IEEE Open Journal of Vehicular Technology最新文献

{"title":"Contextual Graph Modeling for Open-Set Object Detection in Multi-View Autonomous Driving","authors":"Muhammad Asad;Ihsan Ullah;Ganesh Sistu;Michael G. Madden","doi":"10.1109/OJVT.2026.3687800","DOIUrl":"https://doi.org/10.1109/OJVT.2026.3687800","url":null,"abstract":"Autonomous vehicles require robust perception systems capable of accurately identifying both known and unknown objects in complex driving environments. Traditional closed-set detectors risk safety by failing to recognize novel or out-of-distribution (OOD) objects. This work introduces a contextual graph-based framework for open-set object detection in multi-view autonomous driving. We also introduce a novel dataset, NuScenes-Novel, which is enhanced with additional and unusual objects from the MS-COCO dataset to form a more realistic benchmark. Detected objects are represented as nodes in graphs, where spatial and contextual relationships are established through distance metrics. Each object’s appearance and geometric features are embedded using a DINOv2 backbone, and a Variational Graph Auto-Encoder (VGAE) is trained to learn the contextual structure and relationships within each scene. During inference, VGAE reconstruction errors are fused with a context-aware Mahalanobis distance to identify OOD objects effectively. On NuScenes-Novel (aggregated over six cameras), the proposed fusion achieves an AUROC of 0.98 and an FPR@95 of 0.07, which shows strong separation between in-distribution and novel objects across multi-view scenes. Integrating appearance and relational context improves robustness under occlusion and varying objects. The results demonstrate that modeling object relationships and leveraging graph-based reasoning significantly improve the reliability and robustness of perception in autonomous vehicles. These findings show the importance of considering contextual information and a hybrid scoring approach for open-set detection.","PeriodicalId":34270,"journal":{"name":"IEEE Open Journal of Vehicular Technology","volume":"7 ","pages":"1323-1336"},"PeriodicalIF":4.8,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11494509","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148003356","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":"RIS-Aided RSMA Improves the Latency vs. Energy Trade-Off in the Finite Block Length MIMO Downlink","authors":"Mohammad Soleymani;Bruno Clerckx;Robert Schober;Lajos Hanzo","doi":"10.1109/OJVT.2026.3675569","DOIUrl":"https://doi.org/10.1109/OJVT.2026.3675569","url":null,"abstract":"We simultaneously minimize the latency and improve energy efficiency (EE) of the multi-user multiple-input multiple-output (MU-MIMO) rate splitting multiple access (RSMA) downlink, aided by a reconfigurable intelligent surface (RIS). Our results show that RSMA improves the EE and may reduce the delay to 13% of that of spatial division multiple access (SDMA). Moreover, RIS and RSMA support each other synergistically, while an RIS operating without RSMA provides limited benefits in terms of latency and cannot effectively mitigate interference. Furthermore, increasing the RIS size amplifies the gains of RSMA more significantly than those of SDMA, without altering the fundamental EE-latency trade-offs. Results also show that latency increases with more stringent reliability requirements, and RSMA yields more significant gains under such conditions, making it eminently suitable for energy-efficient ultra-reliable low-latency communication (URLLC) scenarios.","PeriodicalId":34270,"journal":{"name":"IEEE Open Journal of Vehicular Technology","volume":"7 ","pages":"1036-1048"},"PeriodicalIF":4.8,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11442953","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147606186","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":"Physics-Guided Tire-Force Estimation for Intelligent Vehicles With Differentiable Magic Formula and Friction-Ellipse Projection","authors":"Ping Li;Feng Zhang;Yong Zhang;Liang Su;Jiming Lin","doi":"10.1109/OJVT.2026.3722881","DOIUrl":"https://doi.org/10.1109/OJVT.2026.3722881","url":null,"abstract":"Accurate tire-force estimation is essential for chassis stability control and motion planning in intelligent vehicles, especially under transient load transfer, combined-slip operation, and variations in tire-road friction. This paper presents a physics-guided tire-force estimator based on a Physics-Informed Neural Network with a Differentiable Friction Ellipse Projection (PINN-DFEP) layer, termed the PINN-DFEP estimator. The estimator predicts longitudinal, lateral, and vertical tire forces from vehicle and suspension-related signals. In the proposed architecture, the Magic Formula is used as a differentiable nominal shape-prior term to regularize the learned force-slip relationship, while the DFEP layer projects the raw longitudinal-lateral force estimates onto the admissible friction domain at the network output. This design embeds tire-mechanics priors into the learning process and enforces friction feasibility with respect to the supplied road-friction coefficient and boundary-defining vertical load. The estimator is trained using CarSim-MATLAB/Simulink co-simulation data from sinusoidal steering maneuvers and evaluated under unseen fishhook and J-turn maneuvers with prescribed Gaussian noise added to the corresponding input or measurement channels. Compared with the learning-based baselines and a physics-based UKF observer under the same test maneuvers and force references, the PINN-DFEP estimator achieves lower tire-force estimation errors while enforcing longitudinal-lateral feasibility with respect to the supplied friction boundary in the tested co-simulation scenarios.","PeriodicalId":34270,"journal":{"name":"IEEE Open Journal of Vehicular Technology","volume":"7 ","pages":"2518-2530"},"PeriodicalIF":6.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11651589","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148925226","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":"GAPPO-AVNs: GA-Bounded PPO for DAG-Aware Resource Optimization in MEC-Enabled Vehicular Networks","authors":"Maha Alruwaili;Karim Djemame;Li Zhang","doi":"10.1109/OJVT.2026.3716033","DOIUrl":"https://doi.org/10.1109/OJVT.2026.3716033","url":null,"abstract":"Autonomous vehicular networks (AVNs) supported by multi-access edge computing (MEC) require reliable execution of latency-critical workloads under dynamic traffic, wireless, and computing conditions. Existing resource-allocation methods often lack the ability to jointly provide system-level coordination, real-time adaptability, and dependency-aware execution for directed acyclic graph (DAG)-based autonomous-driving workloads. This paper proposes GAPPO-AVNs, a hierarchical GA-bounded Proximal Policy Optimization (PPO) framework for delay-constrained energy optimisation in MEC-enabled AVNs. The key novelty lies in a feasibility-guided GA–PPO coordination mechanism, where a genetic algorithm (GA) periodically generates global resource bounds for CPU frequency, transmission power, bandwidth allocation, and task offloading, while PPO performs online refinement within the GA-bounded feasible region through a Monitor–Analyse–Plan–Execute (MAPE) control loop. A DAG-based task model is also incorporated to capture subtask precedence constraints and support dependency-aware parallel execution. Simulation results show that GAPPO-AVNs improves the energy–delay trade-off, task-completion reliability, and policy stability compared with GA-only, PPO-only, PSO-based, DE-based, and heuristic baselines. These results confirm the effectiveness of GA-bounded action control for scalable resource allocation in MEC-enabled AVNs.","PeriodicalId":34270,"journal":{"name":"IEEE Open Journal of Vehicular Technology","volume":"7 ","pages":"2606-2622"},"PeriodicalIF":6.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11617324","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148986089","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":"Frequency-Aware Decision Transformer for Offline Policy Learning in Adaptive Twin Control","authors":"Cao Han;Ran Chen;Shuaizhen Yao;Chunyan Xu;Xiaowei Mei","doi":"10.1109/OJVT.2026.3696721","DOIUrl":"https://doi.org/10.1109/OJVT.2026.3696721","url":null,"abstract":"Adaptive twin control provides a promising paradigm for learning and evaluating control policies using historical trajectories and simulated system responses, especially when direct interaction with physical systems is costly or unsafe. In this paper, we focus on offline policy learning as an algorithmic component in adaptive twin control pipelines, rather than constructing a full-stack vehicular digital twin system. To improve trajectory representation learning under noisy, heterogeneous, and non-stationary offline data, we propose a Frequency Decision-Making Transformer (FDMT). Built upon the MetaFormer framework, FDMT introduces two complementary modules: Time-Aware Diverse Trend (TDT), which captures local temporal trends and mitigates noisy or suboptimal behaviors, and Frequency Periodic Tracking (FPT), which exploits frequency-domain attention to encode long-term periodic and structural dependencies in trajectory sequences. By integrating time-domain trend modeling and frequency-domain dependency learning, FDMT enhances the robustness of return-conditioned policy learning. Experiments on standard D4RL MuJoCo continuous control benchmarks demonstrate that FDMT consistently improves normalized returns over representative offline reinforcement learning baselines. These results suggest that frequency-aware offline policy learning can serve as a useful decision-making module for adaptive twin control scenarios, while high fidelity vehicular digital twin deployment is left for future work.","PeriodicalId":34270,"journal":{"name":"IEEE Open Journal of Vehicular Technology","volume":"7 ","pages":"1656-1666"},"PeriodicalIF":6.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11535007","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148443238","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":"LAMP: Learning Automata-Based Multi-Objective Optimization for RSU Placement in IoV","authors":"Jasmini Kumari;Harinath Ankarboina;Sambit Kumar Mishra","doi":"10.1109/OJVT.2026.3692780","DOIUrl":"10.1109/OJVT.2026.3692780","url":null,"abstract":"In the Internet of Vehicles (IoV), roadside units (RSUs) deployed at road junctions enable vehicle-to-infrastructure (V2I) communication and support various intelligent transportation systems (ITS). However, since RSU deployment is constrained by installation cost, determining an optimal placement strategy becomes a challenging problem. Improper placement may lead to insufficient network coverage, increased relay-induced latency, and higher relay energy consumption due to multi-hop vehicle relaying. To address these issues, this paper proposes a learning automata-based multi-objective RSU placement framework, termed LAMP, which jointly optimizes spatial coverage, traffic-weighted coverage, relay-induced latency, and relay energy consumption. In the proposed approach, each candidate junction is modeled as an independent learning automaton that adaptively learns the probability of hosting an RSU using the Linear Reward–Inaction update mechanism. To improve convergence and solution quality, LAMP integrates an adaptive decaying learning rate, a multi-run restart strategy, and a two-phase local search refinement. The framework is evaluated on a SUMO-simulated urban road network of Mangalagiri, AP, India, consisting of 854 junctions, where vehicle arrivals follow a Poisson distribution to represent stochastic traffic flow. Experimental results show that LAMP achieves the highest combined fitness for RSU budgets of 15, 20, and 25 units, outperforming the nearest competing methods by 0.4%–4.2% depending on RSU budget, while requiring significantly lower computation time than population-based optimization methods.","PeriodicalId":34270,"journal":{"name":"IEEE Open Journal of Vehicular Technology","volume":"7 ","pages":"1641-1655"},"PeriodicalIF":6.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11517224","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148433160","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":"Adaptive Split Federated Learning With Behavioral Digital Twins and Multi-Head Temporal Networks for Internet of Vehicles Security","authors":"Farhan Ullah;Khalid Mahmood;Umar Raza;Jawad Ahmad;Jenhui Chen;Rubal Jeet;Yue Zhao","doi":"10.1109/OJVT.2026.3694907","DOIUrl":"10.1109/OJVT.2026.3694907","url":null,"abstract":"The rapid growth of the Internet of Vehicles (IoV) ecosystem has produced complicated security concerns that require a modern Intrusion Detection System (IDS) that can respond to emerging threats and maintain data protection. The dynamic behavior of clients and the heterogeneity of the system can affect the convergence and performance of models in federated learning for IoV security. Traditional federated learning may overlook data quality, computational resources, and client confidence. This issue is resolved by behavioral digital twin technology, which generates virtual replicas of real-world IoV clients that continuously monitor, analyze, and forecast client behavior. This paper proposes a novel approach, DT-ASFL, an adaptive Split Federated Learning (SFL) system that combines behavioral digital twins with multi-head temporal networks to improve IoV security. The lightweight multi-head Temporal Convolutional Network (TCN) extracts temporal features, and SFL stores sensitive vehicle data on edge devices. These attributes are analyzed by a more advanced classifier on a distant server, so raw data never leaves the device. For each client, behavioral digital twins dynamically modify aggregation weights based on trust and health measures, evaluate data quality, and monitor performance. This adaptive method provides protection against malicious users and system heterogeneity. TCN architecture captures diverse temporal patterns for optimal feature fusion using parallel processing heads with different dilation rates and an attention mechanism. Evaluation on several security datasets demonstrates considerable increases in attack detection accuracy and robustness. The approach ensures privacy and system flexibility while outperforming conventional federated learning techniques.","PeriodicalId":34270,"journal":{"name":"IEEE Open Journal of Vehicular Technology","volume":"7 ","pages":"1614-1626"},"PeriodicalIF":6.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11524213","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148433295","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":"Route-Based Multi-Vehicle Simulation Framework for Energy-Efficiency Evaluation of Class-8 Heavy-Duty Trucks","authors":"Daliang Shen;Jihun Han;Yaozhong Zhang;Namdoo Kim;Dominik Karbowski","doi":"10.1109/OJVT.2026.3710380","DOIUrl":"https://doi.org/10.1109/OJVT.2026.3710380","url":null,"abstract":"This paper presents a simulation-based framework for evaluating the energy efficiency of connected and automated Class-8 heavy-duty trucks (HDTs) under realistic route and traffic conditions. The framework is designed to accommodate different HDT powertrain types through a vehicle-model library; the present demonstration uses an internal-combustion-engine Class-8 tractor-trailer. The emphasis of this paper is on integrating validated HDT models with realistic route, traffic, and control context. The framework combines real-world HDT data analysis, high-fidelity powertrain modeling, HDT-specific driver-model tuning, route extraction, traffic scenario generation, and closed-loop multi-vehicle simulation. A validated HDT powertrain model developed in Autonomie is converted into a RoadRunner-compatible vehicle agent, enabling interaction with traffic signals, surrounding vehicles, and route-level information. Real-world and regional-model-derived routes are used to compare a connected and automated HDT eco-driving controller against a tuned human-driver baseline. Across the studied scenarios, the connected and automated HDT achieved energy savings of 3%–30% and travel-time savings of 4%–21% relative to the baseline. The results demonstrate that route characteristics, traffic conditions, driver-model calibration, and high-fidelity powertrain dynamics strongly affect HDT energy-consumption estimates.","PeriodicalId":34270,"journal":{"name":"IEEE Open Journal of Vehicular Technology","volume":"7 ","pages":"2201-2213"},"PeriodicalIF":6.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11602001","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148626027","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":"Hybrid-Index Modulation in Fading Channels With Integrated Sensing and Communication","authors":"Jaswinder Lota;Andreas Demosthenous","doi":"10.1109/OJVT.2026.3705145","DOIUrl":"https://doi.org/10.1109/OJVT.2026.3705145","url":null,"abstract":"Ultra-high-speed data transfer and energy efficiency are rapidly emerging as challenging and simultaneous requirements for next-generation vehicle-to-vehicle communication networks. This paper presents the first semi-analytical bit-error-rate analysis of hybrid index modulation with hexagonal-quadrature amplitude modulation over Rician and lognormal fading channels, combined with an integrated sensing and communication framework for simultaneous vehicle detection. The moment generating function of the instantaneous signal-to-noise ratio is employed to derive the bit-error-rate; an algorithm is developed to compute the bit-error-rate for any channel with a known probability density function of the instantaneous signal-to-noise ratio. At equal total transmitted power and bit-error-rate of 10⁻⁴, a 2-4 bits per slot gain is achieved over legacy systems with a lower 4–5 dB signal-to-noise ratio. Analytical Cramér-Rao lower bounds and Monte Carlo root-mean square errors are derived for a joint two-dimensional (range-azimuth) position and velocity estimation; a position root-mean square error below 10 cm is demonstrated at 142 GHz for a vehicle at 50 m. The joint bit-error-rate and root-mean-square error as functions of the sensing-communication power allocation are quantified, and the Doppler impact on symbol orthogonality is characterised.","PeriodicalId":34270,"journal":{"name":"IEEE Open Journal of Vehicular Technology","volume":"7 ","pages":"2091-2108"},"PeriodicalIF":6.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11570035","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148626461","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":"Providing Differential Privacy for Federated Learning Over Wireless: A Cross-Layer Framework","authors":"Jiayu Mao;Tongxin Yin;Aylin Yener;Mingyan Liu","doi":"10.1109/OJVT.2026.3672638","DOIUrl":"https://doi.org/10.1109/OJVT.2026.3672638","url":null,"abstract":"Federated Learning (FL) is a distributed machine learning framework that inherently allows edge devices to maintain their local training data, thus providing some level of privacy. However, FL’s model updates still pose a risk of privacy leakage, which must be mitigated. Over-the-air FL (OTA-FL) is an adapted FL design for wireless edge networks that leverages the natural superposition property of the wireless medium. We propose a wireless physical layer (PHY) design for OTA-FL which improves differential privacy (DP) through a fully decentralized, dynamic power control strategy that utilizes both inherent Gaussian noise in the wireless channel and a cooperative jammer (CJ) for additional artificial noise generation when higher privacy levels are required. Although primarily implemented within the Upcycled-FL framework, where a resource-efficient method with first-order approximations is used at every even iteration to decrease the required information from clients, our power control strategy is applicable to any FL framework, including FedAvg and FedProx as shown in the paper. This adaptation showcases the flexibility and effectiveness of our design across different learning algorithms while maintaining a strong emphasis on privacy. Our design removes the need for client-side artificial noise injection for DP, improving client transmission efficiency by shifting the privacy-inducing perturbation to a cooperative jammer, which introduces additional system-level energy consumption. Privacy analysis is provided using the Moments Accountant method. We perform a convergence analysis for non-convex objectives to tackle heterogeneous data distributions, highlighting the inherent trade-offs between privacy and accuracy. Numerical results confirm that our approach with different FL algorithms outperforms the state-of-the-art under the same DP conditions on the non-i.i.d. dataset FEMNIST. The results also demonstrate the effectiveness of cooperative jammer in meeting stringent privacy demands. We further extend to imperfect CSI scenarios and show that adaptive power control combined with Upcycled-FL remains robust under channel estimation errors.","PeriodicalId":34270,"journal":{"name":"IEEE Open Journal of Vehicular Technology","volume":"7 ","pages":"1049-1065"},"PeriodicalIF":6.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11429089","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148482288","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}