期刊
IEEE TRANSACTIONS ON MOBILE COMPUTING
卷 22, 期 7, 页码 4056-4069出版社
IEEE COMPUTER SOC
DOI: 10.1109/TMC.2022.3146881
关键词
Navigation; Task analysis; Training; Stochastic processes; Base stations; Ad hoc networks; Autonomous aerial vehicles; UAV control; communication coverage; deep reinforcement learning; graph learning; stochastic policy
In this paper, a deep reinforcement learning (DRL) based control solution is proposed to navigate a swarm of unmanned aerial vehicles (UAVs) in an unexplored target area under partial observation, serving as Mobile Base Stations (MBSs) for optimal communication coverage. A novel network architecture called Deep Recurrent Graph Network (DRGN) is introduced to handle information loss and obtain spatial information through inter-UAV communication. Based on DRGN and maximum-entropy learning, a stochastic DRL policy named Soft Deep Recurrent Graph Network (SDRGN) is proposed. Extensive experiments demonstrate the superior performance and scalability of SDRGN compared to state-of-the-art approaches.
In this paper, we aim to design a deep reinforcement learning (DRL) based control solution to navigating a swarm of unmanned aerial vehicles (UAVs) to fly around an unexplored target area under partial observation, which serves as Mobile Base Stations (MBSs) providing optimal communication coverage for the ground mobile users. To handle the information loss caused by the partial observability, we introduce a novel network architecture named Deep Recurrent Graph Network (DRGN), which could obtain extra spatial information through graph-convolution based inter-UAV communication, and utilize historical features with a recurrent unit. Based on DRGN and maximum-entropy learning, we propose a stochastic DRL policy named Soft Deep Recurrent Graph Network (SDRGN). In SDRGN, a heuristic reward function is elaborated, which is based on the local information of each UAV instead of the global information; thus, SDRGN reduces the training cost and enables distributed online learning. We conducted extensive experiments to design the structure of DRGN and examine the performance of SDRGN. The simulation results show that the proposed model outperforms four state-of-the-art DRL-based approaches and three heuristic baselines, and demonstrate the scalability, transferability, robustness, and interpretability of SDRGN.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据