估计员
人工神经网络
网络数据包
计算机科学
国家(计算机科学)
控制理论(社会学)
理论(学习稳定性)
事件(粒子物理)
地铁列车时刻表
传输(电信)
数学
李雅普诺夫函数
数学优化
算法
控制(管理)
人工智能
统计
机器学习
操作系统
物理
电信
非线性系统
量子力学
计算机网络
作者
Juanjuan Yang,Lifeng Ma,Yonggang Chen,Xiaojian Yi
标识
DOI:10.1080/00207721.2022.2055192
摘要
In this paper, we study the memory event-triggering L2-L∞ state estimation for a type of continuous stochastic neural networks (NNs) subject to time-varying delays. The information of some recent released packets is made use in the proposed triggering conditions to schedule the data propagation, thereby reducing communication frequency and saving energy. By taking into account network-induced complexities (i.e. transmission delays and random disturbances), we first formulate the evolutions of estimation error in an augmented form, and then propose the conditions under with the design goals could be met. By using certain novel Lyapunov–Krasovskii (L–K) functionals in combination with stochastic analysis technique, sufficient conditions have been provided for the existence of desired estimator, guaranteeing both the globally asymptotically mean-square stability and the prescribed L2-L∞ performance simultaneously. Moreover, the estimator gains are obtained by virtue of certain convex optimisation algorithms. Finally, we use an illustrative example to verify the obtained theoretical algorithm.
科研通智能强力驱动
Strongly Powered by AbleSci AI