3维未知复杂环境下无人机群协同围捕  

Cooperative Hunting by UAV Swarm in 3D Unknown Complex Environment

作　　者：张红强[1] 蒋萍 吴亮红[1] 王汐[1] 左词立 陈祖国 陈亮 ZHANG Hongqiang;JIANG Ping;WU Lianghong;WANG Xi;ZUO Cili;CHEN Zuguo;CHEN Liang(College of Information and Electrical Engineering,Hunan University of Science and Technology,Xiangtan 411201,China)

机构地区：[1]湖南科技大学信息与电气工程学院,湖南湘潭411201

出　　处：《信息与控制》2024年第6期723-738,共16页Information and Control

基　　金：国家重点研发计划(2023YFC30111000);国家自然科学基金项目(52104192,62271199);湖南省教育厅重点项目(23A0382);湖南省教育厅优秀青年项目(22B0476,21B0456);湖南省自然科学基金项目(2022JJ30265);国防基础科研计划项目(JCKY2019403D006);湖南省科技托举工程青年英才项目(2022TJ-Q03)

摘　　要：针对3维未知复杂环境下无人机群协同围捕问题,提出了一种基于膨胀算法和3维简化虚拟受力围捕模型相结合的围捕算法(3D-ESVFH)。首先构建无人机个体运动模型,针对3维未知复杂环境中的非凸障碍物,采用半球和球形膨胀相结合的膨胀算法,并对自然界中围捕行为进行分解,抽象出3维简化虚拟受力围捕模型,接着采用李雅普诺夫函数对系统进行稳定性分析。不同情况下的仿真结果表明,所提的围捕方法可以使无人机群避障的同时保持良好的围捕队形。将膨胀算法和3维简化虚拟受力围捕模型分别与循障算法、松散偏好规则方法进行对比分析,本文方法在时间消耗、路径消耗上分别减少了12.72%、 9.79%和20.05%、 8.35%。To address the challenge of cooperative hunting by unmanned aerial vehiclel(UAV)swarms in 3D unknown complex environments,we present a hunting algorithm that combines an expansion algorithm with a 3D simplified virtual force hunting model(3D⁃ESVFH).First,we develop a UAV individual motion model.For navigating non⁃convex obstacles in 3D unknown complex envi⁃ronments,we employ an expansion algorithm that combines hemispherical and spherical expan⁃sion.This process abstracts natural hunting behavior into a 3D simplified virtual force hunting model.We then use the Lyapunov function to analyze system stability.Simulation results under different conditions show that our method maintains effective hunting formations while avoiding ob⁃stacles.Compared to the following obstacle algorithm and the loose⁃preference rule,the proposed method reduces time and path consumption by 12.72%,9.79%,and 20.05%,8.35%,respec⁃tively.

关 键 词：无人机群 3维未知复杂环境 膨胀算法 3维简化虚拟受力围捕模型 

分 类 号：V249[航空宇航科学与技术—飞行器设计] V279