This work details a scalable framework to orchestrate a swarm of rotary-wing UAVs serving as cellular relays to facilitate beyond line-of-sight connectivity and traffic offloading for ground users. First, a Multiscale Adaptive Energy-conscious Scheduling and TRajectory Optimization (MAESTRO) framework is developed for a single UAV. Aiming to minimize the time-averaged latency to serve user requests, subject to an average UAV power constraint, it is shown that the optimization problem can be cast as a semi-Markov decision process, and exhibits a multiscale structure: outer actions on radial wait velocities and terminal service positions minimize the long-term delay-power trade-off, optimized via value iteration; given these outer actions, inner actions on angular wait velocities and service trajectories minimize a short-term delay-energy cost. A novel hierarchical competitive swarm optimization scheme is developed in the inner optimization, to devise high-resolution trajectories via iterative pair-wise updates. Next, MAESTRO is eXtended to UAV swarms (MAESTRO-X) via scalable policy replication: enabled by a decentralized command-and-control network, the optimal single-agent policy is augmented with spread maximization, consensus-driven conflict resolution, adaptive frequency reuse, and piggybacking. Numerical evaluations show that, for user requests of 10 Mbits, generated according to a Poisson arrival process with rate 0.2 req/min/UAV, single-agent MAESTRO offers 3.8x faster service than a high-altitude platform and 29% faster than a static UAV deployment; moreover, for a swarm of 3 UAV-relays, MAESTRO-X delivers data payloads 4.7x faster than a successive convex approximation scheme; and remarkably, a single UAV optimized via MAESTRO outclasses 3 UAVs optimized via a deep-Q network by 38%.

翻译：本文详细介绍了一个可扩展的框架，用于编排作为蜂窝中继的旋转翼无人机群，以支持地面用户的超视距连接和流量卸载。首先，针对单架无人机，开发了多尺度自适应能量感知调度与轨迹优化（MAESTRO）框架。以在平均无人机功率约束下最小化服务用户请求的时间平均延迟为目标，研究表明该优化问题可建模为半马尔可夫决策过程，并呈现多尺度结构：外层动作（径向等待速度和终端服务位置）通过值迭代优化最小化长期延迟-功率权衡；给定这些外层动作，内层动作（角向等待速度和服务轨迹）最小化短期延迟-能量成本。在内层优化中，提出了一种新颖的层次竞争性群体优化方案，通过迭代成对更新设计高分辨率轨迹。随后，通过可扩展策略复制将MAESTRO扩展到无人机群（MAESTRO-X）：借助分散式指挥控制网络，将单智能体最优策略与扩散最大化、共识驱动的冲突解决、自适应频率复用及捎带机制相结合。数值评估表明，对于速率为0.2请求/分钟/无人机、按泊松到达过程生成的10兆比特用户请求，单智能体MAESTRO的服务速度比高空平台快3.8倍，比静态无人机部署快29%；此外，对于3架无人机中继群，MAESTRO-X的数据载荷传输速度比逐次凸近似方案快4.7倍；值得注意的是，通过MAESTRO优化的单架无人机比通过深度Q网络优化的3架无人机性能高出38%。