Abstract
The effectiveness of autonomous Unmanned Aerial Vehicle (UAV) swarms in providing reliable Mobile Edge Computing (MEC) services is critically dependent on two conflicting performance metrics: the overall task completion time, which dictates service responsiveness, and the swarm's total energy consumption, which determines the network's operational longevity. This optimization challenge is further complicated by the unique operational constraints of aerospace systems, such as maintaining safe separation distances and managing finite onboard battery energy. To address this core challenge, we propose the Nutcracker-Inspired Hybrid Genetic Algorithm, hereinafter the Nutcracker Genetic Algorithm (NGA), a scheduling framework designed to find superior solutions for this trade-off. Unlike traditional approaches that can prematurely converge on inefficient schedules, NGA employs a unique cache-and-recovery mechanism, reinforced by specialized crossover and mutation operators, to autonomously optimize the fleet size and repair bottleneck flight trajectories. We rigorously validate NGA's performance on large-scale mission scenarios incorporating rigorous rotary-wing aerodynamic power models against the original NOA and other state-of-the-art algorithms. The results demonstrate that our framework can reduce task completion time by up to 42.8% while ensuring physical feasibility. This substantial performance gain highlights NGA's potential as a key enabling method for enhancing the scheduling efficiency and operational sustainability of resource-constrained UAV-assisted MEC networks.