摘要
无人机自组织网络(FANET)被广泛应用于军事、应急救灾和环境监测等情况下的网络通信服务,良好的路由协议能为其在通信条件恶劣场景下的可靠传输提供保障。利用强化学习将路由选择描述为一个马尔可夫决策过程进行路由决策成为研究热点。为了更进一步地介绍和挖掘基于强化学习的FANET路由协议研究现状,首先介绍近几年来FANET传统路由协议上的一些改进;其次,基于强化学习的FANET路由协议研究的最新调研结果进行详细的介绍;同时,对路由研究算法中的状态、动作和奖励等建模规律进行深度挖掘,从路由的优化标准和强化学习优化过程等方面进行了比较;最后,根据目前基于强化学习FANET路由协议的研究现状进行总结和展望。
Flying Ad hoc network(FANET)are widely used for network communication services in situations such as military,emergency disaster relief and environmental monitoring,and good routing protocols can provide reliable transmission in scena-rios with harsh communication conditions.Using reinforcement learning to describe routing as a Markov decision process for routing decisions has become a hot research topic.In order to further introduce and dig into the current state of research on FANET routing protocols based on reinforcement learning,this paper firstly introduced some improvements on FANET traditional routing protocols in recent years.Secondly,it provided a detailed introduction on the latest research results on the research of FANET routing protocols based on reinforcement learning.At the same time,this paper deeply explored the modeling laws of state,action and reward in reinforcement learning-based routing research algorithms,and compared them in terms of routing optimization criteria and reinforcement learning optimization process.Finally,it presented a summary and outlook based on the current research status of reinforcement learning based FANET routing protocols.
作者
孙晨
莫国美
舒坚
Sun Chen;Mo Guomei;Shu Jian(School of Software,Nanchang Hangkong University,Nanchang 330063,China)
出处
《计算机应用研究》
CSCD
北大核心
2023年第7期1937-1946,共10页
Application Research of Computers
基金
国家自然科学基金资助项目(62062050)
江西省教育厅科技项目(GJJ170615)。
关键词
无人机自组织网络
路由协议
强化学习
flying Ad hoc network
routing protocol
reinforcement learning