Asynchronous Dynamic Output Feedback Control for Discrete Nonlinear Networked Semi-Markov Jump Models With Cyber Attacks and Applications

Abstract

In this work, the asynchronous dynamic output feedback control is investigated for discrete nonlinear networked semi-Markov jump models with cyber attacks, in which asynchronous phenomenon refers to the mode mismatch between the controller and the system. For the potential uncertainty of system parameters, the interval type-2 fuzzy method is adopted to characterize nonlinear semi-Markov jump models. In light of actual state information unavailable in complex environment, the dynamic output feedback technique is proposed. The main novelty is to construct an appropriate dynamic output feedback control scheme, fully consider the asynchronous phenomenon of the controller mode, and introduce auxiliary variables to solve the matrix dimensional problem, so that the IT2 fuzzy network semi-Markov jump models under the influence of cyber attacks have better dynamic performance. According to stochastic theory, semi-Markov kernel, interval type-2 fuzzy method, and dwell-time-dependent Lyapunov function, sufficient criteria are established to ensure that the networked system is $\vartheta $ -error mean-square stable under random denial-of-service attacks. Furthermore, numerically checkable conditions are formulated to solve the asynchronous dynamic output feedback controller gain. Finally, a two-degree-freedom quarter-car suspension model is given to explain the superiorities of the proposed design approach. Note to Practitioners—As one of the research hotspots, networked control systems show outstanding advantages of low maintenance cost, easy installation, and high flexibility. However, cyber attacks often occur in networked control systems, posing a great threat to signal transmission. With the development of modern science and technology, semi-Markov jump models, owing to their excellent engineering background in modeling complex system, have a wide range of application prospects in solar receiver control, power electronics, chemical processes, and network communication. Note that some factors in the practical dynamical systems, such as parameter uncertainty and dwell information, cannot be completely obtained, and actual controller mode has switching delay with the system mode. In this paper, the asynchronous output feedback control is studied for discrete nonlinear networked semi-Markov jump models with cyber attacks. On the basis of interval type-2 fuzzy and dwell-time-dependent Lyapunov function, the $\vartheta $ -error mean-square stability is realized under random denial-of-service attacks. This research provides a new approach to study output feedback control strategies for discrete networked systems.