Communications in Transportation Research最新文献_第4页

Enhanced trajectory reconstruction from sparse and noisy GPS data: A progressive chunked transformer approach 基于稀疏和噪声GPS数据的增强轨迹重建：一种渐进式分块变压器方法

IF 14.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-08-02 DOI: 10.1016/j.commtr.2025.100200

Yonghui Liu , Qian Li , Inhi Kim

Trajectory reconstruction from sparse and noisy GPS data is critical for applications such as urban mobility analysis, transportation planning, and navigation systems. However, large sampling intervals and the typically long output sequences required to reconstruct coherent travel trajectories significantly increase computational complexity, particularly in the presence of noise. To address these challenges, we propose a progressive chunked transformer (ProChunkFormer), which is a deep learning method for trajectory reconstruction that employs self-attention mechanisms and chunked processing to balance efficiency with accuracy. ProChunkFormer first generates intermediate trajectories at a semi-high frequency from low-frequency sampled data, and then the remaining trajectory is divided into manageable blocks and reconstructed parallelly in the condition of the semi-high-frequency trajectory. By combining progressive reconstruction with chunk processing, ProChunkFormer not only mitigates the cumulative errors commonly observed in autoregressive models but also alleviates the rapid increase in complexity associated with reconstructing ultralong trajectories. Specifically, our approach achieves quadratic optimization in time and space for attention modules, with cubic time savings compared with autoregressive decoding. A case study using an open-source taxi trajectory dataset confirms the effectiveness of our approach. The performance of ProChunkFormer is comparable to that of autoregressive transformers while offering better running efficiency. It improves the accuracy, F1 score (F1), mean absolute error (MAE), and road network mean absolute error (MAE_RN) by 23.1%, 18.6%, 22.3%, and 25.1%, respectively, for trajectories with a long interval time of up to 240 s. Furthermore, we investigate incorporating heuristic information to guide trajectory reconstruction for each block. The experimental results indicate an improvement in both the overall performance and convergence speed of the model.

从稀疏和噪声GPS数据中重建轨迹对于城市交通分析、交通规划和导航系统等应用至关重要。然而，重建相干旅行轨迹所需的大采样间隔和典型的长输出序列显着增加了计算复杂性，特别是在存在噪声的情况下。为了解决这些挑战，我们提出了一种渐进式分块变压器（ProChunkFormer），这是一种用于轨迹重建的深度学习方法，采用自注意机制和分块处理来平衡效率和准确性。ProChunkFormer首先从低频采样数据中生成半高频的中间轨迹，然后将剩余的轨迹划分为可管理的块，并在半高频轨迹条件下并行重构。通过将渐进式重建与块处理相结合，ProChunkFormer不仅减轻了自回归模型中常见的累积误差，而且还减轻了重建超长轨迹时复杂性的快速增加。具体来说，我们的方法在时间和空间上实现了注意力模块的二次优化，与自回归解码相比节省了三次时间。一个使用开源出租车轨迹数据集的案例研究证实了我们方法的有效性。ProChunkFormer的性能可与自回归变压器相媲美，同时提供更好的运行效率。对于长达240 s的长间隔时间的轨迹，该算法将准确率、F1分数（F1）、平均绝对误差（MAE）和路网平均绝对误差（MAE_RN）分别提高了23.1%、18.6%、22.3%和25.1%。此外，我们研究了结合启发式信息来指导每个块的轨迹重建。实验结果表明，该模型的综合性能和收敛速度均有提高。

{"title":"Enhanced trajectory reconstruction from sparse and noisy GPS data: A progressive chunked transformer approach","authors":"Yonghui Liu , Qian Li , Inhi Kim","doi":"10.1016/j.commtr.2025.100200","DOIUrl":"10.1016/j.commtr.2025.100200","url":null,"abstract":"<div><div>Trajectory reconstruction from sparse and noisy GPS data is critical for applications such as urban mobility analysis, transportation planning, and navigation systems. However, large sampling intervals and the typically long output sequences required to reconstruct coherent travel trajectories significantly increase computational complexity, particularly in the presence of noise. To address these challenges, we propose a progressive chunked transformer (ProChunkFormer), which is a deep learning method for trajectory reconstruction that employs self-attention mechanisms and chunked processing to balance efficiency with accuracy. ProChunkFormer first generates intermediate trajectories at a semi-high frequency from low-frequency sampled data, and then the remaining trajectory is divided into manageable blocks and reconstructed parallelly in the condition of the semi-high-frequency trajectory. By combining progressive reconstruction with chunk processing, ProChunkFormer not only mitigates the cumulative errors commonly observed in autoregressive models but also alleviates the rapid increase in complexity associated with reconstructing ultralong trajectories. Specifically, our approach achieves quadratic optimization in time and space for attention modules, with cubic time savings compared with autoregressive decoding. A case study using an open-source taxi trajectory dataset confirms the effectiveness of our approach. The performance of ProChunkFormer is comparable to that of autoregressive transformers while offering better running efficiency. It improves the accuracy, F1 score (F1), mean absolute error (MAE), and road network mean absolute error (MAE_RN) by 23.1%, 18.6%, 22.3%, and 25.1%, respectively, for trajectories with a long interval time of up to 240 s. Furthermore, we investigate incorporating heuristic information to guide trajectory reconstruction for each block. The experimental results indicate an improvement in both the overall performance and convergence speed of the model.</div></div>","PeriodicalId":100292,"journal":{"name":"Communications in Transportation Research","volume":"5 ","pages":"Article 100200"},"PeriodicalIF":14.5,"publicationDate":"2025-08-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144757300","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

CCDSReFormer: Traffic flow prediction with a criss-crossed dual-stream enhanced rectified transformer model CCDSReFormer：交通流量预测与交叉双流增强整流变压器模型

IF 12.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-07-23 DOI: 10.1016/j.commtr.2025.100189

Zhiqi Shao , Michael G.H. Bell , Ze Wang , D. Glenn Geers , Xusheng Yao , Junbin Gao

Accurate, efficient, and rapid traffic forecasting is essential for intelligent transportation systems and plays a pivotal role in urban traffic planning, management, and control. While existing spatiotemporal transformer models have demonstrated effectiveness in traffic flow prediction, they face notable challenges in achieving a balance between computational efficiency and accuracy. Additionally, they often prioritize global trends over local time series information and treat spatial and temporal data separately, limiting their ability to capture complex spatiotemporal interactions. To overcome these limitations, we propose the criss-crossed dual-stream enhanced rectified transformer (CCDSReFormer). This model introduces a novel rectified linear self-attention (ReLSA) mechanism combined with enhanced convolution (EnCov) to reduce computational overhead and sharpen the local feature focus. Furthermore, our cross-learning strategy seamlessly integrates spatial and temporal data, improving the model's ability to capture intricate traffic dynamics. Extensive experiments on six real-world datasets show that CCDSReFormer outperforms existing models in both accuracy and efficiency. An ablation study further validates the contributions of each component, confirming the model's superior ability to forecast traffic flow accurately and efficiently.

准确、高效、快速的交通预测是智能交通系统的基础，在城市交通规划、管理和控制中起着至关重要的作用。虽然现有的时空变换模型在交通流预测中已经证明了有效性，但它们在实现计算效率和准确性之间的平衡方面面临着显著的挑战。此外，它们往往优先考虑全球趋势而不是本地时间序列信息，并分别处理空间和时间数据，从而限制了它们捕捉复杂时空相互作用的能力。为了克服这些限制，我们提出了交叉双流增强整流变压器（CCDSReFormer）。该模型引入了一种新的校正线性自注意（ReLSA）机制，结合增强卷积（EnCov）来减少计算开销并锐化局部特征焦点。此外，我们的交叉学习策略无缝地集成了空间和时间数据，提高了模型捕捉复杂交通动态的能力。在六个真实数据集上的大量实验表明，CCDSReFormer在精度和效率方面都优于现有模型。消融研究进一步验证了各分量的贡献，证实了该模型准确有效地预测交通流量的优越能力。

{"title":"CCDSReFormer: Traffic flow prediction with a criss-crossed dual-stream enhanced rectified transformer model","authors":"Zhiqi Shao , Michael G.H. Bell , Ze Wang , D. Glenn Geers , Xusheng Yao , Junbin Gao","doi":"10.1016/j.commtr.2025.100189","DOIUrl":"10.1016/j.commtr.2025.100189","url":null,"abstract":"<div><div>Accurate, efficient, and rapid traffic forecasting is essential for intelligent transportation systems and plays a pivotal role in urban traffic planning, management, and control. While existing spatiotemporal transformer models have demonstrated effectiveness in traffic flow prediction, they face notable challenges in achieving a balance between computational efficiency and accuracy. Additionally, they often prioritize global trends over local time series information and treat spatial and temporal data separately, limiting their ability to capture complex spatiotemporal interactions. To overcome these limitations, we propose the criss-crossed dual-stream enhanced rectified transformer (CCDSReFormer). This model introduces a novel rectified linear self-attention (ReLSA) mechanism combined with enhanced convolution (EnCov) to reduce computational overhead and sharpen the local feature focus. Furthermore, our cross-learning strategy seamlessly integrates spatial and temporal data, improving the model's ability to capture intricate traffic dynamics. Extensive experiments on six real-world datasets show that CCDSReFormer outperforms existing models in both accuracy and efficiency. An ablation study further validates the contributions of each component, confirming the model's superior ability to forecast traffic flow accurately and efficiently.</div></div>","PeriodicalId":100292,"journal":{"name":"Communications in Transportation Research","volume":"5 ","pages":"Article 100189"},"PeriodicalIF":12.5,"publicationDate":"2025-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144686069","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

Advancing transportation research: interdisciplinary insights from emerging technologies and diverse perspectives 推进交通研究：来自新兴技术和不同视角的跨学科见解

IF 12.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-07-22 DOI: 10.1016/j.commtr.2025.100199

Mingyang Pei , Zhuoyan Wei , Xin Pei , Yu Zhang , Xiaokun Cara Wang , Yang Liu , Ronghui Liu

引用次数: 0

Compensation scheme and split delivery in a collaborative passenger-parcel transportation system 客包协同运输系统中的补偿方案与分送

IF 12.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-07-19 DOI: 10.1016/j.commtr.2025.100197

Yitong Yu, Kechen Ouyang, Qingyun Tian, David Z.W. Wang

The emerging collaborative passenger-parcel transport (CPT) mode aims to address the significant imbalance between passenger and parcel transport demand for last-mile delivery. By enabling passengers and parcels to share a single vehicle’s capacity, CPT reduces resource underutilization during off-peak hours and alleviates traffic congestion during peak hours. However, the successful implementation of such systems is not guaranteed, as passengers may decline shared rides due to reduced service quality. Compensation mechanisms, which incentivize passengers’ acceptance, offer a promising solution to such an issue. However, the design of optimal compensation scheme has not yet been investigated in the existing literature of collaborative transport. To fill this gap, this study incorporates compensation-affected behavior into a typical routing problem of the CPT system, where the routing problem allows delivery requests to be split across multiple trips and permits multiple visits to each node. We formulate this problem as the compensation scheme design in split delivery vehicle routing problem with time windows for a collaborative passenger-parcel transport system (C-SDVRPTW-CPT). We solve it by developing a Surrogate-based Adaptive Large Neighborhood Search framework (SOT-ALNS). Numerical experiments validate the model and algorithm, demonstrating the fast convergence of the algorithm and the advantages of collaborative transport and compensation, which improves profit by 3%–10%.

新兴的客包裹协同运输（CPT）模式旨在解决乘客和包裹在最后一英里运输需求之间的严重不平衡。通过使乘客和包裹共享一辆车的容量，CPT减少了非高峰时段的资源利用不足，缓解了高峰时段的交通拥堵。然而，这种系统的成功实施并不能保证，因为乘客可能会因为服务质量下降而拒绝共享乘车。激励乘客接受的补偿机制为这一问题提供了一个有希望的解决方案。然而，在现有的协同运输文献中，尚未对最优补偿方案的设计进行研究。为了填补这一空白，本研究将补偿影响行为纳入CPT系统的典型路由问题，其中路由问题允许将交付请求拆分为多个行程，并允许对每个节点进行多次访问。我们将此问题表述为客包协同运输系统（C-SDVRPTW-CPT）中带时间窗的分运车辆路线问题的补偿方案设计。我们通过开发一个基于代理的自适应大邻域搜索框架（SOT-ALNS）来解决这个问题。数值实验验证了模型和算法的有效性，证明了算法的快速收敛性以及协同运输和补偿的优势，可使利润提高3% ~ 10%。

{"title":"Compensation scheme and split delivery in a collaborative passenger-parcel transportation system","authors":"Yitong Yu, Kechen Ouyang, Qingyun Tian, David Z.W. Wang","doi":"10.1016/j.commtr.2025.100197","DOIUrl":"10.1016/j.commtr.2025.100197","url":null,"abstract":"<div><div>The emerging collaborative passenger-parcel transport (CPT) mode aims to address the significant imbalance between passenger and parcel transport demand for last-mile delivery. By enabling passengers and parcels to share a single vehicle’s capacity, CPT reduces resource underutilization during off-peak hours and alleviates traffic congestion during peak hours. However, the successful implementation of such systems is not guaranteed, as passengers may decline shared rides due to reduced service quality. Compensation mechanisms, which incentivize passengers’ acceptance, offer a promising solution to such an issue. However, the design of optimal compensation scheme has not yet been investigated in the existing literature of collaborative transport. To fill this gap, this study incorporates compensation-affected behavior into a typical routing problem of the CPT system, where the routing problem allows delivery requests to be split across multiple trips and permits multiple visits to each node. We formulate this problem as the compensation scheme design in split delivery vehicle routing problem with time windows for a collaborative passenger-parcel transport system (C-SDVRPTW-CPT). We solve it by developing a Surrogate-based Adaptive Large Neighborhood Search framework (SOT-ALNS). Numerical experiments validate the model and algorithm, demonstrating the fast convergence of the algorithm and the advantages of collaborative transport and compensation, which improves profit by 3%–10%.</div></div>","PeriodicalId":100292,"journal":{"name":"Communications in Transportation Research","volume":"5 ","pages":"Article 100197"},"PeriodicalIF":12.5,"publicationDate":"2025-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144662810","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

Strategic roles of female scholars in steering transportation research agendas 女性学者在引导交通研究议程中的战略作用

IF 12.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-07-18 DOI: 10.1016/j.commtr.2025.100198

Mingyang Pei , Zisen Lin , Xiao Fu , Xin Pei

In recent years, female scientists have contributed to advancements in the transportation sector through technological innovation and unique perspectives, playing pivotal roles across various domains of the field. This study analyzes 54,511 publications from 20 Science Citation Index (SCI) Q1 transportation journals (2014–2024), encompassing over 100,000 scholars, to advance the understanding of the status of female scientists in transportation academia. Female authors constitute only 22.91% of first authors and 20.86% of corresponding authors, revealing persistent underrepresentation despite incremental progress in mixed-gender collaborations. This study uses a mixed-methods framework that includes data mining, the mean normalized log-transformed citation score (MNLCS), probabilistic gender identification, keyword co-occurrence, and clustering analysis to investigate macrolevel trends and longitudinally compare four collaboration modes. The key findings include that (1) mixed-gender teams exhibit significant growth, with MNLCS exceeding single-gender teams by 0.048–0.067, and (2) female-led collaborations exhibit a stronger tendency to drive sustained exploration in research fields. These findings support gender-equality policies and guide early-career scholars in collaboration strategies and frontier tracking, promoting inclusive development in transportation research.

近年来，女性科学家通过技术创新和独特的视角为交通运输领域的进步做出了贡献，在该领域的各个领域发挥了关键作用。本研究分析了2014-2024年间20种SCI Q1交通期刊的54,511篇论文，涵盖10万多名学者，旨在提高对女性科学家在交通学界地位的认识。女性作者仅占第一作者的22.91%和通讯作者的20.86%，尽管混合性别合作取得了渐进式的进展，但女性作者的比例仍然不足。本研究采用混合方法框架，包括数据挖掘、平均归一化日志转换引文评分（MNLCS）、概率性别识别、关键词共现和聚类分析，研究宏观层面的趋势，并纵向比较四种协作模式。主要发现包括：(1)混合性别团队呈现显著增长，MNLCS比单一性别团队高出0.048-0.067;(2)女性领导的合作在推动研究领域的持续探索方面表现出更强的趋势。这些研究结果支持性别平等政策，并指导早期职业学者制定合作战略和前沿跟踪，促进交通研究的包容性发展。

{"title":"Strategic roles of female scholars in steering transportation research agendas","authors":"Mingyang Pei , Zisen Lin , Xiao Fu , Xin Pei","doi":"10.1016/j.commtr.2025.100198","DOIUrl":"10.1016/j.commtr.2025.100198","url":null,"abstract":"<div><div>In recent years, female scientists have contributed to advancements in the transportation sector through technological innovation and unique perspectives, playing pivotal roles across various domains of the field. This study analyzes 54,511 publications from 20 Science Citation Index (SCI) Q1 transportation journals (2014–2024), encompassing over 100,000 scholars, to advance the understanding of the status of female scientists in transportation academia. Female authors constitute only 22.91% of first authors and 20.86% of corresponding authors, revealing persistent underrepresentation despite incremental progress in mixed-gender collaborations. This study uses a mixed-methods framework that includes data mining, the mean normalized log-transformed citation score (MNLCS), probabilistic gender identification, keyword co-occurrence, and clustering analysis to investigate macrolevel trends and longitudinally compare four collaboration modes. The key findings include that (1) mixed-gender teams exhibit significant growth, with MNLCS exceeding single-gender teams by 0.048–0.067, and (2) female-led collaborations exhibit a stronger tendency to drive sustained exploration in research fields. These findings support gender-equality policies and guide early-career scholars in collaboration strategies and frontier tracking, promoting inclusive development in transportation research.</div></div>","PeriodicalId":100292,"journal":{"name":"Communications in Transportation Research","volume":"5 ","pages":"Article 100198"},"PeriodicalIF":12.5,"publicationDate":"2025-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144654927","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

A hybrid centralized-decentralized traffic control framework for unmanned aerial vehicles in urban low-altitude airspace 城市低空空域无人机集中-分散混合交通控制框架

IF 12.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-07-17 DOI: 10.1016/j.commtr.2025.100195

Xiangdong Chen , Shen Li , Meng Li

Urban air mobility (UAM) represents a transformative approach to alleviating ground-level congestion by transitioning from two-dimension (2D) to three-dimension (3D) transportation systems. Envisioned as a safe, sustainable, and efficient mode of urban transit, UAM leverages aerial space to reduce dependence on traditional road infrastructure while addressing traffic congestion challenges in urban mobility. However, the rapid growth in aerospace transportation demand, coupled with the complexity of managing large-scale unmanned aerial vehicle (UAV) operations in 3D airspace, challenges the effectiveness of traditional traffic management systems. To address these challenges, this study proposes a hybrid framework for UAV air traffic control that integrates centralized and decentralized approaches. A 3D air traffic network is modeled in low-altitude airspace, capturing detailed 2D and 3D conflict relationships. The concept of a “virtual flight container” (VFC) is introduced to regulate UAV space–time trajectories, ensuring conflict-free, low-delay operations while minimizing real-time computational requirements, especially in high demands. The problem is addressed using a bi-level optimization approach: The upper level focuses on solving the traffic assignment problem, considering airway capacity constraints, while the lower level designs space–time trajectories to ensure conflict-free operations and enhance traffic efficiency, thereby complementing the traffic control scheme. Numerical experiments validate the proposed framework, highlighting its effectiveness in improving traffic efficiency and network throughput. Key insights are provided regarding the role of network structure, the placement of take-off and landing points, and control parameters in optimizing UAM operations.

城市空中交通（UAM）代表了一种通过从二维（2D）到三维（3D）交通系统过渡来缓解地面拥堵的变革性方法。UAM被设想为一种安全、可持续和高效的城市交通模式，它利用空中空间来减少对传统道路基础设施的依赖，同时解决城市交通中的交通拥堵挑战。然而，航空运输需求的快速增长，加上在三维空域管理大规模无人机（UAV）操作的复杂性，对传统交通管理系统的有效性提出了挑战。为了应对这些挑战，本研究提出了一种集成集中式和分散式方法的无人机空中交通管制混合框架。在低空空域建模三维空中交通网络，捕获详细的二维和三维冲突关系。引入了“虚拟飞行容器”（VFC）的概念来调节无人机的时空轨迹，确保无冲突、低延迟的操作，同时最小化实时计算需求，特别是在高需求时。该问题采用双层优化方法解决：上层着重解决交通分配问题，考虑航路容量约束；下层设计时空轨迹，保证无冲突运行，提高交通效率，与交通管制方案形成互补。数值实验验证了该框架的有效性，证明了该框架在提高网络传输效率和吞吐量方面的有效性。提供了关于网络结构的作用、起飞和降落点的位置以及优化UAM操作的控制参数的关键见解。

{"title":"A hybrid centralized-decentralized traffic control framework for unmanned aerial vehicles in urban low-altitude airspace","authors":"Xiangdong Chen , Shen Li , Meng Li","doi":"10.1016/j.commtr.2025.100195","DOIUrl":"10.1016/j.commtr.2025.100195","url":null,"abstract":"<div><div>Urban air mobility (UAM) represents a transformative approach to alleviating ground-level congestion by transitioning from two-dimension (2D) to three-dimension (3D) transportation systems. Envisioned as a safe, sustainable, and efficient mode of urban transit, UAM leverages aerial space to reduce dependence on traditional road infrastructure while addressing traffic congestion challenges in urban mobility. However, the rapid growth in aerospace transportation demand, coupled with the complexity of managing large-scale unmanned aerial vehicle (UAV) operations in 3D airspace, challenges the effectiveness of traditional traffic management systems. To address these challenges, this study proposes a hybrid framework for UAV air traffic control that integrates centralized and decentralized approaches. A 3D air traffic network is modeled in low-altitude airspace, capturing detailed 2D and 3D conflict relationships. The concept of a “virtual flight container” (VFC) is introduced to regulate UAV space–time trajectories, ensuring conflict-free, low-delay operations while minimizing real-time computational requirements, especially in high demands. The problem is addressed using a bi-level optimization approach: The upper level focuses on solving the traffic assignment problem, considering airway capacity constraints, while the lower level designs space–time trajectories to ensure conflict-free operations and enhance traffic efficiency, thereby complementing the traffic control scheme. Numerical experiments validate the proposed framework, highlighting its effectiveness in improving traffic efficiency and network throughput. Key insights are provided regarding the role of network structure, the placement of take-off and landing points, and control parameters in optimizing UAM operations.</div></div>","PeriodicalId":100292,"journal":{"name":"Communications in Transportation Research","volume":"5 ","pages":"Article 100195"},"PeriodicalIF":12.5,"publicationDate":"2025-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144654928","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

Should autonomous vehicles be subsidized to reduce parking fees? A productivity perspective 自动驾驶汽车应该得到补贴以降低停车费吗？生产力视角

IF 12.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-07-16 DOI: 10.1016/j.commtr.2025.100196

Yao Li , Ziyue Yang , Tao Wang , Shuxian Xu , Jiancheng Long

Governments often advocate for and implement policies to promote the development of new technologies, such as electric vehicles. Are these policies promoting new mobility modes applicable to autonomous vehicles (AVs)? In this study, we develop an economic model to capture residents' behaviors, including mode choice, location choice, and parking choice. Two parking choices (parking downtown or at home) for AV users are considered. We construct utility maximization models under a user equilibrium state to capture government planning and residents' choices. By deriving the first-order conditions of the model, we analyze the influence of AVs on urban characteristics. We emphasize how the parking subsidy affects AV users’ behavior, thereby influencing urban productivity. The results indicate that parking subsidies for AVs undermine urban productivity, whereas cash-out policies, such as providing subsidies for public transit, can effectively enhance urban productivity.

政府经常倡导和实施促进新技术发展的政策，如电动汽车。这些促进新出行模式的政策是否适用于自动驾驶汽车？在本研究中，我们建立了一个经济模型来捕捉居民的行为，包括模式选择、位置选择和停车选择。为自动驾驶汽车用户提供两种停车选择（市中心或家中）。我们构建了用户均衡状态下的效用最大化模型来捕捉政府规划和居民选择。通过推导模型的一阶条件，分析了自动驾驶汽车对城市特征的影响。我们强调停车补贴如何影响自动驾驶汽车用户的行为，从而影响城市生产力。研究结果表明，自动驾驶汽车停车补贴对城市生产力造成了损害，而公共交通补贴等套现政策则能有效提高城市生产力。

引用次数: 0

Few-shot learning for novel object detection in autonomous driving 基于少镜头学习的自动驾驶新目标检测

IF 12.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-07-16 DOI: 10.1016/j.commtr.2025.100194

Yifan Zhuang , Pei Liu , Hao Yang , Kai Zhang , Yinhai Wang , Ziyuan Pu

Artificial intelligence and advanced sensing technologies have significantly advanced the intelligent transportation system and autonomous vehicles. Perception, a critical component, extracts real-time traffic information essential for various system functionalities, such as agent behavior prediction. However, the quality of information derived from perception greatly influences overall system performance. This study focuses on enhancing perception robustness in autonomous vehicles, particularly in detecting rare objects, which pose a challenge due to limited training samples. While deep learning-based vision methods have shown promising accuracy, they struggle with rare object detection. To address this, we propose a few-shot learning training strategy tailored for improved detection accuracy of rare or novel objects. Additionally, we design a one-stage object detector for efficient object detection in autonomous driving scenarios. Experiments on a self-driving dataset augmented with rare objects alongside the popular few-shot object detection (FSOD) benchmark, the pattern analysis, statical modeling, and computational learning PASCAL Visual Object Classes (PASCAL-VOC), demonstrate state-of-the-art accuracy in rare categories and superior inference speed compared to alternative algorithms. Furthermore, we investigate the impact of intra-class variance on detection accuracy, providing insights for data annotation in the preparation stage.

人工智能和先进的传感技术极大地推动了智能交通系统和自动驾驶汽车的发展。感知是一个关键的组成部分，它提取实时交通信息，这些信息对各种系统功能至关重要，例如代理行为预测。然而，从感知中获得的信息质量极大地影响了系统的整体性能。本研究的重点是增强自动驾驶汽车的感知鲁棒性，特别是在检测稀有物体方面，由于训练样本有限，这构成了挑战。虽然基于深度学习的视觉方法显示出了很好的准确性，但它们在罕见的物体检测方面仍存在困难。为了解决这个问题，我们提出了一种专为提高稀有或新物体的检测精度而定制的少量学习训练策略。此外，我们设计了一种单阶段目标检测器，用于自动驾驶场景下的高效目标检测。在带有稀有物体的自动驾驶数据集上进行的实验，以及流行的少量物体检测（FSOD）基准、模式分析、静态建模和计算学习PASCAL视觉对象类（PASCAL- voc），证明了与其他算法相比，稀有类别的最先进精度和卓越的推理速度。此外，我们还研究了类内方差对检测精度的影响，为准备阶段的数据注释提供了见解。

{"title":"Few-shot learning for novel object detection in autonomous driving","authors":"Yifan Zhuang , Pei Liu , Hao Yang , Kai Zhang , Yinhai Wang , Ziyuan Pu","doi":"10.1016/j.commtr.2025.100194","DOIUrl":"10.1016/j.commtr.2025.100194","url":null,"abstract":"<div><div>Artificial intelligence and advanced sensing technologies have significantly advanced the intelligent transportation system and autonomous vehicles. Perception, a critical component, extracts real-time traffic information essential for various system functionalities, such as agent behavior prediction. However, the quality of information derived from perception greatly influences overall system performance. This study focuses on enhancing perception robustness in autonomous vehicles, particularly in detecting rare objects, which pose a challenge due to limited training samples. While deep learning-based vision methods have shown promising accuracy, they struggle with rare object detection. To address this, we propose a few-shot learning training strategy tailored for improved detection accuracy of rare or novel objects. Additionally, we design a one-stage object detector for efficient object detection in autonomous driving scenarios. Experiments on a self-driving dataset augmented with rare objects alongside the popular few-shot object detection (FSOD) benchmark, the pattern analysis, statical modeling, and computational learning PASCAL Visual Object Classes (PASCAL-VOC), demonstrate state-of-the-art accuracy in rare categories and superior inference speed compared to alternative algorithms. Furthermore, we investigate the impact of intra-class variance on detection accuracy, providing insights for data annotation in the preparation stage.</div></div>","PeriodicalId":100292,"journal":{"name":"Communications in Transportation Research","volume":"5 ","pages":"Article 100194"},"PeriodicalIF":12.5,"publicationDate":"2025-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144633711","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

Urban visual clusters and road transport fatalities: A global city-level image analysis 城市视觉集群与道路交通死亡：全球城市级图像分析

IF 12.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-07-03 DOI: 10.1016/j.commtr.2025.100193

Zhuangyuan Fan, Becky P.Y. Loo

Road traffic crashes are among the leading causes of death and injury worldwide. While urban planning and design are known to influence road safety, it is not clear how features of the built environment contribute to traffic fatalities. In this study, we analyze road fatality data from 106 cities across six continents via a combination of computer vision and unsupervised clustering on 26.8 million Google Street View images. We use deep learning tools to extract 25 features from the images. Among these features, 19 are relatively static built environment features, and 6 are dynamic usage-related features (such as pedestrians, cars, buses, and bikes). On the basis of the built environment features, we group the urban streetscapes into six distinct visual clusters. We then examine how these clusters are related to city-level road fatality rates when various control variables (e.g., population size, carbon emissions, income, road length, road safety policy, and continent) and dynamic features are combined. Our findings show that cities with Open Arterials streetscape (extensive road surface, open-sky views, and railings) tend to have higher road fatality rates. After accounting for differences in the built environment, cities with better public transit (proxied by buses detected) tend to have fewer traffic deaths—specifically, a 1% increase in bus presence is linked to a 0.35% decrease in fatalities per 100,000 people. This study demonstrates the power of using widely available street view imagery to uncover global disparities in urban design and their connection to road safety.

道路交通碰撞是全世界造成死亡和伤害的主要原因之一。众所周知，城市规划和设计会影响道路安全，但目前尚不清楚建筑环境的特点如何导致交通事故死亡。在这项研究中，我们通过计算机视觉和无监督聚类的结合，对2680万谷歌街景图像分析了来自六大洲106个城市的道路死亡数据。我们使用深度学习工具从图像中提取了25个特征。在这些特征中，19个是相对静态的建筑环境特征，6个是与使用相关的动态特征（如行人、汽车、公共汽车和自行车）。根据建筑环境特征，将城市街景划分为6个不同的视觉集群。然后，我们研究了在各种控制变量（如人口规模、碳排放、收入、道路长度、道路安全政策和大陆）和动态特征相结合时，这些集群与城市一级道路死亡率的关系。我们的研究结果表明，具有开放动脉街景（宽阔的路面、开阔的天空景观和栏杆）的城市往往具有更高的道路死亡率。在考虑了建筑环境的差异之后，拥有更好的公共交通（以检测到的公交车为代表）的城市往往交通死亡人数更少——具体来说，每10万人中公交车数量增加1%，死亡人数就会减少0.35%。本研究展示了使用广泛可用的街景图像来揭示城市设计的全球差异及其与道路安全的联系的力量。

{"title":"Urban visual clusters and road transport fatalities: A global city-level image analysis","authors":"Zhuangyuan Fan, Becky P.Y. Loo","doi":"10.1016/j.commtr.2025.100193","DOIUrl":"10.1016/j.commtr.2025.100193","url":null,"abstract":"<div><div>Road traffic crashes are among the leading causes of death and injury worldwide. While urban planning and design are known to influence road safety, it is not clear how features of the built environment contribute to traffic fatalities. In this study, we analyze road fatality data from 106 cities across six continents via a combination of computer vision and unsupervised clustering on 26.8 million Google Street View images. We use deep learning tools to extract 25 features from the images. Among these features, 19 are relatively static built environment features, and 6 are dynamic usage-related features (such as pedestrians, cars, buses, and bikes). On the basis of the built environment features, we group the urban streetscapes into six distinct visual clusters. We then examine how these clusters are related to city-level road fatality rates when various control variables (e.g., population size, carbon emissions, income, road length, road safety policy, and continent) and dynamic features are combined. Our findings show that cities with Open Arterials streetscape (extensive road surface, open-sky views, and railings) tend to have higher road fatality rates. After accounting for differences in the built environment, cities with better public transit (proxied by buses detected) tend to have fewer traffic deaths—specifically, a 1% increase in bus presence is linked to a 0.35% decrease in fatalities per 100,000 people. This study demonstrates the power of using widely available street view imagery to uncover global disparities in urban design and their connection to road safety.</div></div>","PeriodicalId":100292,"journal":{"name":"Communications in Transportation Research","volume":"5 ","pages":"Article 100193"},"PeriodicalIF":12.5,"publicationDate":"2025-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144535777","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

SAFER-predictor: Sparse adversarial training framework for robust traffic prediction under missing and noisy data SAFER-predictor：稀疏对抗训练框架，用于缺失和噪声数据下的鲁棒交通预测

IF 12.5 Q1 TRANSPORTATION

Communications in Transportation Research

Pub Date : 2025-06-26 DOI: 10.1016/j.commtr.2025.100192

Yutian Liu , Chengfeng Jia , Soora Rasouli , Jian Gong , Tao Feng , Melvin Wong , Tianjin Huang

Accurate traffic flow forecasting is essential for developing intelligent transportation systems (ITSs) to reduce congestion, optimize road management, and improve safety. While data-driven traffic prediction approaches have shown high accuracy, they rely heavily on precise measurements, making them vulnerable to perturbed environmental factors, like sensor malfunctions, data storage issues, and adverse weather conditions. To overcome the limitation, we propose SAFER-Predictor, a novel sparse adversarial training (Sparse AT) framework for enhancing the reliability of deep learning based spatiotemporal traffic prediction models. Sparse AT extends traditional adversarial training (AT) through a two-phase process: pre-training and fine-tuning. In the pre-training phase, the model is optimized to capture normal traffic patterns, enhancing predictive performance by understanding standard dynamics without external disruptions. In the fine-tuning phase, the focus shifts to strengthening robustness against corrupted inputs by employing an iterative min-max strategy during AT, optimizing performance for worst-case scenarios. Furthermore, we derive theoretical formulations that establish an upper bound on the model's prediction error following Sparse AT under certain noise levels. Experimental results indicate that incorporating Sparse AT into the representative traffic flow prediction models improves stability and ensures high accuracy under various perturbation scenarios.

准确的交通流量预测对于发展智能交通系统（ITSs）以减少拥堵、优化道路管理和提高安全性至关重要。虽然数据驱动的交通预测方法显示出很高的准确性，但它们严重依赖于精确的测量，这使得它们容易受到环境因素的干扰，如传感器故障、数据存储问题和恶劣天气条件。为了克服这一限制，我们提出了一种新的稀疏对抗训练（sparse AT）框架SAFER-Predictor，用于提高基于深度学习的时空交通预测模型的可靠性。稀疏AT通过两个阶段的过程扩展了传统的对抗训练（AT）：预训练和微调。在预训练阶段，优化模型以捕获正常的交通模式，通过理解标准动态而不受外部干扰来提高预测性能。在微调阶段，重点转移到通过在AT期间采用迭代最小-最大策略来增强对损坏输入的鲁棒性，优化最坏情况下的性能。此外，我们推导了理论公式，在一定的噪声水平下建立了稀疏AT模型预测误差的上界。实验结果表明，将稀疏AT引入代表性的交通流预测模型中，提高了模型的稳定性，保证了模型在各种扰动情况下的高精度。

{"title":"SAFER-predictor: Sparse adversarial training framework for robust traffic prediction under missing and noisy data","authors":"Yutian Liu , Chengfeng Jia , Soora Rasouli , Jian Gong , Tao Feng , Melvin Wong , Tianjin Huang","doi":"10.1016/j.commtr.2025.100192","DOIUrl":"10.1016/j.commtr.2025.100192","url":null,"abstract":"<div><div>Accurate traffic flow forecasting is essential for developing intelligent transportation systems (ITSs) to reduce congestion, optimize road management, and improve safety. While data-driven traffic prediction approaches have shown high accuracy, they rely heavily on precise measurements, making them vulnerable to perturbed environmental factors, like sensor malfunctions, data storage issues, and adverse weather conditions. To overcome the limitation, we propose SAFER-Predictor, a novel sparse adversarial training (Sparse AT) framework for enhancing the reliability of deep learning based spatiotemporal traffic prediction models. Sparse AT extends traditional adversarial training (AT) through a two-phase process: pre-training and fine-tuning. In the pre-training phase, the model is optimized to capture normal traffic patterns, enhancing predictive performance by understanding standard dynamics without external disruptions. In the fine-tuning phase, the focus shifts to strengthening robustness against corrupted inputs by employing an iterative min-max strategy during AT, optimizing performance for worst-case scenarios. Furthermore, we derive theoretical formulations that establish an upper bound on the model's prediction error following Sparse AT under certain noise levels. Experimental results indicate that incorporating Sparse AT into the representative traffic flow prediction models improves stability and ensures high accuracy under various perturbation scenarios.</div></div>","PeriodicalId":100292,"journal":{"name":"Communications in Transportation Research","volume":"5 ","pages":"Article 100192"},"PeriodicalIF":12.5,"publicationDate":"2025-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144481778","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