Learning event‐triggered control based on evolving data‐driven fuzzy granular models

This article proposes a data‐stream‐driven event‐triggered control strategy using evolving fuzzy models learned by granulation of input–output samples of nonlinear systems with unknown time‐varying dynamics. The evolving fuzzy model is obtained online from a data stream ensuring data coverage based on the principle of justifiable granularity and controlled by an event‐triggering learning mechanism dependent on the model accuracy. This evolving fuzzy model is used to design event‐triggered fuzzy controller to stabilize networked control systems while reducing the used communication resources. The event‐triggered learning mechanism is employed to determine the instants in which the event‐triggered fuzzy controller should be redesigned. Numerical examples illustrate the effectiveness of the proposed learning event‐triggered fuzzy control algorithm.