Event-triggered H∞ predictive control for networked control systems with sensor and actuator saturation

Abstract

This paper concerns an event-triggered H∞ predictive control scheme for a networked control system in which sensor and actuator saturation as well as external disturbances are involved. In the scheme, a state estimator is firstly designed to acquire state estimations from the saturated output with measurement noises. An event generator is given to pick out necessary state estimations for saving feedback channel transmission resources. Furthermore, a buffer-based predictive controller is constructed to compensate for network-induced delays and lighten the communication burden of forward channel networks. Then a co-design criterion is shown to find estimator and controller parameters for the resulting closed-loop system with desired H∞ performance. An inverted pendulum model is employed in numerical simulation to verify the proposed control scheme.

Keywords

• Networked control system
• Predictive control
• Sensor and actuator saturation
• Event-triggered mechanism
• H∞ performance