The integration of Unmanned Aerial Vehicles (UAVs) into Mobile Edge Computing (MEC) systems has emerged as a transformative solution for latency-sensitive applications, leveraging UAVs’ unique advantages in mobility, flexible deployment, and on-demand service provisioning. This paper proposes a novel multi-agent reinforcement learning framework, termed Multi-Agent Twin Delayed Deep Deterministic Policy Gradient for Task Offloading and Resource Allocation (MATD3-TORA), to optimize task offloading and resource allocation in UAV-assisted MEC networks. The framework enables collaborative decision making among multiple UAVs to efficiently serve sparsely distributed ground mobile devices (MDs) and establish an integrated mobility, communication, and computational offloading model, which formulates a joint optimization problem aimed at minimizing the weighted sum of task processing latency and UAV energy consumption. Extensive experiments demonstrate that the algorithm achieves improvements in system latency and energy efficiency compared to conventional approaches. The results highlight MATD3-TORA’s effectiveness in addressing UAV-MEC challenges, including mobility–energy tradeoffs, distributed decision making, and real-time resource allocation.

With the rapid advancement of drone technology, the efficient distribution of drones has garnered significant attention. Central to this discourse is the energy consumption of drones, a critical metric for assessing energy-efficient distribution strategies. Accordingly, this study delves into the energy consumption factors affecting drone distribution. A primary challenge in drone distribution lies in devising optimal, energy-efficient routes for drones. However, traditional routing algorithms, predominantly heuristic-based, exhibit certain limitations. These algorithms often rely on heuristic rules and expert knowledge, which can constrain their ability to escape local optima. Motivated by these shortcomings, we propose a novel multi-agent deep reinforcement learning algorithm that integrates a drone energy consumption model, namely EMADRL. The EMADRL algorithm first formulates the drone routing problem within a multi-agent reinforcement learning framework. It subsequently designs a strategy network model comprising multiple agent networks, tailored to address the node adjacency and masking complexities typical of multi-depot vehicle routing problem. Training utilizes strategy gradient algorithms and attention mechanisms. Furthermore, local and sampling search strategies are introduced to enhance solution quality. Extensive experimentation demonstrates that EMADRL consistently achieves high-quality solutions swiftly. A comparative analysis against contemporary algorithms reveals EMADRL’s superior energy efficiency, with average energy savings of 5.96% and maximum savings reaching 12.45%. Thus, this approach offers a promising new avenue for optimizing energy consumption in last-mile distribution scenarios.