TL;DR
Simulation results show that MOMUS outperforms state-of-the-art VEC scheduling approaches, particularly under high-demand scenarios, achieving higher task completion rates while reducing monetary cost and maintaining acceptable latency.
The advancement of vehicle technology enables it to handle complex tasks such as augmented reality, automatic parking, and driving, which demand significant computational power and are sensitive to delays. However, the computational capabilities of these vehicles are limited. As a result, Vehicle Edge Computing (VEC) has been introduced to offload some computational burdens to neighbouring vehicles and edge servers. The main challenge of this system lies in managing offloading in scenarios with frequent disconnections and high-speed driving. To address this challenge, we have formulated the task offloading problem with the goal of minimizing overall time, considering its NP-hard nature and multiple constraints. Our proposed solution is a meta-heuristic approach that integrates the honey bee Food Foraging process with Genetics Algorithm (FFGA) for vehicular task offloading. In the food foraging process, scouts are sent to locate nearby area, which in our case are edge servers and gather contextual details, and upon their return, they perform a dance resembling the infinity symbol, where the central angle indicates the direction of the food field and the speed of the dance signifies the quality of the found food source. The vehicle client uses this information to select one edge server to be in charge of the offloading process. Subsequently, we apply the genetic algorithm to generate an optimized allocation scheme of tasks by considering contextual details of nearby nodes. These details include computational capacity, availability, location, and driving speed. The optimization operators of the genetic algorithm are used to determine the assignment of tasks to nodes. Extensive simulations have demonstrated that the proposed FFGA system outperforms other existing schemes, including the hybrid vehicular edge cloud (HVC), particle swarm optimization (PSO), and the multi-decision based offloading (MDO).
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.
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.
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.