IEEE Transactions on Intelligent Transportation Systems最新文献

{"title":"Self-Organizing Railway Traffic Management","authors":"Federico Naldini;Fabio Oddi;Leo D’Amato;Grégory Marlière;Vito Trianni;Paola Pellegrini","doi":"10.1109/TITS.2026.3704682","DOIUrl":"https://doi.org/10.1109/TITS.2026.3704682","url":null,"abstract":"Improving traffic management in case of perturbation is one of the main challenges in today’s railway research. The great majority of the existing literature proposes approaches to make centralized decisions to minimize delay propagation. In this paper, we propose a new paradigm to the same aim: we design and implement a modular process to allow trains to self-organize. This process consists in having trains identifying their neighbors, formulating traffic management hypotheses, checking their compatibility and selecting the best ones through a consensus mechanism. Finally, these hypotheses are merged into a directly applicable traffic plan. In a thorough experimental analysis on a portion of the Italian network, we compare the results of self-organization with those of a state-of-the-art centralized approach. In particular, we make this comparison mimicking a realistic deployment thanks to a closed-loop framework including a microscopic railway simulator. The results indicate that self-organization achieves better results than the centralized algorithm, specifically thanks to the definition and exploitation of the instance decomposition allowed by the proposed approach.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"12356-12368"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097138","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Efficient Online Origin-Destination Prediction via a Knowledge-Driven Deep Learning Framework","authors":"Zeren Xing;Beiyu Song;Edward Chung;Ashish Bhaskar;Azusa Toriumi;Takashi Oguchi;Keshuang Tang","doi":"10.1109/TITS.2026.3707838","DOIUrl":"https://doi.org/10.1109/TITS.2026.3707838","url":null,"abstract":"Efficient and accurate Origin-destination (OD) Prediction is critical for enhancing the efficiency of demand-responsive urban mobility systems. Existing Deep Learning (DL)-based OD Prediction models face significant challenges including extensive training time, limited adaptability to new data, and lack of explainability, which limits their real-world applications. To address these issues, this study proposes a knowledge-driven hierarchical framework, integrating a Gravity Model-Inspired Network (GIN) with a DL model (GIN-DL). The framework predicts real-time OD matrices via a hierarchical process: First, outflow prediction via a DL model. Second, flow probability estimation via GIN, which aligns with the principle of the singly constrained gravity model. The OD matrix is obtained by multiplying the predicted outflows and estimated flow probabilities. The study also introduces a two-stage training strategy combining offline pre-training using Keyframes (KF) and online fine-tuning through incremental learning. Experiments conducted on two standard benchmark datasets demonstrate that the proposed GIN-DL framework achieves substantial improvements in training time (up to 90%) while maintaining competitive prediction accuracy. Meanwhile, the integration of domain knowledge enhances explainability. The findings underscore the potential of applying the proposed framework and training strategy in real-world applications. The code is available at: <uri>https://github.com/xingzr1005/GIN-DL</uri>","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"13241-13256"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Vectorized HD Map Learning for Autonomous Driving: A Survey of Representations, Architectures, and Future Directions","authors":"Iyad Abuhadrous;Fawzi Nashashibi;Benazouz Bradai","doi":"10.1109/TITS.2026.3705871","DOIUrl":"https://doi.org/10.1109/TITS.2026.3705871","url":null,"abstract":"High-definition (HD) maps are essential for perception, localization, and planning in autonomous driving, providing geometric and semantic descriptions of road environments. This survey reviews recent advances in learning-based HD map construction with a focus on vectorized and structured representations that support high-level autonomy tasks. We analyze the evolution from rasterized to vectorized and graph-based approaches, discussing their respective advantages, limitations, and implications for downstream planning and forecasting. The paper categorizes dominant architectural paradigms, including transformer-based set prediction, diffusion models, and graph neural networks, and compares their performance on major benchmarks such as nuScenes, Argoverse2, and OpenLane-V2. We further examine sensor fusion strategies, topological reasoning mechanisms, and evaluation metrics tailored to structured map learning. Finally, we identify key open challenges, including generalization, real-time scalability, and topological consistency, and outline future research directions toward deployable learning-based HD mapping systems.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"11929-11948"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097369","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cooperative Control in Bottleneck Scenarios for Dynamically Networked USVs Using Graph Attention and Weighted Mean-Field Approximation","authors":"Dongfang Ma;Jie Wang;Songjian Lv;Ziling Wang;Huarong Zheng","doi":"10.1109/TITS.2026.3700649","DOIUrl":"https://doi.org/10.1109/TITS.2026.3700649","url":null,"abstract":"With the rapid evolution of autonomous shipping hubs, high-frequency logistics operations increasingly require multiple independent uncrewed surface vehicles (USVs) to navigate through confined waterways. In such scenarios, multi-vessel interactions degenerate into highly-coupled conflict resolution problems. While Multi-Agent Reinforcement Learning (MARL) offers a potential solution for this problem, it is typically hindered by the dimensionality curse in large-scale coordination, poor generalization across stochastic traffic distributions, and training instability in congested maritime environments. To address these systemic bottlenecks, this paper proposes GAT-WMF-SAC, a cooperative control framework for bottleneck scenarios. By embedding Graph Attention Networks (GAT) and Weighted Mean-Field (WMF) into a soft actor-critic (SAC) architecture, it transforms myopic conflict resolution into system-level operational stability through advanced situational awareness. Our approach introduces three key contributions: 1) a WMF Q-function that adaptively estimates interaction weights to mitigate dimensionality explosion in large-scale port traffic conditions; 2) a GAT-driven feature extractor that uniformly processes dynamic maritime traffic topologies with fluctuating participant densities, enhanced by a dual-coordinate transformation and curiosity-driven rewards to encourage spontaneous deconfliction and prevent deadlocks; and 3) a centralized training paradigm with a decentralized execution framework that fuses macroscopic WMF spatial interaction features with topological traffic observations to ensure stable value estimation within complex maritime traffic patterns. In benchmark multi-vehicle coordination scenarios, GAT-WMF-SAC achieved a 100% success rate, improved trajectory smoothness by 18.2%, and reduced curvature variance by 40.7% over the baseline, with all other metrics being optimal or near-optimal, demonstrating its significant superiority in managing complex maritime traffic.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"13363-13378"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097534","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"UAV Fleet Intrusion Detection and Fault Diagnosis for ITS Operations","authors":"Ning Bai;Xiaoya Hu;Shouyue Wang","doi":"10.1109/TITS.2026.3703012","DOIUrl":"https://doi.org/10.1109/TITS.2026.3703012","url":null,"abstract":"UAV fleets are increasingly deployed as aerial sensing and communication nodes that support intelligent transportation systems operations, including traffic monitoring, infrastructure inspection, and incident response. However, cyber intrusions in air–ground communication and physical faults in onboard subsystems can jointly disrupt data timeliness and service continuity, undermining operational reliability. This paper presents an integrated deep learning framework that jointly performs intrusion detection and fault diagnosis from multi-source network-traffic and flight-telemetry features. A task-aware static feature partitioning strategy decouples network and physical domain representations to mitigate cross-task interference while retaining shared learning. Experiments on a combined multi-source dataset show that the proposed method achieves 96.11% accuracy for intrusion detection and 97.49% accuracy for fault diagnosis, outperforming representative machine-learning and deep-learning baselines. Lightweight profiling further indicates the suitability of the proposed framework for real-time onboard or edge deployment, with an average inference time of 0.23 milliseconds per sample and 0.38 megabytes of parameters.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"12628-12642"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097501","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"TAHALight: A Temporal-Aware, Hybrid-Action Reinforcement Learning Approach for Traffic Signal Plan Generation","authors":"Meng Zhang;Dianhai Wang;Xu Wang;Meng Sun;Huangjun You;Hongxin Yu;Zhengyi Cai","doi":"10.1109/TITS.2026.3710506","DOIUrl":"https://doi.org/10.1109/TITS.2026.3710506","url":null,"abstract":"Most existing deep reinforcement learning (DRL) methods for traffic signal control rely on a high-frequency, second-by-second phase switching paradigm. While this paradigm often demonstrates superior performance in simulations, it faces significant challenges in real-world implementation. Traffic management practitioners prefer a cycle-based control paradigm, which, however, presents two fundamental challenges: (1) the curse of dimensionality in the action space, and (2) the difficulty of representing traffic states effectively given the long control intervals. This paper introduces TAHALight, a Temporal-Aware, Hybrid-Action Reinforcement Learning approach, that overcomes these obstacles through dedicated designs for the action space and state representation. To tackle the dimensionality curse, we propose a hierarchical hybrid action space. In this framework, a high-level policy selects a discrete cycle length from a predefined set, while a low-level policy determines the corresponding splits. This decomposition significantly reduces the action space dimensionality, thereby enhancing learning efficiency. To address the state representation challenge, we devise a state observation mechanism based on multi-period sampling and employ a Transformer-based state encoder. This encoder leverages a multi-head attention mechanism to model the spatio-temporal dependencies of traffic flow, enabling proactive and context-aware decision-making. Comprehensive 24-hour experiments conducted on both synthetic and real-world scenarios demonstrate that TAHALight can dynamically adapt to traffic flow variations. It significantly outperforms baseline methods in the synergistic optimization of cycle length and splits, reducing average delay and queue length by over 10% compared to both fixed-time control and state-of-the-art DRL baselines.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"13082-13096"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097463","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Maximum Likelihood Approach to Cooperative Localization for Partially Connected Networks","authors":"Eric Chaumette;Jordi Vilà-Valls;Yoan Gregoire","doi":"10.1109/TITS.2026.3703063","DOIUrl":"https://doi.org/10.1109/TITS.2026.3703063","url":null,"abstract":"Reliable and precise positioning information is paramount in a plethora of applications involving distributed sensing systems or networks of interconnected agents. This contribution focuses on the centralized cooperative localization problem exploiting inter-agent ranging observations, where: i) only a subset of distances is available, i.e., the network is partially connected, which leads to an incomplete Euclidean distance matrix (EDM), and ii) a position estimate is available for a subset of agents. For fully connected networks, or complete EDMs, a typical solution is given by suboptimal multidimensional scaling and mapping (MDS-MAP), which can be extended to incomplete EDMs if combined with a matrix completion strategy. An alternative is the maximum likelihood estimator (MLE), but in its standard form (grid search) it cannot be used when the number of agents in the network grows, as its convergence is severely impacted by the initialization point. In this article we propose a maximum likelihood approach to EDM matrix completion, being asymptotically optimal in the mean square error sense, which in turn allows one to obtain a MDS-MAP initialization to the corresponding MLE, ensuring a feasible and asymptotically efficient method. Illustrative examples are provided to support the discussion and show that the proposed method outperforms MDS-MAP and semidefinite programming approaches.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"12653-12660"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097468","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Coordinative Multi-Scale Feature Augmentation Network for Multi-Task Driving Perception","authors":"Tao Deng;Haoran Liu;Wenbo Liu;Fei Yan","doi":"10.1109/TITS.2026.3713620","DOIUrl":"https://doi.org/10.1109/TITS.2026.3713620","url":null,"abstract":"With the rapid advancement of autonomous driving technology, panoramic driving perception systems, as a core component of autonomous driving systems (ADS), have gained significant attention. The constraints of onboard computational resources have intensified the demand for lightweight, efficient, and precise perception systems. This paper proposes a multi-task network model based on a coordinative multi-scale feature augmentation network (CMFANet), capable of simultaneously addressing traffic object detection, drivable area segmentation, and lane line detection, thereby enabling comprehensive perception of traffic scenarios. The proposed CMFANet consists of a shared backbone, three independent neck layers, and task heads. To achieve the lightweight and high-precision perception of the driving environment, we design a powerful backbone network for feature extraction and develop an efficient feature fusion strategy tailored to the specific requirements of each task. We construct two versions of the proposed model: nano and small. The nano version features a parameter count of only 3.8 M, significantly lower than existing models such as YOLOP. The model was evaluated on the challenging BDD100K dataset, achieving outstanding results: a mAP50 of 82.5% for traffic object detection, a mIoU of 91.5% for drivable area segmentation, and an IoU of 29.8% for lane line detection. Experiments confirm our model’s balance of efficiency, lightweight design, and accuracy for real-world driving scenarios.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"12185-12200"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097616","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Estimating City-Scale Pedestrian Movement With Sensor Networks for Count Data","authors":"Wentao Chen;Anders Johansson;Nikolai W. F. Bode","doi":"10.1109/TITS.2026.3703235","DOIUrl":"https://doi.org/10.1109/TITS.2026.3703235","url":null,"abstract":"Walking remains an integral part of modern transportation and therefore pedestrian behaviour influences the functioning, management, and further design of intelligent transportation systems. The recent widespread deployment of sensors that collect anonymous pedestrian counts in city locations presents an opportunity. Data on pedestrian dynamics are now recorded on a scale beyond the commonly studied individual level while preserving information on how pedestrians utilize transportation facilities, such as roads and pathways. However, it remains unclear how the pedestrian dynamics on a city scale can be recovered from such aggregated measurements. Here, we integrate information captured by sensor networks to estimate city-scale pedestrian movement represented by flows between locations with sensors. We propose an estimator based on the maximum entropy principle that uses only information collected by the sensing system. Notably, the estimation problem is inherently underdetermined, as the observations are insufficient relative to the complexity of city-scale pedestrian movement. We address this by examining the ensemble average of feasible systems under the observation, which gradually yields an exact representation of the system as it becomes determined, such as for high sensor densities. In a case study, we apply our estimator to three real-world datasets, assuming stronger interactions between counts recorded by nearby sensors. The estimator recovers stable, time-dependent, and spatially heterogeneous pedestrian flows between sensor locations. The proposed framework, leveraging emerging sensor data, opens new possibilities for accounting for pedestrian dynamics in intelligent transportation systems.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"12925-12934"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097574","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Guest Editorial Introduction to the Special Issue on AI-Empowered Intelligent Rail Transit in the Age of Digitalization","authors":"Soufiene Djahel;Zonghua Zhang;Celimuge Wu;Guohui Zhang","doi":"10.1109/TITS.2026.3733898","DOIUrl":"https://doi.org/10.1109/TITS.2026.3733898","url":null,"abstract":"","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"13488-13491"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11720470","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097742","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}