Data-driven second-order iterative sliding mode control for cyber–physical systems under prescribed performance and DoS attacks

Abstract

This work investigates the second-order iterative sliding mode control problem of cyber–physical systems under prescribed performance and denial of service (DoS) attacks. An equivalent model of the controlled system is derived utilizing the dynamic linearization methodology, solely relying on process data. Through the novel tangent-type error transformation function, the confined error is equivalently transformed into the unconfined error. On this basis, a new second-order sliding function is devised to ensure that the error converges to the prearranged asymmetric region from the traditional time axis to the iteration axis. Based upon historical iterative data, an attack compensation mechanism is constructed to eliminate the negative impacts of attacks on the sensor. Finally, the effectiveness of the presented approach is validated via two examples.