DECCo-A Dynamic Task Scheduling Framework for Heterogeneous Drone Edge Cluster

• DOI: 10.3390/drones7080513
• Publication Date: 2023-08-03T15:23:03Z
• Authors: Zhiyang Zhang; Die Wu; Fengli Zhang; Ruijin Wang
• Keywords: Drone

The heterogeneity of unmanned aerial vehicle (UAV) nodes and the dynamic service demands make task scheduling particularly complex in drone edge cluster (DEC) scenario. In this paper, we provide a universal intelligent collaborative framework, named DECCo, which schedules dynamically changing requests for heterogeneous DEC. Benefiting from latest advances deep reinforcement learning (DRL), DECCo autonomously learns strategies with high response rates low communication latency through Advantage Actor–Critic algorithm, avoids interference resource overload local downtime while ensuring load balancing. To better adapt to real scenario, switches between heuristic DRL-based solutions based on real-time performance, thus avoiding suboptimal decisions that severely affect Quality Service (QoS) Experience (QoE). With flexible parameter control, can various clusters. Google Cluster Usage Traces are used verify effectiveness DECCo. Therefore, our work represents state-of-the-art method