Communications in Transportation Research最新文献_第2页

HUTFormer: Hierarchical U-Net transformer for long-term traffic forecasting huformer：用于长期流量预测的分层U-Net变压器

IF 14.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-11-14 DOI: 10.1016/j.commtr.2025.100218

Zezhi Shao , Fei Wang , Tao Sun , Chengqing Yu , Yuchen Fang , Guangyin Jin , Zhulin An , Yang Liu , Xiaobo Qu , Yongjun Xu

Traffic forecasting, which aims to predict traffic conditions based on historical observations, has been an enduring research topic and is widely recognized as an essential component of intelligent transportation. Recent proposals on Spatial-Temporal Graph Neural Networks (STGNNs) have made significant progress by combining sequential models with graph convolution networks. However, due to high complexity issues, STGNNs only focus on short-term traffic forecasting (e.g., 1-h ahead), while ignoring more practical long-term forecasting. In this paper, we make the first attempt to explore long-term traffic forecasting (e.g., 1-day ahead). To this end, we first reveal its unique challenges in exploiting multi-scale representations. Then, we propose a novel Hierarchical U-net TransFormer (HUTFormer) to address the issues of long-term traffic forecasting. HUTFormer consists of a hierarchical encoder and decoder to jointly generate and utilize multi-scale representations of traffic data. Specifically, for the encoder, we propose window self-attention and segment merging to extract multi-scale representations from long-term traffic data. For the decoder, we design a cross-scale attention mechanism to effectively incorporate multi-scale representations. In addition, HUTFormer employs an efficient input embedding strategy to address the complexity issues. Extensive experiments on four traffic datasets show that the proposed HUTFormer significantly outperforms state-of-the-art traffic forecasting and long time series forecasting baselines.

交通预测是基于历史观测对交通状况进行预测的研究课题，被广泛认为是智能交通的重要组成部分。近年来关于时空图神经网络（stgnn）的研究在时序模型与图卷积网络的结合方面取得了重大进展。然而，由于较高的复杂性问题，stgnn只关注短期的交通预测（如1小时前），而忽略了更实际的长期预测。在本文中，我们首次尝试探索长期交通预测（例如，提前1天）。为此，我们首先揭示了它在利用多尺度表示方面的独特挑战。然后，我们提出了一种新的分层u网变压器（huttformer）来解决长期流量预测问题。HUTFormer由分层编码器和解码器组成，共同生成和利用交通数据的多尺度表示。具体来说，对于编码器，我们提出了窗口自关注和片段合并来从长期交通数据中提取多尺度表示。对于解码器，我们设计了一个跨尺度注意机制来有效地融合多尺度表示。此外，HUTFormer采用高效的输入嵌入策略来解决复杂性问题。在四个交通数据集上的大量实验表明，所提出的huttformer显著优于最先进的交通预测和长时间序列预测基线。

{"title":"HUTFormer: Hierarchical U-Net transformer for long-term traffic forecasting","authors":"Zezhi Shao , Fei Wang , Tao Sun , Chengqing Yu , Yuchen Fang , Guangyin Jin , Zhulin An , Yang Liu , Xiaobo Qu , Yongjun Xu","doi":"10.1016/j.commtr.2025.100218","DOIUrl":"10.1016/j.commtr.2025.100218","url":null,"abstract":"<div><div>Traffic forecasting, which aims to predict traffic conditions based on historical observations, has been an enduring research topic and is widely recognized as an essential component of intelligent transportation. Recent proposals on Spatial-Temporal Graph Neural Networks (STGNNs) have made significant progress by combining sequential models with graph convolution networks. However, due to high complexity issues, STGNNs only focus on short-term traffic forecasting (e.g., 1-h ahead), while ignoring more practical long-term forecasting. In this paper, we make the first attempt to explore long-term traffic forecasting (e.g., 1-day ahead). To this end, we first reveal its unique challenges in exploiting multi-scale representations. Then, we propose a novel Hierarchical U-net TransFormer (HUTFormer) to address the issues of long-term traffic forecasting. HUTFormer consists of a hierarchical encoder and decoder to jointly generate and utilize multi-scale representations of traffic data. Specifically, for the encoder, we propose window self-attention and segment merging to extract multi-scale representations from long-term traffic data. For the decoder, we design a cross-scale attention mechanism to effectively incorporate multi-scale representations. In addition, HUTFormer employs an efficient input embedding strategy to address the complexity issues. Extensive experiments on four traffic datasets show that the proposed HUTFormer significantly outperforms state-of-the-art traffic forecasting and long time series forecasting baselines.</div></div>","PeriodicalId":100292,"journal":{"name":"Communications in Transportation Research","volume":"5 ","pages":"Article 100218"},"PeriodicalIF":14.5,"publicationDate":"2025-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145519370","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Optimal speed limit under multi-class user equilibrium: A prescriptive approach using mathematical programming 多类用户均衡下的最优速度限制：使用数学规划的规定性方法

IF 14.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-11-13 DOI: 10.1016/j.commtr.2025.100221

Xiao Lin , Ludovic Leclercq , Lóránt Tavasszy

In practice, speed limits on road networks are often determined pragmatically, which can give suboptimal solutions for traffic performance and unfair results for the underlying user classes. This study presents an elegant approach to determine optimal speed limits on a traffic network with asymmetric user classes under congested conditions, that minimizes individual user travel time and does justice to differences in economic importance. Existing prescriptive approaches typically lack one or more of these features, cannot guarantee optimality or are difficult to solve. We formulate a new prescriptive method using mixed-integer quadratic programming. The model can be solved with well-established operation research approaches and commercial solvers such as Cplex or Gurobi. To demonstrate the approach, we apply it to a regional network in the Netherlands. The result shows a reduction of travel time of passenger cars by 6% and of trucks by 13%, with mild changes in speed limits compared to the base situation, of between −20% and +10%. The speed limit changes and impacts are in line with the relatively high economic importance of freight traffic. Also we find in this case that the speed limit changes are ordered by major routes through the network, which makes implementation relatively straightforward.

在实践中，道路网络上的速度限制通常是实际确定的，这可能会给出交通性能的次优解决方案，并对底层用户类别造成不公平的结果。本研究提出了一种优雅的方法来确定在拥挤条件下具有非对称用户类别的交通网络上的最佳速度限制，该方法可以最大限度地减少个人用户的旅行时间，并公正地对待经济重要性的差异。现有的规定性方法通常缺乏这些特征中的一个或多个，不能保证最优性或难以解决。利用混合整数二次规划，提出了一种新的规定性方法。该模型可以用成熟的运筹学方法和商业求解器（如Cplex或Gurobi）来求解。为了演示该方法，我们将其应用于荷兰的一个区域网络。结果表明，与基本情况相比，乘用车的行驶时间减少了6%，卡车的行驶时间减少了13%，速度限制的变化在- 20%到+10%之间。限速的变化和影响与货运相对较高的经济重要性是一致的。我们还发现，在这种情况下，速度限制的变化是按通过网络的主要路线排序的，这使得实施相对简单。

{"title":"Optimal speed limit under multi-class user equilibrium: A prescriptive approach using mathematical programming","authors":"Xiao Lin , Ludovic Leclercq , Lóránt Tavasszy","doi":"10.1016/j.commtr.2025.100221","DOIUrl":"10.1016/j.commtr.2025.100221","url":null,"abstract":"<div><div>In practice, speed limits on road networks are often determined pragmatically, which can give suboptimal solutions for traffic performance and unfair results for the underlying user classes. This study presents an elegant approach to determine optimal speed limits on a traffic network with asymmetric user classes under congested conditions, that minimizes individual user travel time and does justice to differences in economic importance. Existing prescriptive approaches typically lack one or more of these features, cannot guarantee optimality or are difficult to solve. We formulate a new prescriptive method using mixed-integer quadratic programming. The model can be solved with well-established operation research approaches and commercial solvers such as Cplex or Gurobi. To demonstrate the approach, we apply it to a regional network in the Netherlands. The result shows a reduction of travel time of passenger cars by 6% and of trucks by 13%, with mild changes in speed limits compared to the base situation, of between −20% and +10%. The speed limit changes and impacts are in line with the relatively high economic importance of freight traffic. Also we find in this case that the speed limit changes are ordered by major routes through the network, which makes implementation relatively straightforward.</div></div>","PeriodicalId":100292,"journal":{"name":"Communications in Transportation Research","volume":"5 ","pages":"Article 100221"},"PeriodicalIF":14.5,"publicationDate":"2025-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145519371","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

STFC: Spatio-temporal formation control for connected and autonomous vehicles in multi-lane traffic STFC：多车道交通中网联与自动驾驶车辆的时空编队控制

IF 14.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-11-13 DOI: 10.1016/j.commtr.2025.100219

Jianghong Dong , Jiawei Wang , Mengchi Cai , Yibin Yang , Qing Xu , Jianqiang Wang , Keqiang Li

Formation control of Connected and Autonomous Vehicles (CAVs) has shown significant potential for improving traffic safety and efficiency in multi-lane traffic. However, previous work has primarily focused on spatial coordination while neglecting temporal optimization, which significantly limits their cooperation capability and practical applicability in real-world traffic. In this study, we propose a Spatio-Temporal Formation Control (STFC) method that integrates centralized formation generation with distributed trajectory planning. Precisely, we propose a graph-based formation maintenance representation, and show that the interlaced geometric structure is optimal for multi-lane formation. Then, we develop a distributed spatio-temporal joint formation trajectory planning method that simultaneously optimizes spatial positions and temporal duration, with consideration of multiple objectives such as formation maintenance and obstacle avoidance. Further, we design a polynomial vehicle-to-target assignment algorithm that inherently resolves conflicts. Simulation experiments demonstrate the superiority of our method over baselines in terms of formation maintenance and transition, achieving a 53% and 58% reduction in transition time and distance, respectively.

网联和自动驾驶车辆（cav）的编队控制在提高多车道交通的安全性和效率方面显示出巨大的潜力。然而，以往的研究主要集中在空间协调上，而忽略了时间优化，这极大地限制了它们的协同能力和在现实交通中的实际应用。在本研究中，我们提出了一种将集中式编队生成与分布式轨迹规划相结合的时空编队控制（STFC）方法。具体地说，我们提出了一种基于图的队形维护表示，并证明了交错的几何结构对于多车道队形是最优的。在此基础上，提出了一种兼顾编队维护和避障等多目标，同时优化空间位置和时间持续时间的分布式时空联合编队轨迹规划方法。此外，我们设计了一个多项式车辆到目标分配算法，该算法固有地解决了冲突。仿真实验表明，该方法在地层维护和过渡方面优于基线，分别减少了53%和58%的过渡时间和距离。

{"title":"STFC: Spatio-temporal formation control for connected and autonomous vehicles in multi-lane traffic","authors":"Jianghong Dong , Jiawei Wang , Mengchi Cai , Yibin Yang , Qing Xu , Jianqiang Wang , Keqiang Li","doi":"10.1016/j.commtr.2025.100219","DOIUrl":"10.1016/j.commtr.2025.100219","url":null,"abstract":"<div><div>Formation control of Connected and Autonomous Vehicles (CAVs) has shown significant potential for improving traffic safety and efficiency in multi-lane traffic. However, previous work has primarily focused on spatial coordination while neglecting temporal optimization, which significantly limits their cooperation capability and practical applicability in real-world traffic. In this study, we propose a Spatio-Temporal Formation Control (STFC) method that integrates centralized formation generation with distributed trajectory planning. Precisely, we propose a graph-based formation maintenance representation, and show that the interlaced geometric structure is optimal for multi-lane formation. Then, we develop a distributed spatio-temporal joint formation trajectory planning method that simultaneously optimizes spatial positions and temporal duration, with consideration of multiple objectives such as formation maintenance and obstacle avoidance. Further, we design a polynomial vehicle-to-target assignment algorithm that inherently resolves conflicts. Simulation experiments demonstrate the superiority of our method over baselines in terms of formation maintenance and transition, achieving a 53% and 58% reduction in transition time and distance, respectively.</div></div>","PeriodicalId":100292,"journal":{"name":"Communications in Transportation Research","volume":"5 ","pages":"Article 100219"},"PeriodicalIF":14.5,"publicationDate":"2025-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145519369","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

FollowGen: A scaled noise conditional diffusion model for car-following trajectory prediction FollowGen：一种用于汽车跟随轨迹预测的尺度噪声条件扩散模型

IF 14.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-10-16 DOI: 10.1016/j.commtr.2025.100215

Junwei You , Rui Gan , Weizhe Tang , Zilin Huang , Jiaxi Liu , Zhuoyu Jiang , Haotian Shi , Keshu Wu , Keke Long , Sicheng Fu , Sikai Chen , Bin Ran

Vehicle trajectory prediction is critical for advancing autonomous driving and advanced driver assistance systems (ADASs). Deep learning-based approaches, especially those using transformer-based and generative models, have significantly improved prediction accuracy by capturing complex, non-linear patterns in vehicle dynamics and traffic interactions. However, they often overlook detailed car-following behaviors and the inter-vehicle interactions essential for real-world driving, particularly in fully autonomous or mixed traffic scenarios. Moreover, existing generative approaches in trajectory prediction are inefficient at conditioning predictions on relevant constraints. To address these issues, this study proposes FollowGen, a novel scaled noise conditional diffusion model for car-following trajectory prediction. FollowGen incorporates detailed inter-vehicular interactions and car-following dynamics within a generative framework, enhancing both the accuracy and realism of the predicted trajectories. The model uses a novel pipeline to capture historical vehicle behaviors. It leverages a noise scaling conditioning strategy to scale the noise with encoded historical features within the forward diffusion process to ensure history-constrained noise transformation. A cross-attention-based transformer architecture is employed in the reverse process to model intricate inter-vehicle dependencies, effectively guiding the denoising process and enhancing prediction accuracy. Experimental results in various real-world driving scenarios demonstrate the state-of-the-art performance and robustness of the proposed method.

车辆轨迹预测对于推进自动驾驶和高级驾驶辅助系统（ADASs）至关重要。基于深度学习的方法，特别是那些使用基于变压器和生成模型的方法，通过捕获车辆动力学和交通相互作用中的复杂非线性模式，显著提高了预测精度。然而，他们往往忽略了详细的汽车跟随行为和车辆之间的互动，这对现实世界的驾驶至关重要，特别是在完全自动驾驶或混合交通场景中。此外，现有的轨迹预测生成方法在相关约束条件下的预测效率较低。为了解决这些问题，本研究提出了一种新型的缩放噪声条件扩散模型FollowGen，用于汽车跟随轨迹预测。FollowGen在生成框架内整合了详细的车辆间交互和车辆跟随动力学，提高了预测轨迹的准确性和真实感。该模型使用一种新颖的管道来捕获历史车辆行为。它利用噪声缩放调节策略在正向扩散过程中对具有编码历史特征的噪声进行缩放，以确保受历史约束的噪声转换。在反向过程中采用了基于交叉注意力的变压器结构，对复杂的车辆间依赖关系进行建模，有效地指导了去噪过程，提高了预测精度。各种真实驾驶场景的实验结果证明了该方法的性能和鲁棒性。

{"title":"FollowGen: A scaled noise conditional diffusion model for car-following trajectory prediction","authors":"Junwei You , Rui Gan , Weizhe Tang , Zilin Huang , Jiaxi Liu , Zhuoyu Jiang , Haotian Shi , Keshu Wu , Keke Long , Sicheng Fu , Sikai Chen , Bin Ran","doi":"10.1016/j.commtr.2025.100215","DOIUrl":"10.1016/j.commtr.2025.100215","url":null,"abstract":"<div><div>Vehicle trajectory prediction is critical for advancing autonomous driving and advanced driver assistance systems (ADASs). Deep learning-based approaches, especially those using transformer-based and generative models, have significantly improved prediction accuracy by capturing complex, non-linear patterns in vehicle dynamics and traffic interactions. However, they often overlook detailed car-following behaviors and the inter-vehicle interactions essential for real-world driving, particularly in fully autonomous or mixed traffic scenarios. Moreover, existing generative approaches in trajectory prediction are inefficient at conditioning predictions on relevant constraints. To address these issues, this study proposes FollowGen, a novel scaled noise conditional diffusion model for car-following trajectory prediction. FollowGen incorporates detailed inter-vehicular interactions and car-following dynamics within a generative framework, enhancing both the accuracy and realism of the predicted trajectories. The model uses a novel pipeline to capture historical vehicle behaviors. It leverages a noise scaling conditioning strategy to scale the noise with encoded historical features within the forward diffusion process to ensure history-constrained noise transformation. A cross-attention-based transformer architecture is employed in the reverse process to model intricate inter-vehicle dependencies, effectively guiding the denoising process and enhancing prediction accuracy. Experimental results in various real-world driving scenarios demonstrate the state-of-the-art performance and robustness of the proposed method.</div></div>","PeriodicalId":100292,"journal":{"name":"Communications in Transportation Research","volume":"5 ","pages":"Article 100215"},"PeriodicalIF":14.5,"publicationDate":"2025-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145319546","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Can combined virtual-real testing speed up autonomous vehicle testing? Findings from AEB field experiments 虚拟与现实的结合测试能否加快自动驾驶汽车的测试速度？AEB现场实验结果

IF 14.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-10-16 DOI: 10.1016/j.commtr.2025.100216

Meng Zhang, Jiatong Xu, Ying Gao, Dandan Shen, Zhigang Xu

Proving ground testing has become a standard methodology for the development and validation of autonomous vehicles in the automotive industry. However, it suffers from inherent limitations in efficiency, cost, and scenario coverage. Combined virtual-real testing (CVRT) offers a promising alternative by integrating virtual scenarios with physical vehicles and the environment, enhancing scenario coverage and test flexibility. Nevertheless, few studies have systematically investigated its effectiveness and applicability. To address this gap, this study develops a digital-twin-based CVRT system and conducts consistency verification experiments, taking the autonomous emergency braking (AEB) system test as a case study. Four typical scenarios selected from C-NCAP (China New Car Assessment Programme) 2024 were tested at speeds of 30, 40, and 50 km/h, utilizing both real-world and CVRT methods, with each experiment repeated 15 times. Vehicle dynamics data were collected, and the Fréchet distance metric was used to quantify similarity, whereas statistical hypothesis testing was used to assess differences in time-to-collision (TTC) trigger times. The results show that the average Fréchet distance ratio between the CVRT and real-world tests almost approaches 1.0, and the differences in the TTC trigger times were not statistically significant. However, the results of the simulation experiments differed significantly from those of the real-world tests (0.528 m/s in speed and 1.150 m/s² in acceleration higher than the CVRT). Additionally, the data communication delay between the CVRT platform and the physical autonomous vehicle under test remained well below tolerable thresholds. These results indicate high consistency between CVRT and real-world testing. Furthermore, CVRT achieved considerable improvements in testing efficiency, saving approximately 40%–70% compared with real-world testing.

试验场测试已经成为汽车行业开发和验证自动驾驶汽车的标准方法。然而，它在效率、成本和场景覆盖方面受到固有的限制。虚拟-真实组合测试（CVRT）通过将虚拟场景与物理车辆和环境集成，增强场景覆盖范围和测试灵活性，提供了一种很有前途的替代方案。然而，很少有研究系统地考察其有效性和适用性。为了解决这一空白，本研究以自动紧急制动（AEB）系统测试为例，开发了基于数字孪生的CVRT系统，并进行了一致性验证实验。在C-NCAP（中国新车评估计划）2024中选择了四种典型场景，分别在30,40和50 km/h的速度下进行了测试，使用真实世界和CVRT方法，每个实验重复15次。收集车辆动力学数据，使用fr距离度量来量化相似性，而使用统计假设检验来评估碰撞时间（TTC）触发时间的差异。结果表明，CVRT与真实世界测试的平均距离比几乎接近1.0，而TTC触发次数的差异无统计学意义。然而，模拟实验结果与实际测试结果存在显著差异（速度比CVRT高0.528 m/s，加速度比CVRT高1.150 m/s2）。此外，CVRT平台与测试中的物理自动驾驶车辆之间的数据通信延迟仍远低于可容忍的阈值。这些结果表明CVRT与实际测试之间具有高度的一致性。此外，CVRT在测试效率方面取得了相当大的提高，与实际测试相比节省了约40%-70%。

{"title":"Can combined virtual-real testing speed up autonomous vehicle testing? Findings from AEB field experiments","authors":"Meng Zhang, Jiatong Xu, Ying Gao, Dandan Shen, Zhigang Xu","doi":"10.1016/j.commtr.2025.100216","DOIUrl":"10.1016/j.commtr.2025.100216","url":null,"abstract":"<div><div>Proving ground testing has become a standard methodology for the development and validation of autonomous vehicles in the automotive industry. However, it suffers from inherent limitations in efficiency, cost, and scenario coverage. Combined virtual-real testing (CVRT) offers a promising alternative by integrating virtual scenarios with physical vehicles and the environment, enhancing scenario coverage and test flexibility. Nevertheless, few studies have systematically investigated its effectiveness and applicability. To address this gap, this study develops a digital-twin-based CVRT system and conducts consistency verification experiments, taking the autonomous emergency braking (AEB) system test as a case study. Four typical scenarios selected from C-NCAP (China New Car Assessment Programme) 2024 were tested at speeds of 30, 40, and 50 km/h, utilizing both real-world and CVRT methods, with each experiment repeated 15 times. Vehicle dynamics data were collected, and the Fréchet distance metric was used to quantify similarity, whereas statistical hypothesis testing was used to assess differences in time-to-collision (TTC) trigger times. The results show that the average Fréchet distance ratio between the CVRT and real-world tests almost approaches 1.0, and the differences in the TTC trigger times were not statistically significant. However, the results of the simulation experiments differed significantly from those of the real-world tests (0.528 m/s in speed and 1.150 m/s<sup>2</sup> in acceleration higher than the CVRT). Additionally, the data communication delay between the CVRT platform and the physical autonomous vehicle under test remained well below tolerable thresholds. These results indicate high consistency between CVRT and real-world testing. Furthermore, CVRT achieved considerable improvements in testing efficiency, saving approximately 40%–70% compared with real-world testing.</div></div>","PeriodicalId":100292,"journal":{"name":"Communications in Transportation Research","volume":"5 ","pages":"Article 100216"},"PeriodicalIF":14.5,"publicationDate":"2025-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145319540","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Domain-enhanced dual-branch model for efficient and interpretable accident anticipation 面向高效可解释事故预测的领域增强双分支模型

IF 14.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-10-14 DOI: 10.1016/j.commtr.2025.100214

Yanchen Guan , Haicheng Liao , Chengyue Wang , Bonan Wang , Jiaxun Zhang , Jia Hu , Zhenning Li

Developing precise and computationally efficient traffic accident anticipation system is crucial for contemporary autonomous driving technologies, enabling timely intervention and loss prevention. In this study, we propose an accident anticipation framework employing a dual-branch architecture that effectively integrates visual information from dashcam videos with structured textual data derived from accident reports. Furthermore, we introduce a feature aggregation method that facilitates seamless integration of multimodal inputs through large models (GPT-4o, Long-CLIP), complemented by targeted prompt engineering strategies to produce actionable feedback and standardized accident archives. Comprehensive evaluations conducted on benchmark datasets (Dashcam Accidents Dataset (DAD), Car Crash Dataset (CCD), and AnAn Accident Detection (A3D)) validate the superior predictive accuracy, enhanced responsiveness, reduced computational overhead, and improved interpretability of our approach, thus establishing a new benchmark for state-of-the-art performance in traffic accident anticipation.

开发精确且计算效率高的交通事故预测系统对于当代自动驾驶技术至关重要，可以及时干预和预防损失。在这项研究中，我们提出了一个采用双分支架构的事故预测框架，该框架有效地集成了来自行车记录仪视频的视觉信息和来自事故报告的结构化文本数据。此外，我们还引入了一种特征聚合方法，通过大型模型（gpt - 40、Long-CLIP）促进多模式输入的无缝集成，并辅以有针对性的快速工程策略，以产生可操作的反馈和标准化的事故档案。在基准数据集(行车记录仪事故数据集（DAD）、汽车碰撞数据集（CCD）和AnAn事故检测（A3D）上进行的综合评估验证了我们的方法具有卓越的预测准确性、增强的响应能力、减少的计算开销以及改进的可解释性，从而为交通事故预测的最先进性能建立了新的基准。

{"title":"Domain-enhanced dual-branch model for efficient and interpretable accident anticipation","authors":"Yanchen Guan , Haicheng Liao , Chengyue Wang , Bonan Wang , Jiaxun Zhang , Jia Hu , Zhenning Li","doi":"10.1016/j.commtr.2025.100214","DOIUrl":"10.1016/j.commtr.2025.100214","url":null,"abstract":"<div><div>Developing precise and computationally efficient traffic accident anticipation system is crucial for contemporary autonomous driving technologies, enabling timely intervention and loss prevention. In this study, we propose an accident anticipation framework employing a dual-branch architecture that effectively integrates visual information from dashcam videos with structured textual data derived from accident reports. Furthermore, we introduce a feature aggregation method that facilitates seamless integration of multimodal inputs through large models (GPT-4o, Long-CLIP), complemented by targeted prompt engineering strategies to produce actionable feedback and standardized accident archives. Comprehensive evaluations conducted on benchmark datasets (Dashcam Accidents Dataset (DAD), Car Crash Dataset (CCD), and AnAn Accident Detection (A3D)) validate the superior predictive accuracy, enhanced responsiveness, reduced computational overhead, and improved interpretability of our approach, thus establishing a new benchmark for state-of-the-art performance in traffic accident anticipation.</div></div>","PeriodicalId":100292,"journal":{"name":"Communications in Transportation Research","volume":"5 ","pages":"Article 100214"},"PeriodicalIF":14.5,"publicationDate":"2025-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145319539","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Corrigendum to “Interaction dataset of autonomous vehicles with traffic lights and signs”[Communications. Transp. Res. 5 (2025) 100201] “自动驾驶车辆与交通信号灯和标志的交互数据集”的勘误表[通信]。透明。Res. 5 (2025) 100201]

IF 14.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-10-13 DOI: 10.1016/j.commtr.2025.100217

Zheng Li , Zhipeng Bao , Haoming Meng , Haotian Shi , Qianwen Li , Handong Yao , Xiaopeng Li

引用次数: 0

The fundamental diagram of autonomous vehicles: Traffic state estimation and evidence from vehicle trajectories 自动驾驶汽车的基本图：交通状态估计和车辆轨迹证据

IF 14.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-10-10 DOI: 10.1016/j.commtr.2025.100212

Michail A. Makridis , Shaimaa K. El-Baklish , Anastasios Kouvelas , Jorge A. Laval

The fundamental diagram (FD) is a key tool in traffic flow theory, describing the relationship between traffic flow and density at the link level. Traditionally, FD estimation relies on data from static sensors, although vehicle trajectory data provides an alternative approach. Driver heterogeneity strongly influences the shape and scatter of the FD and is crucial for traffic management. Autonomous vehicles (AVs), exhibiting distinct driving behavior from human drivers, are expected to alter the FD. However, limited observations of AVs in stationary conditions have constrained research in this area. This study addresses this gap by introducing the platoon fundamental diagram (PFD), a simple method to infer empirical FDs from platoon trajectory data. PFD derives pseudo-states from vehicle trajectories and aggregates them to capture consistent relationships between flow, density, and speed—without requiring stationary conditions or backward wave speed estimation. The results highlight the impact of AVs on traffic flow capacity, driver heterogeneity, and oscillation propagation. Comparative analysis with human-driven experiments provides additional insights. Furthermore, the PFD's potential as a practical tool for traffic state estimation in mixed traffic conditions is demonstrated through real-world applications using NGSIM and I–24 Motion datasets.

基本图（FD）是交通流理论中的一个重要工具，它描述了交通流与交通密度之间的关系。传统上，FD估计依赖于静态传感器的数据，尽管车辆轨迹数据提供了另一种方法。驾驶员异质性强烈影响FD的形状和分布，对交通管理至关重要。自动驾驶汽车（AVs）表现出与人类驾驶员截然不同的驾驶行为，有望改变FD。然而，在固定条件下对自动驾驶汽车的有限观察限制了这一领域的研究。本研究通过引入从排轨迹数据推断经验fd的简单方法——排基本图（PFD）来解决这一问题。PFD从车辆轨迹中提取伪状态，并将它们聚合起来，以捕获流量、密度和速度之间的一致关系，而不需要固定条件或反向波速估计。研究结果强调了自动驾驶汽车对交通流容量、驾驶员异质性和振荡传播的影响。与人为实验的比较分析提供了更多的见解。此外，通过使用NGSIM和I-24运动数据集的实际应用，证明了PFD作为混合交通条件下交通状态估计的实用工具的潜力。

{"title":"The fundamental diagram of autonomous vehicles: Traffic state estimation and evidence from vehicle trajectories","authors":"Michail A. Makridis , Shaimaa K. El-Baklish , Anastasios Kouvelas , Jorge A. Laval","doi":"10.1016/j.commtr.2025.100212","DOIUrl":"10.1016/j.commtr.2025.100212","url":null,"abstract":"<div><div>The fundamental diagram (FD) is a key tool in traffic flow theory, describing the relationship between traffic flow and density at the link level. Traditionally, FD estimation relies on data from static sensors, although vehicle trajectory data provides an alternative approach. Driver heterogeneity strongly influences the shape and scatter of the FD and is crucial for traffic management. Autonomous vehicles (AVs), exhibiting distinct driving behavior from human drivers, are expected to alter the FD. However, limited observations of AVs in stationary conditions have constrained research in this area. This study addresses this gap by introducing the platoon fundamental diagram (PFD), a simple method to infer empirical FDs from platoon trajectory data. PFD derives pseudo-states from vehicle trajectories and aggregates them to capture consistent relationships between flow, density, and speed—without requiring stationary conditions or backward wave speed estimation. The results highlight the impact of AVs on traffic flow capacity, driver heterogeneity, and oscillation propagation. Comparative analysis with human-driven experiments provides additional insights. Furthermore, the PFD's potential as a practical tool for traffic state estimation in mixed traffic conditions is demonstrated through real-world applications using NGSIM and I–24 Motion datasets.</div></div>","PeriodicalId":100292,"journal":{"name":"Communications in Transportation Research","volume":"5 ","pages":"Article 100212"},"PeriodicalIF":14.5,"publicationDate":"2025-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145264828","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Minimum-delay opportunity charging scheduling for electricbuses 电动公交车最小延迟机会充电调度

IF 14.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-10-03 DOI: 10.1016/j.commtr.2025.100209

Dan McCabe , Xuegang (Jeff) Ban , Balaźs Kulcsár

Transit agencies that operate battery-electric buses must carefully manage fast-charging infrastructure to extend daily bus range without degrading on-time performance. To support this need, we propose a mixed-integer linear programming model to schedule opportunity charging that minimizes the amount of departure delay in all trips served by electric buses. Our novel approach directly tracks queuing at chargers in order to set and propagate departure delays. Allowing but minimizing delays makes it possible to optimize performance when delays due to traffic conditions and charging needs are inevitable, in contrast with existing methods that require charging to occur during scheduled layover time. To solve the model, we develop two algorithms based on decomposition. The first is an exact solution method based on combinatorial Benders (CB) decomposition, which avoids directly enumerating the model’s logic-based “big M” constraints and their inevitable computational challenges. The second, inspired by the CB approach but more efficient, is a polynomial-time heuristic based on linear programming that we call Select–Sequence–Schedule (3S). Computational experiments on both a simple notional transit network and the real bus system of King County, Washington, USA demonstrate the performance of both methods. The 3S method appears particularly promising for creating good charging schedules quickly at real-world scale.

运营纯电动公交车的公共交通机构必须仔细管理快速充电基础设施，以延长公交车的日常行驶里程，同时不降低准点率。为了支持这一需求，我们提出了一个混合整数线性规划模型来安排机会收费，以最大限度地减少电动公交车服务的所有行程的出发延误量。我们的新方法直接跟踪收费站的排队情况，以设置和传播出发延误。当交通状况和收费需求不可避免地造成延误时，允许但最小化延误使得优化性能成为可能，而现有的方法要求在计划的中途停留时间内进行收费。为了求解该模型，我们开发了两种基于分解的算法。第一种是基于组合Benders （CB）分解的精确解方法，该方法避免了直接枚举模型基于逻辑的“大M”约束及其不可避免的计算挑战。第二种方法受到CB方法的启发，但效率更高，是一种基于线性规划的多项式时间启发式方法，我们称之为选择-序列-调度（3S）。在美国华盛顿州金县一个简单的概念公交网络和实际公交系统上进行了计算实验，验证了这两种方法的有效性。3S方法似乎特别有希望在实际规模下快速创建良好的充电计划。

{"title":"Minimum-delay opportunity charging scheduling for electricbuses","authors":"Dan McCabe , Xuegang (Jeff) Ban , Balaźs Kulcsár","doi":"10.1016/j.commtr.2025.100209","DOIUrl":"10.1016/j.commtr.2025.100209","url":null,"abstract":"<div><div>Transit agencies that operate battery-electric buses must carefully manage fast-charging infrastructure to extend daily bus range without degrading on-time performance. To support this need, we propose a mixed-integer linear programming model to schedule opportunity charging that minimizes the amount of departure delay in all trips served by electric buses. Our novel approach directly tracks queuing at chargers in order to set and propagate departure delays. Allowing but minimizing delays makes it possible to optimize performance when delays due to traffic conditions and charging needs are inevitable, in contrast with existing methods that require charging to occur during scheduled layover time. To solve the model, we develop two algorithms based on decomposition. The first is an exact solution method based on combinatorial Benders (CB) decomposition, which avoids directly enumerating the model’s logic-based “big M” constraints and their inevitable computational challenges. The second, inspired by the CB approach but more efficient, is a polynomial-time heuristic based on linear programming that we call Select–Sequence–Schedule (3S). Computational experiments on both a simple notional transit network and the real bus system of King County, Washington, USA demonstrate the performance of both methods. The 3S method appears particularly promising for creating good charging schedules quickly at real-world scale.</div></div>","PeriodicalId":100292,"journal":{"name":"Communications in Transportation Research","volume":"5 ","pages":"Article 100209"},"PeriodicalIF":14.5,"publicationDate":"2025-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145219485","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Flying cars and urban air mobility: Redefining cities in three dimensions 飞行汽车和城市空中交通：在三维空间重新定义城市

IF 14.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-10-01 DOI: 10.1016/j.commtr.2025.100213

Di Lv , Kai Wang , Shulu Chen , Xiaobo Qu

引用次数: 0