摘要
深度强化学习是人工智能领域的研究热点,利用深度学习的感知能力与强化学习的决策能力,实现从输入到输出的端对端控制。为研究基于深度强化学习的无人机应用发展现状并分析其发展趋势,对近几年来国内外关于深度强化学习的无人机应用进行了总结与回顾。介绍了深度强化学习基本原理及在无人机系统应用上的发展历程,从基于深度强化学习算法的多无人机协同、无人机避障与路径规划、无人机目标搜索与跟踪、无人机空战决策与控制设计、无人机通信与资源管理等几个方面,对近年来国内外的深度强化学习领域无人机系统应用的发展进行了归纳,总结了基于深度强化学习无人机系统应用的未来发展趋势。
Deep reinforcement learning is a research hotspot in the field of artificial intelligence,which uses the perceptual ability of deep learning and the decision-making ability of reinforcement learning to achieve end-to-end control from input to output.In order to study the development status of UAV applications based on deep reinforcement learning and analyze their development trends,the UAV applications of deep reinforcement learning at home and abroad in recent years have been summarized and reviewed.The basic principles of deep reinforcement learning and the development process in the application of UAV systems are introduced,and the development of UAV systems in the field of deep reinforcement learning at home and abroad in recent years is summarized from several aspects,such as multi-UAV collaboration based on deep reinforcement learning algorithms,UAV obstacle avoidance and path planning,UAV target search and tracking,UAV air combat decision-making and control design,UAV communication and resource management,etc.,and the future development trend of UAV system applications based on deep reinforcement learning is summarized.
作者
李波
黄晶益
万开方
宋超
Li Bo;Huang Jingyi;Wan Kaifang;Song Chao(School of Electronics Information,Northwestern Polytechnical University,Xi’an 710072,China)
出处
《战术导弹技术》
北大核心
2023年第1期58-68,共11页
Tactical Missile Technology
基金
国家自然科学基金(62003267)
陕西省重点研发计划(2023-GHZD-33)
中央高校基本科研业务费专项资金资助(G2022KY0602)
西安市科技计划项目——关键核心技术攻关工程项目计划(21RGZN0016)。