Event-Triggered Predictive Control for Networked Systems Based on a Dual-Rate Sampling Switched Observer

This paper is concerned with predictive control of discrete-time networked control systems (NCSs) based on a dual-rate sampling switched observer, where the multiple output signals of the controlled plant are sampled using a dual-rate sampling mechanism. To reduce the effect of the bilateral network-induced uncertainties on NCSs, a novel predictive control protocol is proposed to compensate for the bilateral network-induced delays and packet dropouts. First, a dual-rate sampled-data-based switched observer with a fixed switching law is designed to estimate the state of the controlled plant, where the switching law is determined by the dual-rate sampled instants. Second, an estimated-state-based predictive controller is designed by synchronizing the control inputs used by the state predictor and the controlled plant, and the networked predictive control system is formulated as a threshold-error-dependent delay system model, where two associated but different event-triggered transmission schemes are introduced to determine whether the estimated states or the control packet should be transmitted or not. Sufficient conditions for co-designing the event-triggered matrix and the predictive controller gain are derived by using the Lyapunov-Krasovskii functional method. Finally, it is shown through a numerical example that the proposed method in this paper is effective for ensuring better system performance.