IEEE Transactions on Intelligent Transportation Systems最新文献

{"title":"Game-Guided Multi-Agent Reinforcement Learning Approach for Vehicle Decision-Making in Narrow-Road Meeting With Hybrid Interaction","authors":"Ting Zhang;Mengying Ruan;Mengyin Fu;Yi Yang","doi":"10.1109/TITS.2026.3704668","DOIUrl":"https://doi.org/10.1109/TITS.2026.3704668","url":null,"abstract":"Narrow-road meeting is one of the most challenging interactive driving scenarios, yet it has not been adequately studied in previous research. Current multi-agent reinforcement learning (MARL)-based approaches face critical limitations when addressing head-on conflict scenarios. One major limitation is the assumption of single-modal interactions, such as cooperative behavior patterns, which fail to depict the complexities of negotiation dynamics. Additionally, the convergence of MARL models is not guaranteed, often leading to local optima or deadlocks in extreme conflict. To address these challenges, this work proposed a game-guided MARL framework that integrates the Stackelberg game theory with a Multi-agent Proximal Policy Optimization (MAPPO) model. By incorporating leader-follower roles into the actor network, the model can handle hybrid interaction modalities. The role of an agent is evaluated based on their priorities regarding right-of-way, reversing conditions, and urgency level. Based on this, the agents make decisions by role identification in a ‘think twice, act once’ way. For training, we created scenarios in which the ego vehicle interacts with random oncoming traffic flows. For each background social vehicle, Social Value Orientation (SVO) is introduced alongside the intelligent driver model (IDM) for controlling. Building on the trained MARL model, a self-play adversarial learning mechanism is further designed to enhance the strategies’ robustness to mixed traffic. Experimental results demonstrate that the proposed game-guided MAPPO model achieves the best performance in terms of reward and success rate. After adversarial training, the model gains greater robustness to varied social behaviors.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"12458-12471"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097029","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":"Chat2Route: A Retrieval-Augmented System for Personalized and Context-Aware Travel Planning","authors":"Chuyi Li;Yiyu Chen;Hengshuo Liu;Ziyi Guo;Xinyi Wang;Wei Zeng;Mengxuan Zhang;Lei Li","doi":"10.1109/TITS.2026.3705615","DOIUrl":"https://doi.org/10.1109/TITS.2026.3705615","url":null,"abstract":"Personalized travel planning involves tailoring itineraries to individual pReferences, constraints, and contexts to enhance user satisfaction and engagement in urban exploration. While traditional navigation systems optimize for route efficiency, they often neglect real-time personalization, leading to a gap between algorithmic itinerary generation and user-centric adaptability. To bridge this gap, we introduce <sc>Chat2Route</small>, a novel retrieval-augmented generation (RAG) system that dynamically generates personalized itineraries from natural language user inputs. Grounded in insights from a formative study on user needs, <sc>Chat2Route</small> integrates large language models (LLMs) for natural language query processing, knowledge graphs for context-aware recommendations, and road-network-aware route ranking for interactive itinerary refinement. Unlike static approaches, <sc>Chat2Route</small> integrates LLM-based query understanding, knowledge-graph-guided POI retrieval, preference-aware reranking, and shortest-path-based route ranking to support personalized and context-aware itinerary planning. Through a controlled user study and a supplemental city-scale offline evaluation over real-world Philadelphia POIs, we show that <sc>Chat2Route</small> improves interaction efficiency, perceived clarity, and route-quality metrics under matched experimental conditions.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"13214-13229"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097056","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":"GUIDE: Generalizable Trajectory Prediction Using Intention-Aware Diffusion With Efficiency","authors":"Yuanbo Yang;Sheng Jin;Claudio Roncoli;Panagiotis Angeloudis;Simon Hu","doi":"10.1109/TITS.2026.3697792","DOIUrl":"https://doi.org/10.1109/TITS.2026.3697792","url":null,"abstract":"The ability to predict the future trajectories of surrounding vehicles is crucial for the safe and efficient operation of autonomous vehicles. However, most existing data-driven trajectory prediction models lack a deep understanding of driving intentions, which complicates the prediction task and often leads to suboptimal results. In this paper, we introduce <monospace>GUIDE</monospace>, a framework that decomposes driving intention into driver personality traits and driving maneuvers. This approach incorporates these inferred intentions into a diffusion-based trajectory prediction model, which better models the probability distribution of future trajectories in a generative manner. Furthermore, to address the limited generalization ability of existing models in unseen scenarios, we focus on increasing the data scale and diversity during training. To this end, we have designed a unified data processing framework and evaluation benchmark that integrates multiple autonomous driving datasets. Upon this foundation, we build our diffusion-based, high-generalization trajectory prediction system. Experimental results demonstrate that by training on this unified dataset, our model learns more robust scene representations, leading to significant improvements in both prediction performance and generalization ability.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"12895-12910"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097092","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":"SINS/GNSS Integrated Navigation Based on Left Equivariant Error Model in Inertial Frame","authors":"Yarong Luo;Chi Guo;Wentao Lu;Yulong Huang","doi":"10.1109/TITS.2026.3701149","DOIUrl":"https://doi.org/10.1109/TITS.2026.3701149","url":null,"abstract":"Reliable and continuous vehicle positioning in urban canyons remains a critical challenge for intelligent transportation systems. To address this issue, this article presents a left equivariant error model for strapdown inertial navigation system and global navigation satellite system (SINS/GNSS) integrated navigation in the inertial frame. Unlike the imperfect invariant extended Kalman filter and its variant, the proposed method copes with the biases states in a more natural way. Also, different from the current equivariant filter based SINS/GNSS integrated navigation, the linearized measurement matrices of this method are independent of the system’s position. This method allows the initial alignment stage and integrated navigation stage to be combined into one stage, enabling faster and more accurate navigation results in challenging urban settings. Firstly, we introduce the two-frame group and associated group actions on the state space and input space of the biased INS from the perspective of left equivariant filter. Secondly, we derive the left equivariant error dynamics in inertial frame and the measurement models for both loosely coupled (LC) SINS/GNSS and tightly coupled (TC) SINS/GNSS, respectively. Thirdly, we conduct the simulation and field test to evaluate the practical performance of the proposed method. Experimental results demonstrate that the proposed method achieves superior transient response in urban canyon environments compared to conventional approaches, enabling reliable one-step alignment and continuous positioning for intelligent vehicles operating in complex traffic scenarios.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"12400-12416"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097152","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":"Dual-Mode SPL-SLAM: A Robust Semantic Point-Line SLAM System With Monocular Neural Depth and Sensor Fusion for Intelligent Transportation","authors":"Zhen Tian;Zhihao Lin;Dezong Zhao;Christos Anagnostopoulos;Peng Wang;Wenjing Zhao;Lin Wu;Qi Zhang","doi":"10.1109/TITS.2026.3707726","DOIUrl":"https://doi.org/10.1109/TITS.2026.3707726","url":null,"abstract":"Simultaneous Localization and Mapping (SLAM) is critical for Intelligent Transportation Systems (ITS), yet existing solutions struggle to balance accuracy, sensor versatility, and computational efficiency. Traditional monocular methods suffer from scale drift, while RGB-D systems with heavy semantic networks often exceed the computational budgets of onboard processors. To address these conflicting constraints, we propose Dual-Mode SPL-SLAM, a unified adaptive framework capable of adapting to different intelligent driving agents. The system intelligently switches between two operating paths: Mode I (Sensor-Semantic) leverages physical depth sensors and high-precision semantic segmentation for complex dynamic environments. To further improve accuracy, we use line-feature processing and a refined epipolar-constraint error for robust detection of known and unknown dynamic objects; Mode II (Neural-Clustering) employs a lightweight neural network (LiteMono) and K-means++ clustering within the object bounding box predicted by GPU-accelerated YOLOX to recover scale and filter dynamic objects using fast LK optical flow in monocular setups. Crucially, both modes converge into a shared Adaptive Point-Line Backend, which dynamically weights features based on environmental texture to ensure robustness. Extensive experiments on KITTI dataset demonstrate that our system achieves state-of-the-art accuracy and real-time performance across diverse sensor configurations.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"12687-12700"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097472","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":"HiFTLLM-Rail: A Hierarchical Fine-Tuned LLM for Spatio-Temporal Forecasting in Urban Railway Transit Systems","authors":"Qipeng Song;Tiantian Huang;Yue Li;Wei Shao;Hui Li","doi":"10.1109/TITS.2026.3715544","DOIUrl":"https://doi.org/10.1109/TITS.2026.3715544","url":null,"abstract":"Accurate forecasting of railway transit dynamics, such as passenger flow, is crucial for optimizing scheduling, mitigating congestion, and enhancing operational resilience. Recently, large language models (LLMs) have shown potential for spatio-temporal forecasting because their pre-trained transformer backbones can provide transferable contextual sequence modeling. However, LLM-enabled rail transit forecasting still faces three core challenges. First, railway demand contains multi-scale temporal patterns, including local bursts, periodic commuting rhythms, and long-horizon variations that must be represented without losing station-specific dynamics. Second, rail systems are defined by structured topology and heterogeneous station roles, whereas LLM backbones are inherently sequential and lack graph inductive bias. Third, continuous spatio-temporal observations must be aligned with the hidden representation space of pre-trained LLMs, and the backbone must be adapted efficiently under limited transportation data and compute. To address these challenges, we propose HiFTLLM-Rail, a hybrid framework that integrates topology-aware spatio-temporal encoding with a Hierarchical Fine-Tuned (HiFT) LLM. A dual-branch encoder couples global graph-based diffusion with local temporal aggregation through a gated fusion mechanism, producing topology-aware station embeddings that are aligned to the LLM hidden space. The LLM backbone is then adapted via HiFT, which selectively updates its upper transformer layers and normalization parameters for efficient domain adaptation. Extensive experiments on three large-scale real-world datasets demonstrate that HiFTLLM-Rail consistently outperforms state-of-the-art graph-based, time-series, and LLM-based forecasting baselines across multiple prediction horizons. Ablation and model-analysis results further verify the effectiveness of topology-aware embedding construction, adaptive fusion, the pre-trained backbone, and hierarchical adaptation. Our code is available at <uri>https://github.com/KarryHuang12138/HiFTLLM-Rail-code</uri>","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"13598-13611"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097645","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":"An Adaptive Multimodal End-to-End Autonomous Driving Framework Based on Behavior Discrepancy","authors":"Xiaokai Chen;Xiaoyu Wang;Weihao Chen;Jianping Gao","doi":"10.1109/TITS.2026.3706626","DOIUrl":"https://doi.org/10.1109/TITS.2026.3706626","url":null,"abstract":"Vision-language models (VLMs) have recently advanced end-to-end autonomous driving by providing rich semantic understanding. However, existing approaches rarely consider practical constraints such as computational resource limits and inference latency, which are especially important in large-scale vehicle fleets or dense traffic scenarios with frequent VLM usage. To tackle this, we introduce a multimodal driving framework, G-VLM, which learns from the behavioral discrepancy between a lightweight controller and a powerful perception-language reasoning component. G-VLM incorporates a learned gating mechanism that adaptively determines when to activate the more expensive semantic processor, guided by scene context and confidence estimates. This enables a favorable trade-off between control accuracy and computational efficiency. Comprehensive evaluations on the CARLA closed-loop benchmark demonstrate that G-VLM achieves 96.8% of the fully-invoked VLM baseline’s driving score at only 12.2% of its computational cost, while outperforming compute-matched alternatives by up to 9.8 driving score points. The results highlight a promising direction where behavior-discrepancy-driven gating facilitates scalable and resource-aware autonomous driving, paving the way for efficient cloud-edge collaboration and intelligent mobility in smart city environments.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"13113-13124"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097609","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":"HMFM: A Robust Hierarchical Multimodal Fusion Module to Enhance Map Learning With Additional Information","authors":"Wenjie Gao;Entao Chang;Jiawei Fu;Shitao Chen;Nanning Zheng","doi":"10.1109/TITS.2026.3698777","DOIUrl":"https://doi.org/10.1109/TITS.2026.3698777","url":null,"abstract":"High-definition (HD) maps, rich in geometric and semantic information, are essential for autonomous driving systems. Recent online map learning methods, predicting maps directly from onboard sensor data, offer promising alternatives to traditional manual approaches. However, they are limited in long-range detection and complex environments due to onboard sensors’ inherent constraints. While some works incorporate external priors, they often lack generality due to tight coupling with specific architectures and poor robustness to imperfect data. To address these gaps, we propose the Hierarchical Multimodal Fusion Module (HMFM), a robust, generalizable plug-and-play module that fuses onboard data with external information at two levels. At the feature level, the Masked Attention Module (MAM) selectively enhances features to boost fusion efficiency and minimize interference. At the Bird’s Eye View (BEV) level, the Alignment Module (AM) resolves spatial discrepancies; then, the Unified Attention Module (UAM) adaptively weighs external priors to effectively mitigate inaccuracies from imperfect data. We also introduce a distillation framework to distill benefits from additional information into the model without affecting runtime efficiency. Extensive experiments validate HMFM’s generality, leveraging satellite and SD maps for significant gains across baselines (e.g., +22.0 mIoU on HDMapNet, +12.3 mAP on MapTR). Furthermore, the distillation framework boosts the original HDMapNet by 2.8 mIoU without affecting runtime efficiency. The code and data processing pipeline for this work are publicly available at: <uri>https://github.com/xjtu-cs-gao/HMFM</uri>","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"12432-12445"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097688","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":"Dynamic Parking Pricing for Electric Vehicles in a Loss Queue","authors":"Alix Dupont;Yezekael Hayel;Tania Jiménez;Jean-Baptiste Bréal","doi":"10.1109/TITS.2026.3702272","DOIUrl":"https://doi.org/10.1109/TITS.2026.3702272","url":null,"abstract":"The increasing adoption of Electric Vehicles (EVs) has led to higher demand for public Charging Points (CPs), causing availability issues. This work analyses how parking pricing, structured such that users exceeding a predefined parking duration pay an additional fee, can encourage shorter stays and improve CP turnover. This study explores and compares different pricing strategies for EV charging stations, classified by their level of dynamism: 1) real-time occupancy–based; 2) time-of-day–based; 3) hybrid; and 4) static. In all cases, the objective is to maximize charging profit, excluding revenue from parking fees. The model accounts for EV users heterogeneity in their parking duration and energy needs. User decisions whether to use a CP or not account for bounded rationality and uncertainty in their parking duration through expected utility hypothesis. CP occupancy is modeled as a multiclass and time-inhomogeneous queueing system. The underlying Markov chain is partially controlled by the pricing policy. Results show that the proposed pricing mechanism significantly increases the charging profit. In contrast, the incorporation of real-time occupancy level on the dynamic prices only has a minimal impact.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"12446-12457"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097813","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":"Prediction of Individual Motion Sickness of Bus Passengers Using a 6DoF-SVC Model Driven by Handheld Tablet Motion During Visual Tasks","authors":"Van Trong Dang;Maxence Bourseul;Takahiro Wada","doi":"10.1109/TITS.2026.3710322","DOIUrl":"https://doi.org/10.1109/TITS.2026.3710322","url":null,"abstract":"Motion sickness is a common issue for passengers in public transportation, particularly when engaging in visual tasks using handheld devices. This study investigates the feasibility of capturing individual motion sickness progression in bus passengers using a six-degree-of-freedom subjective vertical conflict (6DoF-SVC) model driven solely by motion data from a handheld tablet. Fifteen participants performed a visual search task on a tablet while riding a public bus, during which motion data from both a head-mounted inertial measurement unit (IMU) and a tablet-integrated IMU were recorded. Individual motion sickness severity was assessed using the Motion Illness Symptoms Classification (MISC). Model parameters were individually identified under different data availability conditions, and estimation accuracy was evaluated using mean absolute error. The results showed that estimation accuracy using tablet motion was not significantly different from that using head motion, and that parameter estimation using data up to MISC <inline-formula> <tex-math>$ge 3$ </tex-math></inline-formula> achieved accuracy comparable to that obtained using the entire dataset. Canonical correlation analysis revealed high correlations between head and tablet motion. These findings demonstrate that high-accuracy, individualized estimation of motion sickness severity is achievable in a real-world bus environment using only handheld device motion data, highlighting the potential for practical, sensor-light personalized monitoring systems.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"27 10","pages":"13461-13472"},"PeriodicalIF":9.1,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11603848","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097798","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}