IEEE Open Journal of Intelligent Transportation Systems最新文献

{"title":"Time-Optimal Speed Planning for Open-Pit Mine Autonomous Driving Based on Segmented S-Curve","authors":"Desheng Zhu;Yunchao Chen;Jun Hao;Qi Zhang;Zhipeng Huang;Kehu Yang","doi":"10.1109/OJITS.2026.3670936","DOIUrl":"https://doi.org/10.1109/OJITS.2026.3670936","url":null,"abstract":"To address complex open-pit mining scenarios featuring terrain constraints and dynamic demands, a segmented S-Curve-based time-optimal planning framework is proposed—one that effectively overcomes the limitations of traditional unconstrained planning approaches. The dynamic trajectory segmentation framework adaptively adjusts segmentation points through identifying geometric features, thereby decomposing the overall planning task into manageable subproblems. To tackle the real-time computational challenges posed by high-order asymmetric S-Curve trajectory planning, a Gröbner basis precomputation framework offsets algebraic complexity to the offline phase. To prevent planned trajectories from exceeding speed constraints, a dynamic path point insertion correction method is put forward by relaxing acceleration rates. Through comprehensive hardware-in-the-loop (HIL) simulations and scaled-down physical tests, the proposed method exhibits significant improvements in both efficiency and robustness. Experimental validation across five mining routes reveals an average 18.8% reduction in travel time while fully satisfying all constraints. The planner achieves real-time performance with a computation time of 15 ms or less, and maintains precise path tracking (0.22 m RMS error) under varying load conditions—verifying its practical applicability to mining operations.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"7 ","pages":"862-874"},"PeriodicalIF":5.3,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11421900","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147557411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"E-YOLO to OMOD: An Enhanced YOLO Framework for Small Moving Object Detection in Omnidirectional Videos","authors":"Yui Nakaoka;Shizuha Asaki;Chinthaka Premachandra","doi":"10.1109/OJITS.2026.3670217","DOIUrl":"https://doi.org/10.1109/OJITS.2026.3670217","url":null,"abstract":"Omnidirectional cameras provide a comprehensive 360-degree field of view and enable efficient data collection, making them valuable for various applications such as surveillance, autonomous navigation, and traffic monitoring. However, challenges such as image distortion, low resolution, and the difficulty of detecting small objects remain significant obstacles. YOLO, a widely adopted object detection model, offers high-speed and accurate recognition but struggles with small object detection in omnidirectional videos due to its grid-based approach, which can result in feature loss in low-resolution images. To address these issues, this study improves the detection accuracy of moving objects in omnidirectional images by leveraging transfer learning. A new dataset was carefully curated from omnidirectional camera footage, with particular emphasis on annotating key feature regions of each moving object to reduce the impact of resolution constraints. By incorporating deep learning-based enhancements, the proposed approach significantly improves the recognition and classification of small moving objects, particularly in outdoor environments such as intersections. Experimental results demonstrate that the proposed method outperforms conventional methods, offering greater robustness and reliability for real-world applications. The proposed method achieved an accuracy of 90%, whereas conventional models achieved a maximum accuracy of only 53%.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"7 ","pages":"838-846"},"PeriodicalIF":5.3,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11421090","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147557452","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"RM3DMOT: Roadside Monocular 3D Multi-Object Tracking With Motion-Appearance Optimization","authors":"Ping Wang;Yuyang Yao;Zuxing Li;Chao Wang;Can Tian;Xinhong Wang","doi":"10.1109/OJITS.2026.3706528","DOIUrl":"https://doi.org/10.1109/OJITS.2026.3706528","url":null,"abstract":"To enhance autonomous driving, vehicle-infrastructure cooperative systems are proposed to extend the vehicles’ perception range and rectify erroneous detection results via roadside perception. As a critical component of the system, multi-object tracking (MOT) provides sequential information of traffic participants, modeling their future motion tendencies. Although 3D MOT has been extensively studied for onboard sensors, roadside monocular 3D MOT remains underexplored. To address this, we propose RM3DMOT, a novel Roadside Monocular 3D Multi-Object Tracking algorithmthat systematically integrates motion and appearance optimization tailored to roadside challenges. In the motion prediction module, the unscented Kalman filter (UKF) is enhanced by incorporating nonlinear motion characteristics of multi-category participants. For the appearance feature extractor, a camera style adaptation mechanism—utilizing generative adversarial networks—is designed to ensure feature precision. Specifically, we customize UKF with category-specific motion models to handle nonlinear trajectories, and we introduce a GAN-based style adaptation to improve appearance feature robustness across diverse cameras. To enable effective motion-appearance association, we design a learning-based association modulethat leverages Multilayer Perceptron (MLP) to adaptively fuse motion and appearance similarities for robust trajectory-detection matching. While learning-based association is an established concept, its integration into a roadside monocular 3D MOT framework—with tailored inputs (HAIoU, appearance cosine, class compatibility) and class-balanced focal loss—constitutes a practical contribution to addressing the unique challenges of roadside scenes. A series of validation experiments are conducted on the V2X-Seq dataset and RCooper dataset. On the V2X-Seq dataset, RM3DMOT achieves state-of-the-art performance (<inline-formula> <tex-math>${72.06}%$ </tex-math></inline-formula> MOTA, <inline-formula> <tex-math>${83.95}%$ </tex-math></inline-formula> IDF1 for Vehicle, and <inline-formula> <tex-math>${59.57}%$ </tex-math></inline-formula> MOTA, <inline-formula> <tex-math>${78.97}%$ </tex-math></inline-formula> IDF1 for Non-vehicle). On the RCooper dataset, RM3DMOT demonstrates superior performance in terms of accuracy and robustness.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"7 ","pages":"1674-1686"},"PeriodicalIF":4.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11575675","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148485777","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"InterTraj: A Short-Term Trajectory Prediction Method Based on Multi-Vessel Interaction and Intention Awareness","authors":"Xingkong Ma;Wuyong Wang;Jiaoyang Li;Houjie Qiu;Weizhi Chen;Fan WU","doi":"10.1109/OJITS.2026.3706795","DOIUrl":"https://doi.org/10.1109/OJITS.2026.3706795","url":null,"abstract":"In complex maritime environments characterized by densely intertwined shipping routes and high vessel mobility, accurate short-term vessel trajectory prediction is critical for traffic safety and operational efficiency. However, existing models do not perform well in short-term vessel trajectory prediction due to the incompleteness and uncertainty of trajectory data. Most methods also suffer from a lack of navigation intentions from surrounding vessels, which ignores the interactions between multiple vessels. To this end, we propose InterTraj, a short-term vessel trajectory prediction model that integrates trajectory reconstruction, multi-vessel interaction modeling, and intention-aware prediction into a unified framework. InterTraj employs an InfoGAN-based trajectory reconstruction module to recover missing trajectory segments with high fidelity and variability. Then, a Social-LSTM-based interaction technique is proposed to model temporal dependencies and spatial interactions among vessels, generating preliminary future trajectories. Furthermore, an intention-aware classifier recognizes the intentions of neighbor vessels, refining predicted trajectories to align with inferred navigational goals. Under various parameter settings, the experimental results demonstrate that the trajectory reconstruction error of InterTraj is reduced by at least 33% compared to the optimal baseline model, and the average displacement error of InterTraj is reduced by at least 35.2% compared to state-of-the-art methods.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"7 ","pages":"1630-1646"},"PeriodicalIF":4.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11578094","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148486185","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Blockchain-Based Enhancement to V2X Security","authors":"Hassan Farran;David Khoury","doi":"10.1109/OJITS.2026.3706660","DOIUrl":"https://doi.org/10.1109/OJITS.2026.3706660","url":null,"abstract":"Current C-ITS security relies on the Public Key Infrastructure (PKI) and Certificate Authority (CA) model called VPKI. This involves a complex hierarchy of Certificate Enrolment Authorities (EAs) and Authorization Authorities (AAs), with the Intelligent Transportation System ITS station (Vehicle OBU or RSU) issuing long-term identifiers and short-term Authorization Tickets (ATs). However, this centralized PKI/CA system faces difficulties in cross-border and cross-organizational scenarios and incurs high processing costs for message verification. This research introduces the Distributed Public Key (DPK) Ethereum Blockchain platform for Vehicle-to-Everything (V2X) communication, designed to store C-ITS long-term certificates and AA public keys. It tackles scalability challenges and enhances cross-domain interoperability, reducing V2V communication latency and simplifying certificate management. Smart contracts in DPK facilitate real-time storage and access to vehicle certificates and AA public keys, thereby easing authentication and enhancing interoperability across various domains. Initial testing indicates a significant reduction in V2V message verification time and the successful establishment of secure, end-to-end communication sessions among C-ITS entities. These results highlight the approach’s potential as a scalable and effective security solution for V2X networks. The system was evaluated through simulation on Ethereum 2.0 and the Sepolia test network. Since this Ethereum implementation is a proof of concept, a private blockchain is recommended for real-world deployment applications.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"7 ","pages":"1789-1808"},"PeriodicalIF":4.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11576043","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148487180","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Semi-Supervised Framework for Road Condition Assessment: From Minor Surface Distress to Post-Disaster Failures","authors":"Eda Tsujimoto;Sunyong Eom;Tsutomu Suzuki","doi":"10.1109/OJITS.2026.3690169","DOIUrl":"https://doi.org/10.1109/OJITS.2026.3690169","url":null,"abstract":"Accurate road-condition segmentation is crucial for intelligent transportation systems (ITS), enabling autonomous driving and post-disaster recovery. Yet, most prior studies focus on binary or low-class segmentation, limiting applicability to complex scenes where subtle surface degradations and large-scale obstructions coexist. To address this gap, we present a unified framework for pixel-level segmentation across 11 road-condition classes using a hybrid dataset integrating the Post-Disaster Road Dataset Japan (PDRDD-J), the Social Media Image Dataset for Disaster Road Damage Object Detection (SoDR), and the Road Damage Dataset 2022 (RDD2022-J). The defined classes include background, road, manhole cover, alligator crack, linear crack, pothole, vehicle, natural blockage, structural blockage, sinkhole, and collapsed road. To learn this label space under limited supervision, we propose F-UNet, a Feature-Level Data Extractor (FLDE)-guided extension of U-Net integrated into a standard teacher–student self-training protocol and trained through progressive pairwise specialization. Unlike conventional U-Net variants that rely solely on global multi-class training, F-UNet derives FLDE guidance through pairwise road–damage patch training and injects it during decoding to better handle class imbalance and thin, low-contrast defects. Extensive experiments demonstrate consistent gains over strong baselines, supported by high-confidence class-wise analysis, confusion-matrix analysis, and error and uncertainty-based failure-mode characterization under disaster edge cases. On a held-out test set, the teacher model achieves 0.8652 mIoU, representing a 15.5% improvement over standard U-Net and outperforming U-Net<monospace>++</monospace>, DeepLabV3<monospace>+</monospace>, SegNet, and FCN under the same supervision setting. In the semi-supervised configuration, the student model further improves to 0.8727 mIoU. The F-UNet source code is available at <uri>https://github.com/EdaTsujimoto/F-UNET</uri>","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"7 ","pages":"1244-1263"},"PeriodicalIF":5.3,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11505821","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147937493","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AIO-NAV: INS-Centric All-In-One Navigation Framework Integrating Mono Camera, GNSS, and HD Map for Large-Scale Autonomous Driving","authors":"Kai-Wei Chiang;Chi-Hsin Huang;Mengchi Ai;Ting-Chun Wu;Pei-Ru Lu;Chin-Chia Hsu;Meng-Lun Tsai;Mohamed Elhabiby;Naser El-Sheimy","doi":"10.1109/OJITS.2026.3688484","DOIUrl":"https://doi.org/10.1109/OJITS.2026.3688484","url":null,"abstract":"Current Visual-Inertial Odometry (VIO) systems degrade rapidly when visual input is compromised, and fail to fully exploit the observability of industrial-grade Inertial Measurement Units (IMUs). Unlike these visual-centric approaches, this paper presents AIO-NAV, an Inertial Navigation System (INS)-centric All-In-One Navigation framework that addresses these limitations. This framework treats INS as the backbone and integrates aiding sources including Global Navigation Satellite System (GNSS), monocular Visual Odometry (VO), and High-Definition (HD) map. All sensors are fused into a single navigator via a unified Extended Kalman Filter (EKF), which employs a modular design. Key update mechanisms are proposed to enable each aiding source to be independently activated or deactivated while maintaining stable operation. Moreover, each module can be easily maintained or replaced with improved solutions. A proposed invariant VO aiding module substitutes the traditional odometer by providing full 3D relative motion measurements instead of distance-only constraints. HD map-based point cloud registration provides advanced aiding, replacing loop-closure mechanisms for absolute pose corrections when the map is available. Comprehensive experiments in GNSS-challenging environments demonstrate that AIO-NAV significantly outperforms benchmarks, including both system-level (Xsens Vision Navigator) and algorithm-level (VINS-Fusion) comparisons. The system achieves a 3D root mean square error (RMSE) of 1.510 m compared to 5.151 m and 4.965 m, respectively. With HD map aiding, AIO-NAV-HDMAP achieves sub-meter level accuracy (0.795 m maximum error), making it suitable for high-level autonomous driving applications.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"7 ","pages":"1209-1222"},"PeriodicalIF":5.3,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11498319","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147828887","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Estimation of Pedestrian Risk Level Using Context Features for Remote Monitoring Alert System of Autonomous Micro-Mobility Vehicles","authors":"Kenta Azuma;Ryoichi Shinkuma;Hayato Itsumi;Koichi Nihei;Takanori Iwai","doi":"10.1109/OJITS.2026.3705874","DOIUrl":"https://doi.org/10.1109/OJITS.2026.3705874","url":null,"abstract":"Smart cities utilize IoT and ICT technologies to optimize urban infrastructure, addressing challenges such as rapid urbanization, traffic congestion, and environmental impact. Micro-mobility services, including autonomous delivery robots and electric scooters, have recently emerged as promising solutions to enhance urban mobility and logistics efficiency. However, autonomous micro-mobility vehicles operating on sidewalks require remote monitoring by human operators for immediate intervention during unexpected events. Existing remote monitoring systems, primarily developed for road-based autonomous vehicles, often generate excessive alerts in crowded sidewalk environments, significantly increasing operator workload. This paper proposes a novel remote monitoring alert system that estimates pedestrian risk levels using context features such as pedestrian size, position, and posture from the perspective of micro-mobility vehicles. A numerical risk level is computed, predicting the likelihood of pedestrians obstructing vehicle movement, and alerts are generated accordingly. The proposed approach was evaluated using the JRDB open dataset, a simulator environment, and a real-world demonstration with an autonomous micro-mobility vehicle. The results show that the estimation performance depends on the future horizon and the combination of context features. Our findings highlight the potential of context-based risk assessment systems to support efficient and scalable remote monitoring of autonomous micro-mobility vehicles.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"7 ","pages":"1563-1572"},"PeriodicalIF":4.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11574031","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148443318","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CACO-VEC: A Charging-Aware Computation Offloading Framework for Parked Electric Vehicles in Vehicular Edge Computing","authors":"Sirawit Makoo;Prachaya Udomparichatr;Peerapon Vateekul;Kultida Rojviboonchai","doi":"10.1109/OJITS.2026.3740444","DOIUrl":"https://doi.org/10.1109/OJITS.2026.3740444","url":null,"abstract":"Vehicular Edge Computing (VEC) has emerged as a promising paradigm for leveraging the computational capabilities of electric vehicles (EVs) to support computation-intensive applications. However, EVs necessitate significant energy consumption to perform task offloading and execution while simultaneously preserving battery levels for mobility. Consequently, the joint optimization of energy and task management remains a formidable challenge in VEC environments. To address these challenges, this paper proposes a collaborative parked-EV charging and task offloading framework incorporating two novel components: (i) a quantile regression-based EV profile prediction module to characterize charging demand of each EVs, and (ii) a charging-aware task offloading scheduler formulated as a Mixed-Integer Linear Programming problem that enables task partitioning support to efficiently allocate workload execution across multiple vehicles. Extensive simulations demonstrate that the proposed framework schedules and offloads 2.53-5.27<inline-formula> <tex-math>$times$ </tex-math></inline-formula> more tasks than baseline methods and achieves at least 8.77<inline-formula> <tex-math>$times$ </tex-math></inline-formula> higher system throughput regarding successfully computed workload while consistently maintaining a high charging satisfaction rate exceeding 81% across all evaluated scenarios, even under increased workload. These results confirm the effectiveness of jointly integrating EV behavior prediction with charging-aware task offloading in VEC systems.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"7 ","pages":"2456-2474"},"PeriodicalIF":4.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11720418","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149188495","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"LiDAR-to-Graph Framework for Traffic Scene Understanding at Urban Intersections","authors":"Elham Binshaflout;Chaima Zaghouani;Charalampos Antoniadis;Hakim Ghazzai;Ahmad Alsharoa;Gianluca Setti","doi":"10.1109/OJITS.2026.3682464","DOIUrl":"https://doi.org/10.1109/OJITS.2026.3682464","url":null,"abstract":"Effective traffic monitoring is a cornerstone of intelligent transportation systems, enabling safer and more efficient urban mobility. Among emerging sensing modalities, Light Detection and Ranging (LiDAR) stands out as a robust technology that delivers high-resolution 3D spatial information, complementing traditional traffic monitoring systems. However, despite its potential, publicly available LiDAR datasets dedicated to traffic monitoring remain scarce, with most existing datasets primarily designed for autonomous driving from the ego-vehicle perspective. In this work, we develop a comprehensive LiDAR-driven pipeline for urban traffic assessment, with a specific focus on road intersections, one of the most complex and safety-critical areas of urban networks involving vehicles and pedestrians. The proposed simulation-based framework, combining SUMO for traffic scenario generation and Blender for high-fidelity 3D reconstruction with LiDAR emulation, integrates perception and relational reasoning through two main components: 1) a LiDAR-based perception module for 3D object detection and segmentation and 2) a graph-construction module that encodes spatial, geometric, and semantic relations between road users and infrastructure. The resulting datasets offer rich annotations, making them suitable for advanced AI-driven analysis of road user interactions and behaviors. We validate our approach using state-of-the-art models for LiDAR perception and graph neural networks, demonstrating high detection and relation-learning accuracy under both clean and distorted point cloud conditions. The results confirm the effectiveness of the proposed pipeline and highlight LiDAR’s ability to provide a holistic, reliable understanding of complex traffic scenes, paving the way for future smart mobility and intelligent traffic management applications.","PeriodicalId":100631,"journal":{"name":"IEEE Open Journal of Intelligent Transportation Systems","volume":"7 ","pages":"1073-1090"},"PeriodicalIF":5.3,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11478406","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147736956","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}