Aug 2026· IEEE transactions on intelligent transportation systems (Print)· Vol 27, pp. 9088-9101· 0 citations· 41 references
Abstract
Achieving efficient task offloading with low latency in cooperative vehicle-infrastructure systems represents a critical challenge influenced by three key factors: heterogeneous vehicle computational capabilities, dynamic traffic conditions, and variable communication reliability. To address the fundamental trade-off between real-time processing performance and cost-effective infrastructure investment, this study develops a multi-layer agent-based modeling (ABM) framework that systematically evaluates how roadside unit (RSU) computation capacity, traffic volume, and vehicular computation capability (VCC) heterogeneity collectively impact collaborative computing latency. The ABM framework comprises three interconnected layers: an infrastructure layer that models RSUs with sensing, communication, and computing functions; a traffic layer that simulates diverse traffic environments; and a collaborative computing layer that manages decentralized task generation and offloading. A modified water-filling algorithm is integrated to dynamically allocate computational tasks based on available resources and latency constraints. Simulation results under signalized intersection scenarios reveal nonlinear latency growth with increasing traffic volume due to queuing effects. When the RSU computing capacity is insufficient, most tasks are offloaded to nearby vehicles. Our results show that when the RSU computational capacity is comparable to the total VCC of all vehicles, the RSU processes approximately 40% of the tasks, primarily due to vehicle–RSU mobility effects and stochastic wireless transmission failures. Furthermore, increasing VCC heterogeneity—while keeping the average constant—leads to both higher average latency and greater variability. These findings provide quantitative insights for optimizing computing resource deployment in real-world cooperative systems.
The rapid advancements of next-generation vehicular networks require intelligent, low-latency, and efficient resource management to support heterogeneous services. In this work, we propose a Traffic-aware Dynamic Resource Allocation (TADRA) architecture for UAV-assisted vehicular O-RAN to address the challenges of dynamic traffic conditions, infrastructure failures, and stringent quality of service (QoS) requirements. Due to the dynamic mobility and flexible deployment characteristics, UAV Open Radio Units (O-RUs) in the TADRA architecture support the terrestrial infrastructure under overload or failure conditions, dynamically extending coverage, balancing traffic loads, and restoring service to maintain uninterrupted QoS across diverse and heterogeneous traffic demands. Unlike existing static or single-layer solutions, our proposed TADRA integrates RAN Intelligent Controllers (RICs) with a Hierarchical Traffic-Aware Multi-Agent Twin-Delayed (TMT) algorithm to optimize the allocation of computation and radio resources. This joint optimization problem is NP-hard, highly dynamic, and coupled across agents, making TMT a tractable and adaptive alternative. This hierarchical framework performs traffic prioritization at the upper (application) layer and resource allocation at the lower (MAC) layer, facilitating adaptive decision-making under diverse vehicular traffic patterns. Numerical results demonstrate that our solution provides substantial gains over MATD3, MADDPG, and GA, achieving 17% lower latency, 10% higher throughput, 14% lower energy consumption, and 6.5% higher reliability.
Hayla Nahom Abishu, Ahmed Badawy, Amr Mohamed et al.· IEEE Transactions on Network...· 0 citations
This paper proposes a hierarchical computation framework that flexibly supports task execution across local vehicles, neighboring vehicles, RSUs, and cloud resources, and designs an efficient task migration and resource scheduling strategy that improves overall system performance under dynamic network conditions.
Beyond 5G and future 6G services require radio access networks to support heterogeneous applications with diverse latency, reliability, throughput, mobility, and computing requirements. These challenges are particularly pronounced in vehicle-to-network (V2N) communications because of high mobility, dynamic channel conditions, frequent handovers, and heterogeneous service requirements. Conventional traffic-steering methods primarily rely on radio-side indicators, while computing-resource availability and traffic-specific computation demands are often considered separately. To address this limitation, this paper proposes a coordinated communication and computing resource management framework for O-RAN-based V2N communications. The framework integrates a traffic-management rApp (TM-rApp) in the non-real-time RIC with a traffic-steering xApp (TS-xApp) in the near-real-time RIC to enable policy-based closed-loop control. Candidate cells are ranked using communication quality, computing-resource capability and availability, predicted throughput, mobility characteristics, and traffic-class priority. As a proof-of-concept supporting component, proactive throughput forecasting is evaluated using standalone LSTM and stacked ensemble (S-LSTM) models based on lagged radio, mobility, load, and throughput features. The S-LSTM provides an adaptive mechanism for combining base learners but does not achieve a statistically significant improvement over the standalone LSTM; moreover, the forecasting evaluation uses fixed, non-optimized hyperparameters and a single chronological train–test split without cross-validation. Accordingly, the prediction results are interpreted as preliminary evidence of forecasting feasibility rather than as a definitive predictive-performance contribution. The framework further incorporates O-RAN-compatible traffic-steering policies, a minimum dwell-time constraint, and priority-aware resource allocation. Evaluation using a real-world corridor based on Al Haramain Expressway Road in Jeddah and a synthetic straight-highway scenario shows that the proposed method improves SLA compliance over RSS and HHAARC, achieves the highest computing-resource satisfaction, and reduces handovers relative to RSS. The results demonstrate a balanced trade-off among SLA compliance, computing-resource satisfaction, delay, throughput, and mobility robustness, while also showing that load-aware steering can provide higher aggregate SLA compliance under specific traffic distributions.
Vehicular edge computing (VEC) has emerged as a key paradigm to support computation-intensive and delay-sensitive vehicular applications by offloading tasks from vehicles to nearby multi-access edge computing (MEC) servers. However, in realistic urban environments, task processing performance is heavily affected by heterogeneous vehicle-MEC interactions, spatiotemporal traffic dynamics, and continuously varying vehicle populations. To address these challenges, this paper considers a traffic-aware embodied edge intelligence-enabled vehicular network (EEIVN), where edge intelligence is grounded in the physical traffic environment by integrating VLM-based semantic perception with edge decision making. Based on this architecture, we formulate a reliability-constrained delay minimization problem (RDMP) by jointly optimizing task offloading ratio, computing resource allocation, and vehicle association, while constraining the queue reliability to mitigate queue-induced tail delay. To solve the NP-hard RDMP, we propose a VLM-multi-agent proximal policy optimization (VLM-MAPPO) approach that integrates a VLM-based traffic awareness method, a vehicle-adaptive MAPPO algorithm, and a vehicle association scoring and selection mechanism. Extensive simulations based on SUMO and CARLA demonstrate that the proposed VLM-MAPPO approach outperforms benchmarks in terms of task completion delay and tail delay, while maintaining comparable vehicle energy consumption and exhibiting robust scalability under dynamic traffic conditions and varying vehicle densities.
Xulong Qiao, Jian Wang, Zemin Sun et al.· IEEE Transactions on Cogniti...· 0 citations
With the rapid growth of Vehicular Edge Computing (VEC) and Mobile Edge Computing, efficient task offloading is essential for enhancing the computing and communication capabilities in vehicular networks. However, many existing methods suffer from slow convergence, load imbalance, and instability in dynamic, latency-sensitive environments. To address these challenges, we propose MAPPO-Lyapunov (MAPPO-L), a multi-agent offloading framework that integrates Multi-Agent Proximal Policy Optimization (MAPPO) with Lyapunov optimization. MAPPO-L enables distributed coordination among vehicles, roadside units (RSUs), and cloud servers, minimizing delay, improving resource utilization, and ensuring long-term stability. Lyapunov theory transforms long-term stability into per-slot optimizations, while MAPPO ensures efficient policy learning. An adaptive exploration mechanism dynamically adjusts exploration rates based on network dynamics, accelerating convergence and stabilizing training. Extensive simulations with real-world data show that MAPPO-L maintains task completion rates above 80%, converges 25%–37.5% faster than baselines, and reduces training fluctuations to 2.3%. Ablation studies confirm the critical roles of location, channel, and queue information, validating the robustness of MAPPO-L in practical VEC environments.
Lu Wei, Yong Yu, Jie Cui et al.· IEEE Transactions on Network...· 0 citations
The rapid growth of Internet of Vehicles (IoV) applications has imposed strict requirements on low-latency and energy-efficient computing services. This letter investigates a multi-Uncrewed Aerial Vehicle (UAV)-assisted IoV system, where multiple Mobile Edge Computing (MEC)-enabled UAVs (MUs) collaboratively provide computing services for vehicular terminals (VTs). To improve service capability, we propose an energy-efficient task offloading and load balancing scheme that jointly considers vehicle mobility, task offloading and migration, and computing resource allocation to formulate an optimization problem. To solve this problem, a collective learning (CL)-enabled multi-agent reinforcement learning (CL-MARL) algorithm is proposed, where each agent learns optimal policies through centralized training and collective cooperative learning. Simulation results demonstrate that the proposed scheme outperforms benchmark strategies in terms of energy efficiency, task completion rate, and load balancing.
Yongbin Wang, Peng Lin, Yan Liu et al.· IEEE Wireless Communications...· 0 citations