摘要
复杂跳跃系统是一类由连续时间变量和离散事件共同驱动的动态系统模型,具有状态混杂性、参数时变性、扰动多样性及结构不确定性等特性。介绍复杂跳跃系统的建模,包括连续时间和离散时间的马尔科夫跳跃系统和半马尔科夫跳跃系统模型;基于完全已知转移速率/概率、部分已知转移速率/概率和一般转移速率/概率等情况,从稳定性分析、控制器与滤波器设计等方面综述连续时间和离散时间复杂跳跃系统的研究进展,总结采用隐马尔科夫跳跃模型处理复杂跳跃系统模态信息未知情况的控制与滤波等问题的研究。指出:如何处理复杂非线性跳跃系统的建模、分析与控制仍是重难点问题;如何建立合适的数学模型描述复杂跳跃系统可能遭遇的多种网络攻击情况,解决其安全控制和滤波问题也是复杂跳跃系统面临的挑战性问题;针对复杂跳跃系统基于学习的控制和滤波、基于学习的系统安全控制与安全估计等问题值得进一步研究。
Complex jump systems are a kind of dynamic system model driven by continuous-time variables and discrete events,which are with hybrid state features,parameter time variability,disturbance diversity and structural uncertainty.Firstly,the models of complex jump systems are elaborated,including the continuous-time and discrete-time Markov jump systems and semi-Markov jump systems.Then,the research progress of continuoustime and discrete-time complex jump systems is introduced from stability analysis,control and filtering problems,respectively,including the completely known transition rate/probability cases,the partially known transition rate/probability cases and the general transition rate/probability cases.Additionally,the research results of hidden Markov jump model for the control and filtering problems of complex jumping systems with unknown mode information are introduced.Finally,the main content of this article is summarized and pointed out that how to deal with the modeling,analysis and control of complex nonlinear jump systems is still a difficult problem,how to establish a suitable mathematical model to describe various cyber-attacks that a complex jump system may encounter and how to solve its security control and filtering problems are also challenge for the complex jump systems,and the learning-based control,filtering and learning-based safety control and safety estimation of the complex jump systems also deserve further study.
作者
苏磊
李峰
夏荣盛
汪婧
沈浩
SU Lei;LI Feng;XIA Rongsheng;WANG Jing;SHEN Hao(School of Electrical&Information Engineering,Anhui University of Technology,Maanshan 243032,China)
出处
《安徽工业大学学报（自然科学版）》
CAS
2022年第3期237-249,共13页
Journal of Anhui University of Technology（Natural Science）
基金
国家自然科学基金项目(62103005,62173001)。
关键词
复杂跳跃系统
系统模型
鲁棒控制
估计
滤波
complex jump systems
system model
robust control
estimation
filtering
