基于5G网联无人机的智慧巡防应用研究  

Research on intelligent patrol application based on 5G networked unmanned aerial vehicle

作　　者：于海洋 董石磊[1] 张学智[1] Yu Haiyang;Dong Shilei;Zhang Xuezhi(Beijing Research Institute of China Telecom Corporation Limited,Beijing 102209 China)

机构地区：[1]中国电信股份有限公司研究院,北京102209

出　　处：《电子技术应用》2021年第5期11-13,18,共4页Application of Electronic Technique

基　　金：国家重点研发计划项目(2020YFB2104400)。

摘　　要：为提升治安巡防等的快速响应与智慧化水平,基于5G、无人机、边缘计算(MEC)以及人工智能(AI),构建了一套网联无人机智慧巡防解决方案。通过专用5G通信终端实现无人机网联接入,借助5G上行超大带宽实现高清视频远程回传,基于5G MEC超低时延实现无人机超视距精准控制。针对低空无人机视频特征,优化人脸识别算法,实时精准捕捉人脸并进行人物特征AI识别,实现智慧巡防。在施工工地进行了应用实践与测试,对工地附近人员是否佩戴口罩进行实时监测,达到了预期效果。In order to improve the rapid response and intelligent level of public security patrol,a set of networked UAV intelligent patrol solution is constructed based on 5G,UAV(unmanned aerial vehicle),MEC(multi-access edge computing)and AI(artificial intelligence).UAV network connection is realized through dedicated 5G communication terminal,high-definition video remote backhaul is realized with 5G uplink ultra-large bandwidth,and ultra-low latency of 5G MEC is used to realize precise control of UAV beyond visual range.Aiming at the video features of low-altitude UAVs,the face recognition algorithm is optimized,the face is accurately captured in real time and the character feature AI recognition is performed to realize smart patrol and defense.Application practices and tests were carried out on the construction site,real-time monitoring of whether people near the construction site were wearing masks,the expected results were achieved.

关 键 词：智慧巡防 5G 无人机 MEC AI 

分 类 号：TN919.5[电子电信—通信与信息系统]