无人艇集群自组织协同围捕控制算法研究  

Research on collaborative self-organizing surrounding control algorithm of USV swarm

作　　者：夏桂华 朱文序 刘浩岩 刘兴 姜享利 XIA Guihua;ZHU Wenxu;LIU Haoyan;LIU Xing;JIANG Xiangli(College of Intelligent Systems Science and Engineering,Harbin Engineering University,Harbin 150006,China)

机构地区：[1]哈尔滨工程大学智能科学与工程学院,黑龙江哈尔滨150006

出　　处：《智能系统学报》2025年第1期162-171,共10页CAAI Transactions on Intelligent Systems

基　　金：国家自然科学基金项目(62403159);国家自然科学基金项目(52171302);中国科协青年人才托举工程项目(2022QNRC001);新时代龙江优秀博士学位论文项目(LJYXL2022-004).

摘　　要：无人艇(unmanned surface vessel,USV)集群协同围捕是无人艇的重要应用场景之一,文章针对无人艇集群协同围捕任务设计了一种自组织控制算法。考虑到被围对象的加速度未知且仅部分集群成员可对其测量的情况,文章设计了分布式目标状态观测器(distributed target state observer,DTSO),实现围捕过程中各成员对被捕对象状态的一致性观测。设计了一种自组织围捕引导律,解决了围捕算法需要为成员预分配期望位置的问题。设计了分布式非线性模型预测控制(distributed nonlinear model predictive control,DNMPC),解决了执行器饱和与艇间距离约束的问题,实现了满足避碰安全的协同围捕控制。文章分析了控制系统的输入-状态稳定性(input-to-state stability,ISS),并通过多组不同初始状态和改变成员数量的仿真实验,验证了围捕算法的有效性。Unmanned surface vehicle(USV)swarm cooperative surrounding is one of the important application scenarios of USVs.This paper designs a self-organizing control algorithm for USV swarm cooperative surrounding tasks.Considering the unknown acceleration of the target to be encircled and the fact that only some members of the swarm can measure its position and velocity,this paper proposes a distributed target state observer(DTSO)to achieve consistency observation of the target state among swarm members during the surrounding process.Subsequently,a self-organizing surrounding guidance law is designed to address the issue of pre-allocating expected positions for members in the surrounding algorithm.Finally,a distributed nonlinear model predictive control(DNMPC)is designed to resolve the problems of actuator saturation and inter-boat distance constraints,achieving collaborative surrounding control that meets collision avoidance safety.The paper analyzes the input-to-state stability(ISS)of the control system and validates the effectiveness of the surrounding algorithm through simulation experiments with multiple sets of different initial states and varying numbers of members.

关 键 词：无人艇 集群控制 协同围捕 多智能体 模型预测控制 输入-状态稳定性 状态观测器 图论 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]