LABOR: Latency and Buffer Occupancy Optimized Multimodal Mission Offloading Strategy in Resource-Constrained UAV Swarm Networks

The rapid development of uncrewed aerial vehicle (UAV) swarm technologies has significantly expanded their applications in fields like road rescue and agricultural monitoring, where multimodal data streams from sensors such as infrared, radar, and visible light are critical. However, processing such data in real-time is computationally intensive, particularly with UAVs’ limited onboard resources. Missions involving multimodal data are typically divided into subtasks, which can be processed either locally or offloaded to edge servers. While offloading subtasks helps alleviate onboard computational load, it introduces transmission delays and potential server overloads. Moreover, the interdependence of subtasks creates a synchronization bottleneck, where the slowest subtask determines the mission completion time. Early completed subtasks also occupy limited UAV onboard buffer resources. To address these challenges, we propose LABOR, a proximal policy optimization (PPO)-based mission offloading strategy. LABOR dynamically determines whether each subtask should be processed locally or offloaded, optimizing the balance between latency, resource availability, and computational requirements. Through extensive simulations in various scenarios, LABOR is shown to significantly outperform other offloading strategies, achieving up to 64.0% and 80.1% reductions in latency and buffer occupancy, respectively. This work provides a promising solution for efficient mission offloading in resource-constrained UAV swarm networks, enhancing both real-time performance and resource utilization.