计算机科学 ›› 2025, Vol. 52 ›› Issue (10): 287-295.doi: 10.11896/jsjkx.240700193
潘云伟, 李敏, 曾祥光, 邢丽静, 黄傲
PAN Yunwei, LI Min, ZENG Xiangguang, XING Lijing, HUANG Ao
摘要： 由多个智能体组成的多智能体系统,能够完成复杂的任务。针对多智能体在复杂环境中的编队成型,以及编队遭受冲击时的编队重组,提出了一种基于编队形状离散层的分布式编队控制方法。首先,对编队形状进行离散和迭代,扩大其影响范围,并将编队信息共享给每个智能体;其次,针对存在障碍物的编队环境,设计动态协商算法实时调整编队集结位置;最后,采用分布式控制的方式,利用传感器信息和编队形状信息设计速度控制器,实现动态避障和复杂编队成型。实验结果表明,所提方法能够很好地引导多智能体形成复杂的编队形状,在有障碍物的环境中,实现编队避障、偏移和重组。通过编队成型时间和编队效果函数对实验结果进行评价和分析,验证了所提方法具有较好的环境适应性和有效性。
中图分类号:
| [1]GELBLUM A,PINKOVIEZKY I,FONIO E,et al.Ant groups optimally amplify the effect of transiently informed individuals [J].Nature Communications,2015,6(1):7729. [2]YIN Z,HE W,ZOU Y,et al.Efficient Formation of Flapping-wing Aerial Vehicles Based on Wild Geese Queue Effect [J].Zidonghua Xuebao/Acta Automatica Sinica,2021,47(6):1355-1367. [3]TSANG K F E,NI Y Q,WONG C F R,et al.A novel ware-house multi-robot automation system with semi-complete and computationally efficient path planning and adaptive genetic task allocation algorithms[C]//2018 15th International Conference on Control,Automation,Robotics and Vision.2018. [4]SUN D,TANG Q R,LIU M H,et al.Research on swarm formation control of modular underwater robots[J].Ship Science and Technology,2024,46(2):68-73. [5]SONG W,WANG T,YANG G X,et al.A Self-OrganizationFormation Configuration Based Assignment Probability and Collision Detection [J].Journal of Systems Engineering and Electronics,2024,35(1):222-232. [6]CHOI Y H,KIM D.Distance-Based Formation Control withGoal Assignment for Global Asymptotic Stability of Multi-Robot Systems [J].IEEE Robotics and Automation Letters,2021,6(2):2020-2027. [7]ZHANG Y J,ZHOU R,LI X,et al.Mean-Shift Shape Formation of Multi-Robot Systems Without Target Assignment [J].IEEE Robotics and Automation Letters,2024,9(2):1772-1779. [8]SLAVKOV I,CARRILLO-ZAPATA D,CARRANZA N,et al.Morphogenesis in robot swarms [J].Science Robotics,2018,3(25):eaau9178. [9]RUBENSTEIN M,CORNEJO A,NAGPAL R.Programmableself-assembly in a thousand-robot swarm [J].Science,2014,345(6198):795-799. [10]LIU X Y,GUO R H,REN C C,et al.Distributed Target Assignment Method for UAV Swarms Using Identity Hungarian Algorithm[J].Acta Armamentarii,2023,44(9):2824-2835. [11]HU C F,SONG S H,XU J J,et al.Distributed Task Allocation Based on Auction-PIO Algorithm for Multi-UAV Tracking[J].Journal of Tianjin University:Science and Technology,2024,57(4):403-414. [12]AVCI S,YILDIRIM M.Solving Weapon-Target AssignmentProblem with Salp Swarm Algorithm [J].Tehnicki Vjesnik,2023,30(1):17-23. [13]KONG X R,ZHOU Y T,LI Z,et al.Multi-UAV simultaneous target assignment and path planning based on deep reinforcement learning in dynamic multiple obstacles environments [J].Frontiers in Neurorobotics,2024,17:1302898. [14]MENG F Q,TIAN K S,WU C F.Deep Reinforcement Lear-ning-Based Radar Network Target Assignment [J].IEEE Sensors Journal,2021,21(14):16315-16327. [15]ALONSO-MORA J,BREITENMOSER A,RUFLI M,et al.Image and animation display with multiple mobile robots [J].International Journal of Robotics Research,2012,31(6):753-773. [16]WANG H L,RUBENSTEIN M,RUBENSTEIN M.Shape Formation in Homogeneous Swarms Using Local Task Swapping [J].IEEE Transactions on Robotics,2020,36(3):597-612. [17]SAKURAMA K,AHN H S.Multi-agent coordination over local indexes via clique-based distributed assignment [J].Automatica,2020,112:108670. [18]ALHAFNAWI M,HAUERT S,O'DOWD P.Self-OrganisedSaliency Detection and Representation in Robot Swarms [J].IEEE Robotics and Automation Letters,2021,6(2):1487-1494. [19]XU X,LI R P,ZHAO Z F,et al.Stigmergic Independent Reinforcement Learning for Multiagent Collaboration [J].IEEE Transactions on Neural Networks and Learning Systems,2022,33(9):4285-4299. [20]REZECK P,CHAIMOWICZ L.Chemistry-Inspired Pattern Formation with Robotic Swarms [J].IEEE Robotics and Automation Letters,2022,7(4):9137-9144. [21]BI Q,HUANG Y.A Self-organized Shape Formation Method For Swarm Controlling[C]//Proceedings of the 37th Chinese Control Conference.2018. |
|
||
首页