Deep Reinforcement Learning Assisted UAV Path Planning Relying on Cumulative Reward Mode and Region Segmentation

IF 5.3 Q1 ENGINEERING, ELECTRICAL & ELECTRONIC IEEE Open Journal of Vehicular Technology Pub Date : 2024-03-16 DOI:10.1109/OJVT.2024.3402129
Zhipeng Wang;Soon Xin Ng;Mohammed EI-Hajjar
{"title":"Deep Reinforcement Learning Assisted UAV Path Planning Relying on Cumulative Reward Mode and Region Segmentation","authors":"Zhipeng Wang;Soon Xin Ng;Mohammed EI-Hajjar","doi":"10.1109/OJVT.2024.3402129","DOIUrl":null,"url":null,"abstract":"In recent years, unmanned aerial vehicles (UAVs) have been considered for many applications, such as disaster prevention and control, logistics and transportation, and wireless communication. Most UAVs need to be manually controlled using remote control, which can be challenging in many environments. Therefore, autonomous UAVs have attracted significant research interest, where most of the existing autonomous navigation algorithms suffer from long computation time and unsatisfactory performance. Hence, we propose a Deep Reinforcement Learning (DRL) UAV path planning algorithm based on cumulative reward and region segmentation. Our proposed region segmentation aims to reduce the probability of DRL agents falling into local optimal trap, while our proposed cumulative reward model takes into account the distance from the node to the destination and the density of obstacles near the node, which solves the problem of sparse training data faced by the DRL algorithms in the path planning task. The proposed region segmentation algorithm and cumulative reward model have been tested in different DRL techniques, where we show that the cumulative reward model can improve the training efficiency of deep neural networks by 30.8% and the region segmentation algorithm enables deep Q-network agent to avoid 99% of local optimal traps and assists deep deterministic policy gradient agent to avoid 92% of local optimal traps.","PeriodicalId":34270,"journal":{"name":"IEEE Open Journal of Vehicular Technology","volume":"5 ","pages":"737-751"},"PeriodicalIF":5.3000,"publicationDate":"2024-03-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10531630","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Open Journal of Vehicular Technology","FirstCategoryId":"1085","ListUrlMain":"https://ieeexplore.ieee.org/document/10531630/","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
引用次数: 0

Abstract

In recent years, unmanned aerial vehicles (UAVs) have been considered for many applications, such as disaster prevention and control, logistics and transportation, and wireless communication. Most UAVs need to be manually controlled using remote control, which can be challenging in many environments. Therefore, autonomous UAVs have attracted significant research interest, where most of the existing autonomous navigation algorithms suffer from long computation time and unsatisfactory performance. Hence, we propose a Deep Reinforcement Learning (DRL) UAV path planning algorithm based on cumulative reward and region segmentation. Our proposed region segmentation aims to reduce the probability of DRL agents falling into local optimal trap, while our proposed cumulative reward model takes into account the distance from the node to the destination and the density of obstacles near the node, which solves the problem of sparse training data faced by the DRL algorithms in the path planning task. The proposed region segmentation algorithm and cumulative reward model have been tested in different DRL techniques, where we show that the cumulative reward model can improve the training efficiency of deep neural networks by 30.8% and the region segmentation algorithm enables deep Q-network agent to avoid 99% of local optimal traps and assists deep deterministic policy gradient agent to avoid 92% of local optimal traps.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
依赖累积奖励模式和区域分割的深度强化学习辅助无人机路径规划
近年来,无人驾驶飞行器(UAV)已被考虑用于灾害预防和控制、物流和运输以及无线通信等许多应用领域。大多数无人飞行器需要使用遥控器进行手动控制,这在许多环境中都具有挑战性。因此,自主无人机引起了人们极大的研究兴趣,而大多数现有的自主导航算法都存在计算时间长、性能不理想等问题。因此,我们提出了一种基于累积奖励和区域分割的深度强化学习(DRL)无人机路径规划算法。我们提出的区域分割旨在降低DRL代理陷入局部最优陷阱的概率,而我们提出的累积奖励模型考虑了节点到目的地的距离和节点附近的障碍物密度,解决了DRL算法在路径规划任务中面临的训练数据稀疏的问题。我们在不同的 DRL 技术中测试了所提出的区域分割算法和累积奖励模型,结果表明累积奖励模型能将深度神经网络的训练效率提高 30.8%,区域分割算法能使深度 Q 网络代理避免 99% 的局部最优陷阱,并帮助深度确定性策略梯度代理避免 92% 的局部最优陷阱。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
CiteScore
9.60
自引率
0.00%
发文量
25
审稿时长
10 weeks
期刊最新文献
Efficient Modeling of Interest Forwarding in Information Centric Vehicular Networks Multi-Agent Deep Reinforcement Learning Based Optimizing Joint 3D Trajectories and Phase Shifts in RIS-Assisted UAV-Enabled Wireless Communications Digital Twin-Empowered Green Mobility Management in Next-Gen Transportation Networks Fairness-Aware Utility Maximization for Multi-UAV-Aided Terrestrial Networks LiFi for Industry 4.0: Main Features, Implementation and Initial Testing of IEEE Std 802.15.13
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1