基于周期事件触发机制的分布式资源分配算法  

Periodic event-triggered mechanism based distributed resource allocation algorithm

作　　者：李志勇 谈世磊 LI Zhi-yong;TAN Shi-lei(School of Electrical Engineering,Southwest Jiaotong University,Chendu 611756,China;School of Computer Science and Technology,University of Science and Technology of China,Hefei 230026,China)

机构地区：[1]西南交通大学电气工程学院,成都611756 [2]中国科学技术大学计算机科学与技术学院,合肥230026

出　　处：《控制与决策》2025年第2期563-571,共9页Control and Decision

基　　金：中央高校基本科研业务费项目(2682024CX006);国家自然科学基金项目(62302475);合肥市自然科学基金项目(202328)。

摘　　要：针对有限通信带宽下的多智能体系统最优资源分配问题,研究一种周期事件触发的分布式优化算法.首先,在连续时间型分布式加权梯度算法上,通过一种新的李雅普诺夫函数综合出一种事件触发通信机制,其触发器只需以一个固定周期采样自身状态信息并评估触发条件来判断是否需要进行通信;然后,通过稳定性分析表明,所提出分布式优化算法以指数速率收敛至最优解,这种周期事件触发机制不仅自然地排除芝诺行为,而且不需要触发器进行实时的检测;最后,通过数值仿真验证了所提出分布式事件触发优化算法的有效性.This article studies a periodic event-triggered distributed optimization algorithm for the optimal resource allocation of multi-agent systems under limited communication bandwidth.Based on the continuous-time distributed weighted gradient algorithm,a new Lyapunov function is proposed to synthesize an event-triggered communication mechanism.In this mechanism,each agent only needs to sample its own state information and evaluate the event-triggered condition at fixed intervals to determine whether communication is required.The stability analysis shows that the distributed optimization algorithm converges to the optimal solution at an exponential rate.The periodic event-triggered mechanism not only naturally eliminates Zeno behavior but also does not require real-time detection triggers.Finally,numerical simulations verify the effectiveness of the proposed distributed event-triggered optimization algorithm.

关 键 词：多智能体系统 资源分配问题 分布式优化 周期事件触发通信 指数收敛 连续时间算法 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]