| 534 | 0 | 1131 |
| 下载次数 | 被引频次 | 阅读次数 |
针对现代战争智能化转型对无人集群协同感知的迫切需求，系统性综述了战场环境下关键技术的研究进展与挑战。首先，在通信机制方面，突破传统带宽限制的三阶段握手机制、空间置信图模型、图神经网络与通信感知一体化等新兴技术显著提升了感知精度与通信效率。其次，在跨域协同领域，“蜂群+狼群”协同、无人机-无人车天地一体化定位与规划体系通过目标检测算法与鲁棒滤波技术增强了复杂环境适应性。再次，在智能算法层面，改进粒子群优化算法、深度强化学习方法以及视觉Transformer及其轻量化变体优化了路径规划与目标检测性能。最后，研究结果表明，当前技术仍面临强对抗环境下动态适应性不足、时空通信效率瓶颈、多模态语义级融合缺失、算力能耗矛盾等挑战。未来需重点发展基于博弈强化学习的智能协同、空地海天多域智能融合、云边端协同计算及轻量化混合架构，推动无人集群从功能协同向智能决策跃升，为联合作战提供颠覆性态势认知能力。
Abstract：Addressing the urgent demand for collaborative perception in unmanned swarms driven by the intelligent transformation of modern warfare, this paper systematically reviews research advances and challenges in key technologies for battlefield environments. Firstly, in terms of communication mechanisms, emerging technologies such as the three-phase handshake mechanism that breaks through traditional bandwidth limiations, spatial confidence map model, Graph Neural Networks(GNN), and Integrated Sensing and Communication(ISAC) have significantly enhanced perception accuracy and communication efficiency. Secondly, in the field of cross-domain collaboration, the "Swarm+Wolf Pack" collaboration and the air-ground integrated positioning and planning system for UAV-UGV improve adaptability in complex environments through object detection algorithms and robust filtering techniques. Thirdly, regarding intelligent algorithms, improved Particle Swarm Optimization(PSO) algorithms, Deep Reinforcement Learning(DRL) methods, and Vision Transformers(ViT) with their lightweight variants enhance the path planning and object detection capabilities. Finally, the study identifies persistent challenges including insufficient dynamic adaptability in highly contested environments, bottlenecks in spatio-temporal communication efficiency, lack of multi-modal semantic-level fusion, and the computing-power-versus-energy-consumption trade-off. Future efforts should prioritize developing game-theoretic reinforcement learning, multi-domain(air-land-sea-space) intelligent fusion, cloud-edge-end collaborative computing, and lightweight hybrid architectures to propel unmanned swarms from functional cooperation towards intelligent decision-making, thereby providing transformative situational awareness capabilities for joint operations.
[1]张洪碧，孟凡松，翟东航.某局部冲突中无人机作战运用及启示[J].指挥控制与仿真，2022, 44(4):16-20.ZHANG H B, MENG F S, ZHAI D H. UAV operational application and its enlightenment in a certain regional conflict[J]. Command Control&Simulation, 2022, 44(4):16-20(in Chinese).
[2]董希旺，于江龙，化永朝，等.集群系统智能协同IOODA技术体系架构与关键技术[J].航空学报，2025, 46(4):6-43.DONG X W, YU J L, HUA Y Z, et al. Architecture and key technologies of intelligent cooperative IOODA technology system for swarm systems[J]. Acta Aeronautica et Astronautica Sinica, 2025, 46(4):6-43(in Chinese).
[3]段海滨，张岱峰，范彦铭，等.从狼群智能到无人机集群协同决策[J].中国科学：信息科学，2019, 49(1):112-118.DUAN H B, ZHANG D F, FAN Y M, et al. From wolf pack intelligence to UAV swarm cooperative decision making[J]. Scientia Sinica(Informationis), 2019, 49(1):112-118(in Chinese).
[4] LIU Y C, TIAN J, MA C Y, et al. Who2com:Collaborative perception via learnable handshake communication[C]//2020 IEEE International Conference on Robotics and Automation(ICRA), Piscataway, NJ:IEEE,2020.
[5] LIU Y C, TIAN J, GLASER N, et al. When2com:Multi-agent perception via communication graph grouping[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR), Piscataway, NJ:IEEE, 2020.
[6] HU Y, FANG S, LEI Z, et al. Where2comm:Communication-efficient collaborative perception via spatial confidence maps[EB/OL]. 2022-09-27[2025-08-10].arXiv:2209.12836.
[7] HAN J Q, CEN J C, WU L M, et al. A survey of geometric graph neural networks:Data structures, models and applications[J]. Frontiers of Computer Science ,2025, 19(11):1-20.
[8]吕扬.仿"萤火虫"通信无人机研究取得新突破[N].陕西日报，2024-05-07.LV Y. New breakthrough in research on firefly-inspired communication drones[N]. Shanxi Daily, 2024-05-07(in Chinese).
[9] LOU L B, LIU X X. Dual-UAV-enabled secure communication and sensing for A2G-ISAC systems with maneuverable jamming[EB/OL]. 2025-05-20[2025-08-10]. arXiv:2505.08523.
[10]中兴.中兴通讯联合中国移动研究院发布“算力路由（CATS）AI推理及视联网应用样板”，共推算力网络商用进程[EB/OL]. 2025-09-10[2025-10-11].https：//www. zte. com. cn/china/solutions_latest/ip_network_cn/_/_-_--cats-ai_-_.html.ZTE. ZTE jointly with China mobile research institute releases“computing-aware routing(CATS)AI inference and video networking application template” to boost commercialization of computing force network[EB/OL].2025-09-10[2025-10-11]. https：//www. zte. com. cn/china/solutions_latest/ip_network_cn/_/_-_--cats-ai_-_.html(inChinese).
[11] DAI J, LIU Y, HAO X Y, et al. Unmanned ground vehicle-unmanned aerial vehicle relative navigation robust adaptive localization algorithm[J]. IET Science, Measurement&Technology, 2023, 17(5):210-220.
[12] WANG D, LIAN B, TANG C. UGV-UAV robust cooperative positioning algorithm with object detection[J].IET Intelligent Transport Systems, 2021, 15(7):945-958.
[13]熊泽昊，李杰，牛轶峰，等.生成对抗网络及其在无人系统中的应用[J].无人系统技术，2023, 6(1):1-13.XIONG Z H, LI J, NIU Y F, et al. Generative adversarial network and its application in unmanned systems[J].Unmanned Systems Technology, 2023, 6(1):1-13(in Chinese).
[14]南京邮电大学.关于2023年度重点实验室开放研究基金课题资助通知[EB/OL]. 2023-11-24[2025-10-11].https：//bwcsn. njupt. edu. cn/2023/1124/c3380a250642/page.htm.Nanjing University of Posts and Telecommunications.Notice on the 2023 open research fund project funding of the key laboratory[EB/OL]. 2023-11-24[2025-10-11].https：//bwcsn. njupt. edu. cn/2023/1124/c3380a250642/page.htm(in Chinese).
[15] LI J G, DENG G Q, LUO C W, et al. A hybrid path planning method in unmanned air/ground vehicle(UAV/UGV)cooperative systems[J]. IEEE Transactions on Vehicular Technology, 2016, 65(12):9585-9596.
[16]张云飞，王宏伦，张梦华，等.基于深度强化学习和导航向量场的卫星规避拦截算法[J].无人系统技术，2023, 6(1):62-71.ZHANG Y F, WANG H L, ZHANG M H, et al. Satellite avoidance algorithm based on deep reinforcement learning and guidance vector field[J]. Unmanned Systems Technology, 2023, 6(1):62-71(in Chinese).
[17] CHEN Z, LUO F, ZHAI C. Obstacle avoidance strategy for quadrotor UAV based on improved particle swarm optimization algorithm[C]//2019 Chinese Control And Decision Conference(CCDC), Piscataway, NJ:IEEE,2019.
[18] SKRZYPECKI S, TARAPATA Z, PIERZCHAA D.Combined PSO methods for UAVs swarm modelling and simulation[C]//2020 International Conference on Military Technologies(ICMT), Piscataway, NJ:IEEE,2020.
[19] HUANG C, FEI J. UAV Path planning based on particle swarm optimization with global best path competition[J]. International Journal of Pattern Recognition and Artificial Intelligence, 2018, 32(6):1859008.
[20] ZHEN X, ENZE Z, QINGWEI C. Rotary unmanned aerial vehicles path planning in rough terrain based on multi-objective particle swarm optimization[J]. Journal of Systems Engineering and Electronics, 2020, 31(1):12-25.
[21] SHANMUGASUNDAR S, PAVAN K Y, HARIHARAN C,et al. Efficient path planning for UAV formation via comprehensively improved particle swarm optimization[J]. ISA Transactions, 2020, 97:415-430.
[22] PHUNG M D, HA Q P. Safety-enhanced UAV path planning with spherical vector-based particle swarm optimization[J]. Applied Soft Computing, 2021, 107:107376.
[23] BHAT S F, BIRKL R, WOFK D, et al. ZoeDepth:Zero-shot transfer by combining relative and metric depth[EB/OL]. 2023-02-24[2025-08-10]. arXiv:2302.12288.
[24] LAGA H, JOSPIN L V, BOUSSAID F, et al. A survey on deep learning techniques for stereo-based depth estimation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(4):1738-1764.
[25] CAI H. Model acceleration for efficient deep learning computing[D]. San Francisco:Stanford University,2024.
[26] HU C Y, ZHENG J. Fusionformer:A multi-sensory fusion in Bird's-Eye-View and temporal consistent transformer for 3D object detection[EB/OL]. 2023-09-11[2025-08-10]. arXiv:2309.05257.
[27] LI Z, ZHOU H, REI M, SPECIA L. DiffuseDef:Improved robustness to adversarial attacks via iterative denoising[EB/OL]. 2024-06-28[2025-08-10]. arXiv:2407.00248.
[28] LAGA H, JOSPIN L V, BOUSSAID F, et al. A survey on deep learning techniques for stereo-based depth estimation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(4):1738-1764.
[29] ZHANG H, LI F, LIU S, et al. DINO:DETR with improved denoising anchor boxes for end-to-end object detection[EB/OL]. 2022-03-27[2025-08-10]. arXiv:2203.03605.
[30] LIU Z, HU H, LIN Y, et al. Swin transformer V2:Scaling up capacity and resolution[C]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR), Piscataway, NJ:IEEE, 2022.
[31] LI Y, MAO H, GIRSHICK R, et al. Exploring plain vision transformer backbones for object detection[EB/OL].2022-03-21[2025-08-10]. arXiv:2203.16527.
[32] QIAN Z J. ECViT:Efficient convolutional vision transformer with local-attention and multi-scale stages[EB/OL].2025-04-21[2025-08-10]. arXiv:2504.14825.
[33] XIANG Y M, LI S Z, LI R P, et al. Decentralized consensus inference-based hierarchical reinforcement learning for multiconstrained UAV pursuit-evasion game[J].IEEE Transactions on Neural Networks and Learning Systems, 2025, 36(5):1-16.
[34] ZHANG X, BHUYAN M. A novel multi-agent deep reinforcement learning-enabled distributed power allocation scheme for mmWave cellular networks[C]//2023IEEE International Conference on Communications(ICC), Piscataway, NJ:IEEE, 2023.
[35] ZHU K, YU Y T, WU B, et al. GO-INO:Graph optimization MEMS-IMU/NHC/odometer integration for ground vehicle positioning[J]. Micromachines, 2022,13(9):1400.
基本信息:
DOI：10.19942/j.issn.2096-5915.2025.05.43
中图分类号:TN92
引用信息:
[1]刘言,胡旭东.战场环境智能无人集群协同感知关键技术综述[J].无人系统技术,2025,8(05):30-44.DOI:10.19942/j.issn.2096-5915.2025.05.43.