Application of Deep Reinforcement Learning in UAV Networks and Mobile Edge Computing

Deep reinforcement learning (DRL) has demonstrated significant potential in managing dynamic and complex networked systems. In this paper, we address the challenge of efficient task offloading and energy-aware trajectory planning in unmanned aerial vehicle (UAV) networks within mobile edge computing (MEC) environments, where limited resources and real-time demands pose critical constraints. We propose a DRL-based optimization framework that leverages the proximal policy optimization (PPO) algorithm to jointly optimize task scheduling and UAV trajectory control, with the goal of minimizing system latency and energy consumption. Our approach integrates adaptive resource allocation with flight-aware decision-making to enhance both responsiveness and operational stability. Extensive simulation results validate the effectiveness of our method, demonstrating improved convergence speed, battery performance, and energy efficiency compared to baseline approaches. These findings further confirm that dynamic UAV networks can substantially benefit from the stability and adaptability offered by PPO, enabling robust task scheduling and trajectory control under resource constraints.