考虑局部特征的无人机集群关键节点识别方法  

Critical Node Identification Method for Unmanned Aerial Vehicle Cluster Considering Localized Features

作　　者：石成泷 华翔 王东[1] 张金金[2] 蒋天启 党元章 Shi Chenglong;Hua Xiang;Wang Dong;Zhang Jinjin;Jiang Tianqi;Dang Yuanzhang(School of Mechatromic Engineering,Xi'an Technological University,Xi'an 710021,China;School of Armament Science and Technology,Xi'an Technological University,Xi'an 710021,China;School of Electronic Information Engineering,Xi'an Technological University,Xi'an 710021,China)

机构地区：[1]西安工业大学机电工程学院,陕西西安710021 [2]西安工业大学兵器科学与技术学院,陕西西安710021 [3]西安工业大学电子信息工程学院,陕西西安710021

出　　处：《系统仿真学报》2024年第12期2824-2833,共10页Journal of System Simulation

基　　金：陕西省重点研发计划(2023-YBGY-227);陕西省自然科学基础研究计划(2023-JC-QN-0705);西安市科技计划(2022JH-RYFW-0138);碑林区科技计划(GX2216)。

摘　　要：针对无人机集群关键节点识别方法注重网络全局,忽略节点与其局部特征之间关联性的问题,提出一种考虑局部特征的无人机集群关键节点识别方法。基于复杂网络理论构建无人机集群网络模型;引入拉普拉斯能量评估两跳范围内节点的重要程度,结合信息熵评估特定模体中节点的重要程度,以综合识别关键节点。结果表明:该方法识别的关键节点表现出较好的区分度,验证了其有效性,相比对比方法,在连续失效下抗毁性下降趋势显著,验证了其准确性。Aiming at the problem that the UAV cluster critical node identification methods focus on the global network and ignore the correlation between nodes and their local features,a critical nodes identification method for unmanned aerial vehicle cluster considering local features is proposed.An unmanned aerial vehicle cluster network model is constructed based on complex network theory.The Laplacian energy is introduced to evaluate the importance of node within two hops,and information entropy is combined to evaluate the importance of node in a specific motif to comprehensive identify the critical nodes.Simulation results demonstrate that this method identifies critical nodes that perform well in terms of differentiation,validating its effectiveness.This method exhibits a significant improvement in resilience against continuous failures compared to the comparative methods,further confirming its superiority.

关 键 词：无人机集群 关键节点 局部特征 拉普拉斯能量 模体 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]