Event-triggered robust MPC of nonlinear cyber-physical systems against DoS attacks

Abstract

This paper proposes an event-triggered robust nonlinear model predictive control (NMPC) framework for cyber-physical systems (CPS) in the presence of denial-of-service (DoS) attacks and additive disturbances. In the framework, a new robustness constraint is introduced to the NMPC optimization problem in order to deal with additive disturbances, and a packet transmission strategy is designed for NMPC such that DoS attacks can be tackled. Then, an event-triggered mechanism, which accommodates DoS attacks occurring in the communication network, is proposed to reduce the communication cost for resource-constrained CPSs. Besides, we prove that the NMPC algorithm is recursively feasible and the closed-loop system is input-to-state practical stable under some sufficient conditions. Simulation examples and comparisons are conducted to show the effectiveness of the proposed NMPC algorithm.

Dual‐mode resilient model predictive control for cyber‐physical systems against DoS attacks

“…To date, several scholars have concentrated on the development of resilient control strategies based on the MPC paradigm. This research area, closely related to the present paper, is further explored in References 13–20. To be more specific, several existing studies have delved into the resilient control problem through event‐triggered MPC approaches for various systems.…”

“…then (i) the satisfaction of ( 12) is ensured at sampling time k; (ii) after the system is regulated for 𝛼 steps in open-loop mode, constraint (16) is still feasible at time k…”

“…Lemma 2 further modifies the conditions in Lemma 1, providing constraints that not only ensure (12) holds, but also ensure the feasibility of constraints (16) when the system is regulated in an open-loop mode. Constraints (17) provided by Lemma 3 is equivalent to constraints (16) in Lemma 2, and the number of constraints ( 17) is finite.…”

“…To be more specific, several existing studies have delved into the resilient control problem through event‐triggered MPC approaches for various systems. For instance, Reference 13 focuses on adaptive cruise control systems in the context of sensor attacks, Reference 14 addresses vehicle platoon systems and their resilience to DoS attacks, while References 15 and 16 explore the resilient control of linear/nonlinear system models under deception/DoS attacks. The resilient MPC algorithms designed above can help conserve communication resources as the event‐triggered MPC scheme effectively reduces data transmission.…”

Resilient predictive control strategy of cyber–physical systems against FDI attack

“…Note that the main advantage of the resilient MPC over other resilient controllers is that the MPC based method could explicitly handle the input/output constraints of the considered system, and thus the physical limitations and/or safety operation ranges of the system variables could be automatically considered to further improve the performance of the resilient controller. Furthermore, compared with the existing results on MPC under attack [3, 28–30], the proposed resilient STMPC could significantly reduce the consumption of the computation and communication resources while still ensuring the system's performance under FDI attack. More specifically, the utilised self‐triggered mechanism makes the controller only solve the optimal control problem at necessary moments, and thus effectively reduces the online computation burden of the resilient STMPC; meanwhile, the proposed sample‐and‐reconstruction mechanism of the input signal further reduces the number of control samples that need to be protected and transmitted through the network, which significantly saves the communication resources.…”