Learning (ML) training, feature analysis using SHapley Additive exPlanation (SHAP), and performance validation. This model, underpinned by rigorous data-driven analysis, enables precise and efficient predictions of channel characteristics, offering a flexible and impactful solution that advances the intelligence of mixed-autonom systems and enhances communication reliability.]]></description>
      <pubDate>Wed, 12 Aug 2026 15:14:54 GMT</pubDate>
      <guid>https://trid.trb.org/View/2731736</guid>
    </item>
    <item>
      <title>Market-Incentivized Aggregation Modeling and Capacity from Distributed Electric Vehicles with AI-Enabled Evolutionary Peer-to-Peer Game</title>
      <link>https://trid.trb.org/View/2743270</link>
      <description><![CDATA[The aggregation of distributed electric vehicles (EVs) initiates a promising avenue for constructing a virtual energy storage system within the distribution electricity market. To unlock the latent storage capacity inside the large-scale electrified transportation system and reduce redundant investment in the traditional power plant, a market-incentivized aggregation framework is developed to integrate those inherently stochastic energy resources. To address the substantial computational complexity posed by large EV populations and align with their distributed operational nature, an artificial intelligence (AI)-enabled evolutionary peer-to-peer game with price-making and probabilistic policy is deployed to emulate market dynamics effectively. The proposed strategy minimizes overall energy demand and shifts peak loads, thereby enhancing system flexibility and enabling the formation of an aggregation-based capacity for grid management. In response to the varying energy dynamics across communities, the framework embeds a price-making scheme to reinforce economic incentives for participation. Through continuous, distributed dispatch enabled by the probabilistic policies, fine-grained management over large-scale energy integration is achieved. Experimental evaluations using real-world distribution system data validate the effectiveness of the proposed framework in stimulating EV participation and establishing a robust, aggregated energy storage system.]]></description>
      <pubDate>Tue, 11 Aug 2026 16:35:12 GMT</pubDate>
      <guid>https://trid.trb.org/View/2743270</guid>
    </item>
    <item>
      <title>AI-assisted Condition Assessment of Roads</title>
      <link>https://trid.trb.org/View/2752288</link>
      <description><![CDATA[The objective of this project is to develop an AI-assisted road monitoring system that enables low-cost, autonomous, and frequent condition-based assessments using a network of mobile sensing units. The system will use computer vision and machine learning to detect and quantify pavement defects, replacing traditional schedule-based inspections with continuous, data-driven monitoring. The proposed system provides transportation agencies with an affordable, scalable, and intelligent tool for real-time pavement monitoring. By using low-cost sensors on existing vehicles and automated data interpretation, it delivers accurate condition insights, reduces inspection costs, and supports timely maintenance decisions.]]></description>
      <pubDate>Thu, 13 Aug 2026 15:31:12 GMT</pubDate>
      <guid>https://trid.trb.org/View/2752288</guid>
    </item>
    <item>
      <title>Integrating Data-Driven and Model-Driven Approaches for Traffic-State Estimation in Data-Deficient Areas</title>
      <link>https://trid.trb.org/View/2752197</link>
      <description><![CDATA[Missing or incomplete traffic data caused by sensor malfunctions and the absence of detectors create data-deficient areas that hinder the efficient and safe operation of road networks. This issue is particularly acute on highly congested road segments where a high-resolution traffic state is essential to mitigate congestion and enhance safety. This study presents a traffic-state estimation model designed to function effectively in such data-deficient environments. The proposed attention-based model integrates model-driven and data-driven approaches, combining the former’s ability to infer unobserved states with the latter’s adaptability to diverse traffic scenarios. Microscopic traffic simulation was employed to generate physically coherent and high-resolution training data, incorporating realistic variations in origin–destination patterns and driving behaviors. The model learns both the temporal dependencies of traffic evolution and the spatial correlations among detectors through gated recurrent units (GRU) and attention mechanisms. Validation was conducted using detector and drone data collected from the Gyeongbu Expressway, one of South Korea’s most heavily traveled corridors. The model achieved a mean absolute error within 17 vehicles per lane for volume, 10 km/h for speed, and 6% for occupancy, successfully reproducing fine-scale traffic dynamics even where no detectors were installed. This research contributes to improving traffic-state estimation practices by demonstrating how a substantial amount of simulated data with simple calibrations offers a versatile model, particularly in areas where data are deficient.]]></description>
      <pubDate>Tue, 11 Aug 2026 10:13:56 GMT</pubDate>
      <guid>https://trid.trb.org/View/2752197</guid>
    </item>
    <item>
      <title>Artificial intelligence supported road vehicle suspension design</title>
      <link>https://trid.trb.org/View/2751968</link>
      <description><![CDATA[This thesis presents an AI-supported framework for vehicle suspension design, combining reinforcement learning (RL) and reverse engineering to automate hardpoint optimization. A case study demonstrates a 50% reduction in design lead time. The proposed framework uses RL to derive suspension kinematics targets from vehicle-level requirements and reverse engineering to convert these targets into hardpoint configurations. The full case study demonstrates the practical application of this integrated methodology. The findings conclude that AI-supported suspension design algorithms significantly enhance both the efficiency and precision of suspension architecture development. The wheel suspension represents one of the most architecture-intensive systems in automotive design, largely determining a vehicle's motion characteristics and performance boundaries. Increasing pressures from electrification and intensifying global competition demand accelerated and more efficient development of new vehicle concepts, even within traditional domains like mechanical wheel suspension design. This system encompasses numerous design parameters with intricate interdependencies. Conventionally, development relies heavily on highly specialized engineering expertise. A significant bottleneck in modern suspension development involves balancing complex performance requirements that currently require time-consuming iterations. Today's development process also involves virtual subjective assessment alongside traditional chassis engineering experience. Addressing these challenges requires a full review of the entire development workflow-from initial target setting through verification and subsequent optimization loops.]]></description>
      <pubDate>Fri, 07 Aug 2026 08:34:47 GMT</pubDate>
      <guid>https://trid.trb.org/View/2751968</guid>
    </item>
    <item>
      <title>A Data-Driven Probabilistic Framework Using Computational Fluid Dynamics, Artificial Intelligence, and Underwater Robotics for Predicting Bridge Scour</title>
      <link>https://trid.trb.org/View/2745258</link>
      <description><![CDATA[Scour is the leading cause of bridge failure in the U.S. Traditional methods of
scour prediction rely on empirical formulas that require flow information at
bridge location, which is scarce and hard to obtain, and scour inspections often
rely on human divers which is costly, involve safety risks, and often lack the
precision and adaptability needed for the complex coastal and estuarine
environments. This project will develop a novel approach for scour prediction
and mapping that addresses these shortcomings by integrating computational
fluid dynamics (CFD) and machine learning (ML) for real-time prediction of flow
velocity and scour, and underwater robots powered by first-principles and
machine learning-driven perception to provide high fidelity maps of the scour
beyond capabilities of human divers. Project tasks include: (1) identify bridges
vulnerable to scour and characterize their environmental conditions and
structural features, (2) development of a CFD model for scour of a vulnerable
bridge, and deployment of a current meter on the channel bed close to the
bridge to measure currents that drive scour, and use of its data to validate the
CFD model, (3) run the CFD model for a variety water level conditions, spanning
regular tides to intense storms to generate training data for a ML model that will
calculate scour in real time given real-time current measurements at operational
gauges, (4) Develop a probabilistic framework for scour prediction using the
trained ML model, (5) deployment of low-cost underwater autonomous vehicles
to map a scour patch pre- and post-storm, and using the data to validate the
scour models. The framework in this proof-of-concept project can be scaled up
to numerous bridges across any region in future studies. By combining novel simulation and in-situ data acquisition techniques, this project aims to enable risk-informed decision making for management of bridge infrastructure.
]]></description>
      <pubDate>Fri, 07 Aug 2026 08:37:48 GMT</pubDate>
      <guid>https://trid.trb.org/View/2745258</guid>
    </item>
    <item>
      <title>Advanced Sensing and AI for Next-Generation Transportation Asset Management</title>
      <link>https://trid.trb.org/View/2745222</link>
      <description><![CDATA[The proposed study aims to create an artificial intelligence (AI)-based framework
that harness data from advanced sensing technologies to advance the next-generation of transportation asset management systems. The development of
AI-driven diagnostic tools for infrastructure is fundamental to optimizing
maintenance plans, thereby enhancing public safety and minimizing the
inefficient use of economic resources.
To address the limitations associated with the availability of training data for
deep learning algorithms, the proposed framework integrates heterogeneous
data collected from multiple sources. Beyond the deployment of conventional
fixed sensors, cutting-edge mobile sensing technologies will be incorporated to
achieve unprecedented temporal and spatial resolution, thereby ensuring the
scalability and adaptability of the AI-based strategy. The dynamic characteristics
extracted from acceleration data acquired through smartphones will be utilized
to: (i) calibrate the digital twin of the structure, (ii) identify and characterize
potential structural damage, and (iii) provide essential physics-based knowledge
to support the development of a physics-informed neural network for life-cycle
assessment.]]></description>
      <pubDate>Fri, 07 Aug 2026 08:35:24 GMT</pubDate>
      <guid>https://trid.trb.org/View/2745222</guid>
    </item>
  </channel>
</rss>