Model-Based Nonlinear Periodic Event-Triggered Control for Continuous-Time Systems with Sampled-Data Prediction

Abstract: In this paper, we present a model-based periodic event-triggered control mechanism for nonlinear continuous-time Networked Control Systems. A sampled-data prediction of the system behavior is used at the actuator to reduce the amount of required communication while maintaining a user-defined performance level. This prediction is based on a possibly inaccurate discretization of the nonlinear system dynamics and can be implemented on simple hardware. Nevertheless, guarantees for asymptotic stability and a user-defined performance level are given for the periodic event-triggered control (PETC) mechanism, whilst the reduction of the required amount of transmissions of state information depends on the quality of the prediction. We discuss furthermore how the prediction can be implemented. The performance of the proposed PETC mechanism is illustrated with a numerical example. This paper is the accepted version of [1], containing also the proofs of the main results.