作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2024, Vol. 50 ›› Issue (3): 78-88. doi: 10.19678/j.issn.1000-3428.0067683

• 人工智能与模式识别 • 上一篇    下一篇

具有规定性能的多智能体动态事件触发编队控制

张建强*(), 杨凯军, 欧阳凌丛   

  1. 陕西科技大学电气与控制工程学院, 陕西 西安 710021
  • 收稿日期:2023-05-23 出版日期:2024-03-15 发布日期:2023-07-28
  • 通讯作者: 张建强
  • 基金资助:
    国家自然科学基金(62003201)

Dynamical Event-Triggered Formation Control of Multi-Agent with Prescribed Performance

Jianqiang ZHANG*(), Kaijun YANG, Lingcong OUYANG   

  1. School of Electrical and Control Engineering, Shaanxi University of Science and Technology, Xi'an 710021, Shaanxi, China
  • Received:2023-05-23 Online:2024-03-15 Published:2023-07-28
  • Contact: Jianqiang ZHANG

摘要:

针对多智能体系统编队控制过程中面临的性能约束、输入饱和、通信资源受限等问题,研究一类具有规定性能的高阶非仿射多智能体系统编队控制问题,提出一种有限时间动态事件触发编队控制策略。首先利用反步法设计控制律,并引入微分跟踪技术解决“计算爆炸”问题,有效避免对复杂虚拟控制律的求导过程;其次采用模糊逻辑系统估计系统内部的不确定性,通过设计性能函数和饱和补偿系统使得多智能体满足规定瞬态性能并防止输入饱和所带来的影响;随后为了解决通信资源受限的问题,构造一个具有动态阈值的自适应事件触发协议,有效减少控制器与执行器之间的通信量;最后利用李雅普诺夫稳定性理论分析闭环系统的稳定性,保证编队误差将渐近收敛至0且系统中所有信号均有界。采用含有1个领导者和5个跟随者的多智能体模型进行仿真验证,结果表明,在该有限时间动态事件触发编队控制策略下,多智能体系统最终形成以领导者为圆心、跟随者都在圆上的编队队形,且编队误差在规定的范围内演化。

关键词: 多智能体系统, 非线性系统, 规定性能, 输入饱和, 动态事件触发控制

Abstract:

In response to the challenges regarding performance constraints, input saturation, and limited communication resources in the formation control process of multi-agent systems, this study investigates a class of high-order nonaffine multi-agent formation control problems with prescribed performance. It proposes a finite-time dynamic event-triggered formation control strategy. First, the control law is designed using the backstepping method, and the differential tracking technique is introduced to address the issue of "computational explosion", effectively avoiding the need for complex virtual control law differentiation. Secondly, a fuzzy logic system is employed to estimate the internal uncertainties of the system. By designing a performance function and a saturation compensation system, the multi-agent system satisfies the prescribed transient performance and mitigates the effects of input saturation. To address limited communication resources, we construct an adaptive event-triggered protocol with a dynamic threshold. This effectively reduces the communication load between the controller and the actuators. Finally, the stability of the closed-loop system is analyzed using the Lyapunov stability theory to ensure the asymptotic convergence of formation errors to zero and the boundedness of all signals in the system. Simulation experiments are conducted with a multi-agent model with one leader and five followers. The results show that, with the finite-time dynamic event-triggering formation control strategy, the multi-agent system ultimately achieves a formation with the leader at the center and the followers arranged in a circle. Moreover, the formation error remains within the prescribed range.

Key words: multi-agent system, nonlinear system, prescribed performance, input saturation, dynamic event-triggered control