Adaptive Sliding Mode Security Control for Stochastic Markov Jump Cyber-Physical Nonlinear Systems Subject to Actuator Failures and Randomly Occurring Injection Attacks

Abstract

This article investigates the issue of security control for stochastic Markov jump cyber-physical systems (SMJCPS) against actuator failures (AF), randomly occurring injection attacks (ROIA), and inaccessible states by virtue of state estimator-based adaptive sliding mode control (SMC) strategy. The knowledge of the states is generated with an estimator not requesting any input information from which a novel switching surface of linear type (SSL) is established. Then, an adaptive SMC input is developed to ensure the attainability of the SSL in limited steps, almost surely under stochastic noise, unknown ROIA, and potential AF. In the light of the arrival of the SSL and stochastic stability theory, a new stochastically stable criterion for the target SMJCPS operating on the defined SSL is deducted in the occurrence of AF, ROIA, and more generally uncertain transition rates. At last, a simulation study is performed, in which the raised control scheme is realized and certified by a tunnel diode circuit model.