High Availability Migration Mechanism for UAV Failures in Edge ScenariosOA
Unmanned Aerial Vehicles(UAVs)are widely used in Internet of Things(IoT)scenarios,but factors like circuit malfunctions and network issues can lead to data loss,disconnections,and service interruptions,causing UAVs to go offline and fail to meet user needs.This problem can be effectively solved using UAV resource migration techniques.However,traditional methods lead to resource redundancy and cannot ensure effective migration decisions or the continuity of services.Therefore,we proposed a new UAV framework with a high-availability migration strategy to improve UAV swarm resource utilization.By combining Software Defined Network and Mobile Edge Computing technologies,we designed a UAV collaboration architecture model and introduced large language model agents in the edge layer for better task execution.Additionally,we created a two-tier collaborative UAV high-availability mechanism for handling failures in edge scenarios during UAV collaborative tasks.Our experiments on the container cloud platform showed a 42%reduction in failure recovery time and higher overall resource utilization compared to baseline algorithms.While our focus is on UAVs,our findings can potentially be applied to broader IoT and mobile edge computing scenarios such as traffic management systems and healthcare IoT systems.
Haoran Jin;Yi Li;Shengye Pang;Jinsong Shu;Chen Zhi;Xuhong Zhang;Yueshen Xu;Junlan Cai;Shaofeng Liu;Xiang Li;Xinkui Zhao
Zhejiang University-University of Illinois Urbana-Champaign Institute,Zhejiang University,Jiaxing 314499,ChinaSchool of Software Technology,Zhejiang University,Ningbo 315048,ChinaCollege of Computer Science and Technology,Zhejiang University,Hangzhou 310058,ChinaSchool of Software Technology,Zhejiang University,Ningbo 315048,ChinaSchool of Software Technology,Zhejiang University,Ningbo 315048,ChinaSchool of Software Technology,Zhejiang University,Ningbo 315048,ChinaSchool of Computer Science and Technology,Xidian University,Xi’an 710126,ChinaBeijing Life Science Academy,Beijing 102200,China Zhengzhou Tobacco Research Institute of China National Tobacco Corporation,Zhengzhou 450001,ChinaBeijing Life Science Academy,Beijing 102200,China Zhengzhou Tobacco Research Institute of China National Tobacco Corporation,Zhengzhou 450001,ChinaBeijing Life Science Academy,Beijing 102200,China Zhengzhou Tobacco Research Institute of China National Tobacco Corporation,Zhengzhou 450001,ChinaSchool of Software Technology,Zhejiang University,Ningbo 315048,China
航空航天
edge computingUnmanned Aerial Vehicle(UAV)collaborationtask migrationhigh availability
《Tsinghua Science and Technology》 2026 (2)
P.1075-1091,17
supported in part by the National Key R&D Program of China(No.2022YFF0902702)in part by the Key R&D Program of Ningbo(No.2023Z235),and in part by the Ningbo Yongjiang Talent Programme(No.2023A-198-G)in part by the Beijing Life Science Academy,Beijing,China(BLSA,No.2023000CB0020)in part by Henan Institute of Chinese Engineering Development Strategies,Zhengzhou,China(No.2023HENZDB01).
评论