<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>A Multimode Cooperative Control Architecture for Connected and Automated Vehicle Platoon Splitting and Merging in Mixed Traffic</title>
      <link>https://trid.trb.org/View/2712065</link>
      <description><![CDATA[Connected and automated vehicle (CAV) platoons provide significant advantages in enhancing traffic efficiency and safety through vehicle-to-vehicle cooperative driving. However, owing to the uncertainty of human-driven vehicles in mixed traffic environments, platoons must frequently split to avoid potential collisions and merging is required to maintain platoon following. To address this challenge, this paper proposes a cooperative control architecture for CAV platoons that includes a single-vehicle cruising control mode and a platoon-following control mode, enabling independent operation of each mode and discrete event transitions around split and merge maneuvers. In single-vehicle mode, a driving safety potential field model is proposed for collision-avoidance trajectory planning, and a distributed model predictive control algorithm is designed to achieve the distinct control objectives of the two modes. Then, a long short-term memory (LSTM) neural network and fuzzy logic are combined to predict collision risk and determine platoon split events. A cooperative control system is implemented to ensure continuous control and flexible switching between the two modes. Finally, joint simulations in PreScan, CarSim, and MATLAB/Simulink were conducted to evaluate the performance of the system across various obstacle scenarios. The results demonstrate that the proposed control architecture effectively coordinates vehicle maneuvers and adapts platoon formation to changes in traffic conditions.]]></description>
      <pubDate>Tue, 09 Jun 2026 14:35:55 GMT</pubDate>
      <guid>https://trid.trb.org/View/2712065</guid>
    </item>
    <item>
      <title>Perception of cyclists while being overtaken by motorized vehicles: A semi-naturalistic field study</title>
      <link>https://trid.trb.org/View/2705494</link>
      <description><![CDATA[To promote sustainable mobility, it is important to strengthen cycling as a mode of transport. A key deterrent to cycling is not only objective risk but also the subjective perception of safety. One particularly critical situation is being overtaken by motorized vehicles – where both risk and safety perception are relevant. Despite this, existing research has rarely combined objective measures of passing distance and other situational characteristics with in-situ perceptions of multiple cyclists during overtaking situations in urban traffic. A field study was conducted in Braunschweig, Germany, with 38 participants cycling along two predefined routes on an equipped bicycle. A total of 720 overtaking maneuvers were recorded, measuring passing distance, traffic and infrastructure characteristics, and cyclists' perceived mental comfort. Only 43% of all overtaking maneuvers met the legally required minimum distance of 150 cm. Passing distance was influenced by factors including street width, cycling infrastructure, oncoming traffic and vehicle size. Larger distances were measured on different types of cycle lanes, while narrow roads, oncoming traffic and larger vehicles led to the smallest distances. Analysis showed that perceptions of mental comfort increased with greater passing distance, particularly when exceeding 150 cm. Heavy vehicles and oncoming traffic reduced mental comfort, while cycling on mandatory or red-marked advisory cycle lanes improved it. The findings indicate that infrastructure design and situational traffic characteristics combine to simultaneously shape drivers' overtaking behavior and influence cyclists' subjective experience of safety.]]></description>
      <pubDate>Thu, 04 Jun 2026 15:13:33 GMT</pubDate>
      <guid>https://trid.trb.org/View/2705494</guid>
    </item>
    <item>
      <title>Responsibility Attribution for Autonomous Vehicle Crashes Based on Causal Inference: Case Study in California, USA</title>
      <link>https://trid.trb.org/View/2709310</link>
      <description><![CDATA[In the autonomous driving environment, the attribution of responsibility becomes complex when multiple crash parties and factors are involved. This study proposes a method to attribute the responsibility of the primary crash vehicle when human drivers and autonomous driving systems coexist, and apply it to the existing Autonomous Vehicles (AVs) crash data from 2019 to 2023 in California, USA. Firstly, a causal network is constructed by integrating the Decision-making Trial and Evaluation Laboratory (DEMATEL) and Interpretative Structural Modeling (ISM) methods to analyze the mutual impact of factors in the crash data. Secondly, Random Forest (RF) is used to obtain the feature importance in AV crashes. Based on the relationship between factors and the main responsible parties, the responsibility among relative stakeholders can be quantified. Under the research data in California, in autonomous driving mode, human drivers of the primary crash vehicle and software developers both account for 31% of the crash. Following behind are other stakeholders at 21% and vehicle manufacturers at 17%. On this basis, adjustments can be made to the responsible proportion in relation to a specific crash. By identifying the impact factors of AV crashes and responsibility attribution, this study offers important insights into safe autonomous driving tests, AV production regulation, and the development of crash responsibility policies. The methodology framework developed in this paper is universal and can be applicable to AV crash analysis in diverse regions and AV penetration rates.]]></description>
      <pubDate>Wed, 03 Jun 2026 09:07:22 GMT</pubDate>
      <guid>https://trid.trb.org/View/2709310</guid>
    </item>
    <item>
      <title>Use of Bicycle Signal Faces in Applications with Concurrent Turning Traffic</title>
      <link>https://trid.trb.org/View/2709246</link>
      <description><![CDATA[Bicycle infrastructure is expanding rapidly across the United States. Conflict involving bicyclists and other users at urban intersections is a growing safety concern. Various signalization strategies have been employed to govern bicycle and turning-vehicle movements, but the use of the bicycle signals has been restricted to locations where bicycle movement is phase-separated from turning traffic.

Because dedicated bicycle phases can substantially reduce green time for both cyclists and turning vehicles, the added delay may reduce red-signal compliance or encourage cyclists to use less-protected routes or ride in mixed traffic.

Research is needed to better understand the safety and operational effects of bicycle signals. This includes evaluating tradeoffs for cyclists and drivers when phase separation is required; documenting current methods that allow cyclists to proceed concurrently with turning vehicles, with or without bicycle signal indications; and comparing compliance and crash outcomes associated with different signalization strategies.

OBJECTIVE: The objective of this research is to explore the feasibility and implications of configuring bicycle signals to permit bicycles to move concurrently with turning traffic.]]></description>
      <pubDate>Tue, 02 Jun 2026 14:44:11 GMT</pubDate>
      <guid>https://trid.trb.org/View/2709246</guid>
    </item>
    <item>
      <title>A Conflicts-Free, Speed-Lossless KAN-Based Reinforcement Learning Decision System for Interactive Driving in Roundabouts</title>
      <link>https://trid.trb.org/View/2617791</link>
      <description><![CDATA[Safety and efficiency are crucial for autonomous driving in roundabouts, especially mixed traffic with both autonomous vehicles (AVs) and human-driven vehicles. This paper presents a learning-based algorithm that promotes safe and efficient driving across varying roundabout traffic conditions. A deep Q-learning network is used to learn optimal strategies in complex multi-vehicle roundabout scenarios, while a Kolmogorov-Arnold Network (KAN) improves the AVs’ environmental understanding. To further enhance safety, an action inspector filters unsafe actions, and a route planner optimizes driving efficiency. Moreover, model predictive control ensures stability and precision in execution. Experimental results demonstrate that the proposed system consistently outperforms state-of-the-art methods, achieving fewer collisions, reduced travel time, and stable training with smooth reward convergence.]]></description>
      <pubDate>Mon, 01 Jun 2026 09:10:53 GMT</pubDate>
      <guid>https://trid.trb.org/View/2617791</guid>
    </item>
    <item>
      <title>Automated Mobility and Cooperation Compliance in Mixed Vehicle Traffic Environments</title>
      <link>https://trid.trb.org/View/2579085</link>
      <description><![CDATA[Automated vehicles have the potential to transform transportation systems through enhanced safety, efficiency, and sustainability. This chapter first reviews cooperative control approaches for Connected Automated Vehicles (CAVs) at conflict areas such as merging roadways, intersections, and lane-changing maneuvers. Formulating and solving optimal control problems with hard safety constraints enables the derivation of cooperative trajectories. When solutions to such problems become computationally intractable, online control methods based on Control Barrier Functions (CBFs) can be used to still guarantee all constraint satisfaction at the expense of some possible performance loss. The chapter examines extending these approaches to mixed traffic environments, introducing new techniques to guarantee safety despite the presence of unpredictable Human-Driven Vehicles (HDVs). Finally, a cooperative compliance framework is proposed to incentivize HDVs to align their behavior with CAV objectives using virtual, refundable tokens, without any monetary transactions. The goal is to provide foundations and specific new techniques aimed at optimizing automated mobility in mixed traffic.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:37 GMT</pubDate>
      <guid>https://trid.trb.org/View/2579085</guid>
    </item>
    <item>
      <title>Eco-Driving for Connected and Automated Vehicles: Managing Uncertainty in Mixed Traffic With a Physics-Enhanced Data-Driven Method</title>
      <link>https://trid.trb.org/View/2658922</link>
      <description><![CDATA[With the growing emphasis on sustainable transportation, enhancing vehicular energy efficiency in mixed traffic scenarios has become a key focus, particularly in cooperative control strategies for Connected and Automated Vehicles (CAVs) amidst the unpredictability of Human-Driven Vehicles (HDVs). These strategies are essential for improving driving safety, traffic efficiency, and reducing energy consumption. This study introduces a novel Physics-Enhanced Data-Enabled Predictive Control (PE-DeePC) method to optimize CAV control in mixed traffic flows, where the presence of HDVs introduces significant uncertainty. By incorporating partial system physics, such as velocity and acceleration, into a unified state-space equation framework, PE-DeePC effectively combines data-driven insights with physical models, enabling optimal control decisions that are both accurate and robust. This method tackles the challenges of improving energy efficiency in the face of noisy measurements and the unpredictable behaviors of HDVs. Extensive simulation results show that the PE-DeePC achieves superior performance in robust CAV control under noisy, unpredictable conditions, and reduces energy consumption in the entire mixed traffic flow. The framework demonstrates energy reductions of 8.57%, 3.59%, and 13.17% over existing benchmark algorithms, highlighting its potential as a reliable and efficient solution for improving traffic management and energy efficiency in real-world mixed traffic scenarios.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:35 GMT</pubDate>
      <guid>https://trid.trb.org/View/2658922</guid>
    </item>
    <item>
      <title>A Dual-Layer Mixture of Expert Model-Based Human-Like Strategy for Autonomous Driving Velocity Control</title>
      <link>https://trid.trb.org/View/2658900</link>
      <description><![CDATA[In the context of mixed traffic flow, it is importance to autonomous vehicles (AVs) that exhibit human-like and safe driving strategies. Models based on deep reinforcement learning (DRL) have strong generalization abilities, but their security is difficult to guarantee. Jointly modeling DRL with rule-based models can improve the security of the model, but it may hinder the learning of an optimal strategy. To address this challenge, this paper proposes a dual-layer mixture of expert model based on DRL and rule model (2L-DU-MOE). First, use the upper DRL model alone to explore the optimal strategy. Then, use DRL to establish an activation layer to connect the rule-based model and the upper model, making it adaptively arbitrate between the strategies of the upper and lower models according to the current environment. The real trajectories are extracted from trajectory dataset in order to construct scenarios with mixed traffic flow. Then, the features of human drivers are extracted from trajectory data and integrated into the reward function as one of the learning objectives. The test results in the mixed traffic flow scenarios show that 2L-DU-MOE achieved a success rate of 96.04%, surpassing the highest baseline model’s success rate by 11.63%. Meanwhile, the test results show that the proposed model’s acceleration, velocity, and TTC characteristics are similar to those of human drivers, exhibiting human-like characteristics. The test results in scenarios involving static obstacles avoidance demonstrate that AV can effectively navigate around obstacles by adjusting velocity and yaw angle. 2L-DU-MOE exhibits remarkable generalization capability and robustness, enabling human-like and safe autonomous driving.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:35 GMT</pubDate>
      <guid>https://trid.trb.org/View/2658900</guid>
    </item>
    <item>
      <title>HGSCO: Heterogeneous Graph Structure Contrast Optimization for Trajectory Prediction</title>
      <link>https://trid.trb.org/View/2658853</link>
      <description><![CDATA[Predicting and planning the future trajectories of various traffic participants is an important task with multiple applications, including autonomous vehicles, service robots, and intelligent transportation. However, the diversity of heterogeneous agents including pedestrians, bicycles, and vehicles in traffic scenarios presents substantial challenges to this task. Current models do not fully capture the implicit and explicit interaction relationships among these heterogeneous agents and often overlook the significance of extracting implicit correlations from agent features. To address these issues, we introduce a novel model for trajectory prediction: Heterogeneous Graph Structure Contrast Optimization (HGSCO). To accurately capture the interaction relationships among heterogeneous agents, HGSCO constructs semantic graph structures representing implicit relationships and meta-path graph structures representing explicit relationships. Then, HGSCO introduces a cross-view contrastive learning approach, which optimizes the heterogeneous graph structure by maximizing mutual information between the two types of graph structures. The model can provide precise interaction relationships among heterogeneous agents by effectively fusing these two graph representations with a gated fusion method. We utilized video data captured by camera sensors in complex environments with multiple agents to conduct experiments. Our proposed model achieved an 11.5% and 6.7% reduction in Average Displacement Error (ADE) across these datasets, respectively, and a reduction of 15.6% and 8.1% in Final Displacement Error (FDE). The results demonstrate that HGSCO significantly surpasses existing state-of-the-art methods regarding trajectory prediction accuracy.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:34 GMT</pubDate>
      <guid>https://trid.trb.org/View/2658853</guid>
    </item>
    <item>
      <title>MRIC: Model-Based Reinforcement-Imitation Learning With Mixture-of-Codebooks for Autonomous Driving Simulation</title>
      <link>https://trid.trb.org/View/2658842</link>
      <description><![CDATA[Accurately simulating the diverse behaviors of heterogeneous agents in varied scenarios is fundamental to autonomous driving simulation. Our first insight is to leverage state-matching through differentiable simulation to provide informative learning signals and enable efficient credit assignment for the policy. We demonstrate this by identifying gradient highways and inter-agent gradient pathways. However, challenges such as gradient explosion and weak supervision in low-density regions emerge. Our second insight addresses these issues by applying dual policy regularizations to constrain the function space. To further enhance diversity, our third insight involves compressing the behaviors of heterogeneous agents into a series of prototype vectors for retrieval. These insights culminate in our model-based reinforcement-imitation learning framework with a temporally abstracted mixture-of-codebooks (MRIC). MRIC integrates open-loop model-based imitation learning to stabilize training and model-based reinforcement learning (RL) to inject domain knowledge. The RL component introduces differentiable rewards for collision avoidance, road adherence, and traffic rule compliance. We further propose a dynamic multiplier mechanism to maintain the effectiveness of the regularizations while avoiding interference. Experiments on the large-scale Waymo Open Motion dataset show that MRIC significantly outperforms strong baselines in diversity, behavioral realism, and distributional fidelity, achieving notable improvements in metrics such as collision rate, minSADE, and time-to-collision JSD.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:34 GMT</pubDate>
      <guid>https://trid.trb.org/View/2658842</guid>
    </item>
    <item>
      <title>Bi-Level Control of Weaving Sections in Mixed Traffic Environments With Connected and Automated Vehicles</title>
      <link>https://trid.trb.org/View/2658838</link>
      <description><![CDATA[Connected and automated vehicles (CAVs) can be beneficial for improving the operation of highway bottlenecks such as weaving sections. This paper proposes a bi-level control approach based on an upper-level deep reinforcement learning controller and a lower-level model predictive controller to coordinate the lane-changings of a mixed fleet of CAVs and human-driven vehicles (HVs) in weaving sections. The upper level represents a roadside controller that collects vehicular information from the entire weaving section and determines the control weights used in the lower-level controller. The lower level is implemented within each CAV, which takes the control weights from the upper-level controller and generates the acceleration and steering angle for individual CAVs based on the local situation. The lower-level controller further incorporates an HV trajectory predictor, which is capable of handling the dynamic topology of vehicles in weaving scenarios with intensive mandatory lane changes. The case study inspired by a real weaving section in Basel, Switzerland, shows that our method consistently outperforms state-of-the-art benchmarks.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:34 GMT</pubDate>
      <guid>https://trid.trb.org/View/2658838</guid>
    </item>
    <item>
      <title>Safety-Critical Multi-Agent MCTS for Mixed Traffic Coordination at Unsignalized Intersections</title>
      <link>https://trid.trb.org/View/2658784</link>
      <description><![CDATA[Decision making at unsignalized intersections presents significant challenges for autonomous vehicles (AVs), particularly in mixed traffic scenarios where both AVs and human-driven vehicles (HDVs) must safely coordinate their movements. This paper proposes a safety-critical multi-agent Monte Carlo tree search (MCTS) framework that integrates deterministic and probabilistic predictions to enable cooperative decision making in complex intersection scenarios. The framework incorporates three main innovations: 1) a safety assessment mechanism that systematically handles AV-to-AV (V2V), AV-to-HDV (V2H), and Vehicle-to-Road (V2R) interactions using dynamic safety thresholds and spatiotemporal risk metrics, 2) an adaptive HDV behavior awareness by combining the Intelligent Driver Model (IDM) with probabilistic distributions, and 3) a multi-objective reward function optimization approach that balances safety, efficiency, and cooperation. Extensive simulations demonstrate our framework’s efficacy and superior capability in ensuring safe and efficient intersection navigation across the fully-autonomous scenario (100% AVs) and challenging mixed traffic scenario (50% AVs +50% HDVs). Compared to benchmarks, our method reduces trajectory deviations by up to 37.56% in the fully-autonomous scenario and 62.43% in the mixed traffic scenario, while maintaining significantly lower Post-Encroachment Time (PET) violations (0% and 2.8%, respectively).]]></description>
      <pubDate>Thu, 28 May 2026 17:09:33 GMT</pubDate>
      <guid>https://trid.trb.org/View/2658784</guid>
    </item>
    <item>
      <title>Longitudinal Car-Following Control of Connected Autonomous Vehicles in Realistic Scenarios: A Survey</title>
      <link>https://trid.trb.org/View/2658742</link>
      <description><![CDATA[By exploiting information from surrounding vehicles and road infrastructure, Connected Autonomous Vehicles (CAVs) are widely expected to enable improved traffic safety and efficiency, as well as fuel savings. In practice, however, the information exchange may suffer from communication delays and packet losses due to limited communication channel bandwidth, fading, and signal interference, and this might have severe negative effects on the longitudinal control performance of CAVs. Furthermore, it is expected that mixed traffic, i.e., vehicles with different automation levels and communication capabilities (Human-driven vehicles, Connected human-driven vehicles, Autonomous vehicles, CAVs), will coexist on roads during a long transition period. This creates additional control challenges for CAVs due to the uncertainty in human’s driving behaviour. While extensive surveys on longitudinal control strategies for CAVs and their impact on traffic performance have been reported in the literature in recent years, they have assumed CAVs’ operation in unrealistic scenarios such as perfect communication, pure CAVs traffic, identical vehicle dynamics, etc. In this paper, we review the existing work on CAVs in realistic scenarios with a focus on the control strategies to address the challenges in unreliable communication and mixed-traffic environments. Open research questions and potential future research directions to facilitate the deployment of CAVs are also highlighted.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:33 GMT</pubDate>
      <guid>https://trid.trb.org/View/2658742</guid>
    </item>
    <item>
      <title>Regional Cooperative Decision-Making Based on Coalition Game for Multilane Merging in Mixed Traffic</title>
      <link>https://trid.trb.org/View/2659035</link>
      <description><![CDATA[The coordinated operation of multiple connected autonomous vehicles (CAVs) is conducive to resolving the bottleneck problem in on-ramp merging areas. However, achieving cooperation under the interference of connected manual vehicles (CMVs) remains challenging. This study integrates lane volume balancing and game theory, developing a regional cooperative decision-making model based on coalition game (RCD-CG) in mixed traffic. First, a progressive cooperation method is proposed, and multiple cooperative regions are designed in the merging area and its upstream section with dynamically adjustable boundaries in response to the time-varying traffic flow. Second, within these regions, CAVs form a coalition to maximize overall efficiency while ensuring safety. Additionally, considering the influence of CMVs outside the coalition on the cooperation of multiple CAVs, the payoffs of CMVs with different driving styles are quantified to solve non-cooperative games between CAVs and CMVs. To validate the model, simulations are conducted on an on-ramp merging scenario with a three-lane mainline, comparing the performance of RCD-CG with other methods across varying CAV penetration, traffic demand, and lane flow unevenness. The results demonstrate that under medium and high CAV penetrations and on-ramp demands, the use of the RCD-CG offers clear benefits in terms of efficiency, with the average speed increase rates reaching 23.2%.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:32 GMT</pubDate>
      <guid>https://trid.trb.org/View/2659035</guid>
    </item>
    <item>
      <title>Distributed Switching Model Predictive Control for Adaptive Human-Lead-Platooning in Mixed Traffic</title>
      <link>https://trid.trb.org/View/2659003</link>
      <description><![CDATA[In this study, we propose an innovative multi-stage control framework for human-lead-platooning of autonomous vehicles in complex, mixed traffic environments. The framework begins with collecting aggressive driving data from expert human drivers under various weather conditions and visibility levels, which inform a Refined Intelligent Driver Model for predicting the driving states of human-driven vehicles. A novel trust mechanism is then introduced to guide the trajectory selection for each autonomous follower, leveraging a reference set provided by the human-driven leader. In parallel, user-centric preferences (e.g., motion sickness, emotional fear, situational urgency) are captured and converted into precise acceleration and control constraints through a Fuzzy Logic System. Finally, a distributed switching model predictive control algorithm coordinates lane changes and vehicle-following tasks in real time for each follower. The proposed approach is validated through hardware-in-the-loop testing, demonstrating both effectiveness and adaptability in diverse traffic scenarios.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:31 GMT</pubDate>
      <guid>https://trid.trb.org/View/2659003</guid>
    </item>
  </channel>
</rss>