文献类型：期刊文献

英文题名：UAV Swarm Collaborative Transhipment Scheduling with Deep Reinforcement Learning

作者：Wang, Ershen[1,2];Wang, Peipei[1];Hong, Chen[3,4];Yu, Tengli[1];Zhao, Hongsheng[5];Wu, Peifeng[6];Chen, Aidong[3,4];Xu, Song[1]

第一作者：Wang, Ershen

通讯作者：Hong, C[1];Hong, C[2]

机构：[1]Shenyang Aerosp Univ, Coll Elect & Informat Engn, Shenyang 110136, Peoples R China;[2]State Key Lab Extreme Environm Optoelect Dynam Mea, Taiyuan 038507, Peoples R China;[3]Beijing Union Univ, Multiagent Syst Res Ctr, Beijing 100101, Peoples R China;[4]Beijing Union Univ, Coll Robot, Beijing 100101, Peoples R China;[5]Aviat Ind Corp China, Chinese Aeronaut Radio Elect Res Inst, Shanghai 200233, Peoples R China;[6]Beijing Union Univ, Sch Urban Rail Transit & Logist, Beijing 100101, Peoples R China

第一机构：Shenyang Aerosp Univ, Coll Elect & Informat Engn, Shenyang 110136, Peoples R China

通讯机构：[1]corresponding author), Beijing Union Univ, Multiagent Syst Res Ctr, Beijing 100101, Peoples R China;[2]corresponding author), Beijing Union Univ, Coll Robot, Beijing 100101, Peoples R China.|[1141739]北京联合大学机器人学院;[11417]北京联合大学;

年份：2026

外文期刊名：UNMANNED SYSTEMS

收录：EI(收录号：20261720559454);WOS:【ESCI(收录号:WOS:001742727600001)】；

基金：The research described in this paper was supported by the National Science Foundation for Young Scientists of China (GrantNo.52502397),the Basic Science Center Program of the National Natural Science Foundation of China (62388101), Fund of State Key Laboratory of Extreme Environment Optoelectronic Dynamic Measurement Technology and Instrument (2023-SYSJJ-04), the Aeronautical Science Foundation of China (20240055054001),Open Fund of Key Laboratory of Technology and Equipment of Tianjin Urban Air Transportation System(TJKL-UAM-202305), Joint Fund of Ministry of Natural Resources Key Laboratory of Spatiotemporal Perception and Intelligent Processing(232203),the Applied Basic Research Programs of Liaoning Province(2025JH2/101300011) and the General Program of Liaoning Province Education Department (20250054, 310125011).

语种：英文

外文关键词：Multi-agent reinforcement learning; deep learning; UAV swarm; collaborative scheduling

摘要：As intelligent manufacturing continues to emerge as a dominant industrial paradigm, unmanned aerial vehicles (UAVs) have proven instrumental in enhancing workshop transhipment efficiency through their inherent operational flexibility. In this paper, we develop a comprehensive UAV swarm collaborative transhipment scheduling model with respect to three-dimensional continuous environments, and introduce Soft-QMIX that systematically integrates maximum entropy with an order-preserving transformation mechanism into collaborative transhipment scheduling tasks, which shows that it can effectively enhance the strategy exploration capability within three-dimensional continuous action spaces and significantly improve UAV's ability to achieve globally optimal strategies. Besides, a comprehensive simulation environment for three-dimensional UAV swarm scheduling is constructed, with experimental evaluations conducted under different UAV swarm sizes. The results reveal that Soft-QMIX consistently delivers superior cumulative rewards, loss stability, and execution efficiency, approximately achieving 7% increase in cumulative reward for 4-UAV swarm and 5% increase for 8-UAV swarm compared with QMIX. Meanwhile, compared with QMIX, the execution efficiency is improved approximately by 8% and 12.5% for 4-UAV and 8-UAV swarm, respectively. Our work will provide insights for collaborative transhipment scheduling of UAV swarm in complex three-dimensional continuous scenes.