基于actor-critic算法的分数阶多自主体系统最优主-从一致性控制  

Optimal Leader-Following Consensus Control of Fractional-Order Multi-Agent Systems Based on the Actor-Critic Algorithm

作　　者：马丽新 刘晨 刘磊[1] MA Lixin;LIU Chen;LIU Lei(College of Science,Hohai University,Nanjing 211100,P.R.China)

机构地区：[1]河海大学理学院,南京211100

出　　处：《应用数学和力学》2022年第1期104-114,共11页Applied Mathematics and Mechanics

基　　金：国家自然科学基金(面上项目)(61773152);中央高校基本科研业务费(2019B19214)。

摘　　要：研究了分数阶多自主体系统的最优主-从一致性问题.在考虑控制器周期间歇的前提下,将分数阶微分的一阶近似逼近式、事件触发机制和强化学习中的actor-critic算法有机整合,设计了基于周期间歇事件触发策略的强化学习算法结构.最后,通过数值仿真实验证明了该算法的可行性和有效性.Aimed at the optimal leader-following consensus problem of fractional-order multi-agent systems, an reinforcement learning strategy was designed based on the intermittent event trigger. With the periodic intermittent strategy as the basic mechanism, the event trigger and the actor-critic algorithm in reinforcement learning were organically integrated. According to the 1 st-order approximation of the fractional differential, the reinforcement learning algorithm structure based on the periodic intermittent event trigger strategy was proposed. Finally, the feasibility and effectiveness of the algorithm was proved through numerical simulation experiments.

关 键 词：分数阶多自主体系统 actor-critic算法 最优主-从一致性 事件触发 间歇 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置] O232[自动化与计算机技术—控制科学与工程]