系统工程与电子技术 ›› 2024, Vol. 46 ›› Issue (12): 4165-4172.doi: 10.12305/j.issn.1001-506X.2024.12.25

• 制导、导航与控制 • 上一篇    

多智能体编队加权中心点固定时间分布式跟踪控制

孙谷昊, 蔡中泽, 曾庆双   

  1. 哈尔滨工业大学航天学院, 黑龙江 哈尔滨 150001
  • 收稿日期:2023-10-11 出版日期:2024-11-25 发布日期:2024-12-30
  • 通讯作者: 孙谷昊
  • 作者简介:孙谷昊 (1992—), 男, 博士研究生, 主要研究方向为非线性控制及其在多智能体系统中的应用
    蔡中泽 (1993—), 男, 博士研究生, 主要研究方向为滑模控制理论方法及其在电路系统中的应用
    曾庆双 (1963—), 男, 教授, 博士, 主要研究方向为高精密运动控制技术、惯性技术、智能无人系统

Fixed-time distributed tracking and control of multi-agent formation weighted centroid

Guhao SUN, Zhongze CAI, Qingshuang ZENG   

  1. School of Astronautics, Harbin Institute of Technology, Harbin 150001, China
  • Received:2023-10-11 Online:2024-11-25 Published:2024-12-30
  • Contact: Guhao SUN

摘要:

针对多智能体编队加权中心点的目标跟踪问题, 提出一种基于滑模的固定时间收敛控制方法。每个智能体仅利用与邻居之间有限的信息交互, 通过各自的估计器实现对整个队形的加权中心点位置的快速、准确估计。同时, 提出一种基于状态依赖的可变指数系数滑模函数控制方法, 利用多智能体系统之间的相对位置作为队形约束, 解决队形的歧义问题, 并且所设计的滑模函数不存在奇异问题。估计器和控制器均能够保证在固定时间内收敛, 因此所提控制方法能够保证编队的加权中心点在固定时间内跟踪给定的参考轨迹, 且收敛时间与智能体的初始状态无关。最后, 通过仿真验证了所提方法的有效性。

关键词: 多智能体系统, 编队控制, 固定时间收敛, 中心点估计, 滑模控制, 分布式控制

Abstract:

A sliding mode-based fixed time convergence control method is proposed for the target tracking problem of weighted centroid in multi-agent formation. Each agent only utilizes limited mutual information with its neighbors and achieves a fast and accurate estimation of the weighted centroid position of the entire formation through its own estimator. At the same time, a state dependent-based variable exponential coefficient sliding mode function control method is proposed, which utilizes the relative positions between multi-agent systems as formation constraints to solve the ambiguity problem of formation, and the designed sliding mode function does not have singular problems. Both the estimator and controller can ensure convergence within a fixed time, so the proposed control method can ensure that the weighted centroid of the formation can track the given reference trajectory within a fixed time, and the convergence time is independent of the initial state of the agent. Finally, the effectiveness of the proposed method is verified through simulation test.

Key words: multi-agent system, formation control, fixed-time convergence, centroid estimation, sliding-mode control, distributed control

中图分类号: