基于单体视觉投影的集群控制策略  

Swarm Control Strategy Based on Single Visual Projection

作　　者：张胜 胡学峰 刘畅[2] 李娟[2] ZHANG Sheng;HU Xuefeng;LIU Chang;LI Juan(Anhui Military Industry Group Holding Co.,Ltd.,Hefei 230051,China;School of Mechatronical Engineering,Beijing Institute of Technology,Beijing 100081,China)

机构地区：[1]安徽军工集团控股有限公司,合肥230051 [2]北京理工大学机电学院,北京100081

出　　处：《无人系统技术》2022年第5期81-89,共9页Unmanned Systems Technology

基　　金：国家自然科学基金(62003043)。

摘　　要：与常见的无人系统集群需要进行预先规划和控制机制不同,生物集群采用分布式和反应式架构,具备完成复杂任务的强大智能能力。受生物集群系统启发,设想无人系统集群中的个体能够对所遇到的环境进行独立反应,并展现出集群协同行为。针对这一设想,提出基于单体视觉膜投影模型的集群控制方法,并对其进行理论分析和推导,实现了群体内自组织集群控制;通过观察和计算运动过程中群体的稳定性、一致性、协调性等相关指标,探究所提出自组织模型对不同类型任务的适应性。此外,针对群体遭遇外来攻击的情形,设计了合理的群体避障机制,使得集群能够实现紧急安全避障,从而提高集群面临突发情况时的应对能力。Unlike common unmanned system swarms that require pre-planning and control mechanisms,biological swarms adopt a distributed and reactive architecture and possess powerful intelligence to complete complex tasks.Inspired by biological swarm systems,this paper envisages that individuals in a swarm of unmanned systems can inde-pendently respond to the environments encountered and exhibit swarm cooperative behavior.Aiming at this idea,this paper proposes a swarm control method based on the projection model of the single visual projection model,and makes a theoretical analysis and derivation to realize the self-organized swarm control.The adaptability of the proposed self-organization model to different types of tasks is explored by observing and calculating the stability,consistency,coordination and other related indicators of the swarm during the movement.In addition,a reasonable swarm obstacle avoidance mechanism is designed for the situation that the swarm encounters external attacks,so that the swarm can achieve emergency safety obstacle avoidance,thereby improving the ability of the swarm to cope with emergencies.

关 键 词：群体智能 反应式 视觉投影 集群控制 集群避障 

分 类 号：V247.1[航空宇航科学与技术—飞行器设计]