Event-triggered model predictive control for switched systems: a memory-based anti-attack strategy
Yiwen Qi,Wenke Yu,Ziyang Liu,Ning Xing
DOI: https://doi.org/10.1080/00207721.2021.1925993
IF: 2.648
2021-05-13
International Journal of Systems Science
Abstract:This paper investigates H∞ event-triggered model predictive control (MPC) problem for switched systems. In network transmission processes, two problems (including limited resources and denial of service (DoS) attacks) profoundly affect the usability of control systems. Especially, DoS attacks that occur in feedback channel can make the systems open-loop. In this regard, a novel memory-based anti-attack strategy is presented, which is implemented by adding a cache function to network controllers. An event-triggering mechanism is adopted to improve network resource utilisation and reduce the probability of transmission data being attacked. MPC technology is employed for network controllers, which optimises the control input and state by minimising an upper bound of a quadratic cost function. Then, closed-loop switched system is established, and the coupling relationship between switching and attacking is clearly discussed. Based on average dwell time approach, H∞ performance is analysed. The co-design of MPC controllers and event-triggering matrices are given by solving linear matrix inequalities (LMIs). Finally, a numerical example is provided to demonstrate the feasibility.
automation & control systems,computer science, theory & methods,operations research & management science