期刊
NEUROCOMPUTING
卷 346, 期 -, 页码 3-11出版社
ELSEVIER SCIENCE BV
DOI: 10.1016/j.neucom.2018.08.088
关键词
Neural networks; State estimation; Synchronization analysis; Communication protocols; Network-induced phenomena
资金
- National Natural Science Foundation of China [61573246, 61503001]
- Shanghai Rising-Star Program of China [16QA1403000]
- Natural Science Foundation of Shanghai [18ZR1427000]
State estimation and synchronization of neural networks (NNs) have recently received ever-increasing research interests due to a large number of successful applications in various fields such as repetitive learning, classification of patterns, nonlinear control, adaptive control, image processing, and so forth. Owing to limited communication bandwidth, the network-induced phenomena and the adopted communication scheduling may cause inevitable negative effects or the degradation of dynamical performance of NNs. This paper, from the perspective of dynamical behavior, renders a comprehensive summary on the recent advances of communication-embedded NNs and their application in nonlinear control. First, some common network-induced phenomena and communication protocols are roughly introduced from both models and mechanisms. Then, the state estimation and synchronization analysis of various NNs with network-induced phenomena are systematically reviewed and some interesting results are separately presented. Furthermore, the latest advances for the cases with communication protocols are profoundly surveyed from three aspects: state estimation, synchronization analysis and applications in control realm. Finally, some challenged problems are listed and several potential future research directions are highlighted. (C) 2019 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据