College of Automation, Nanjing University of Information Science and Technology, Nanjing, China.
PLoS One. 2024 Apr 18;19(4):e0299535. doi: 10.1371/journal.pone.0299535. eCollection 2024.
This paper focuses on studying the optimization problem of multi-agent systems (MAS) under undirected graph. To reduce the communication frequency among agents, a zero-gradient-sum (ZGS) algorithm based on dynamic event-triggered (DET) mechanism is investigated. The event-triggered condition of each agent only uses its own state information and the neighbor's state information at the previous triggering instants, without requiring continuous state information from the neighbor. In addition, the designed algorithm allows for the sampling period to be arbitrarily large. The Lyapunov method is utilized to derive the sufficient conditions that incorporate time delay and parameters. As the event is only checked at the periodic moment, zeno behavior can be directly excluded. Finally, numerical simulations demonstrate the effectiveness of the theoretical results.
本文主要研究无向图下多智能体系统（MAS）的优化问题。为了减少智能体之间的通信频率，研究了一种基于动态事件触发（DET）机制的零梯度和（ZGS）算法。每个智能体的事件触发条件仅使用其自身的状态信息和前一个触发时刻的邻居的状态信息，而不需要来自邻居的连续状态信息。此外，所设计的算法允许采样周期任意大。利用李雅普诺夫方法推导出包含时滞和参数的充分条件。由于事件仅在周期时刻进行检查，可以直接排除零阶行为。最后，数值仿真验证了理论结果的有效性。
