Variable sampling thresholds for event reduction in event-based control

Event-based control pursues the objective of providing an appropriate process regulation while reducing the amount of information needed to that end. During the last decade, research has primarily focused on providing adequate tuning methods to avoid the induction of limit cycle oscillations, a phenomenon to avoid in this kind of control system. However, as important as the tuning rule, is the sampling strategy, which has remained almost intact. In this work, a new family of sampling strategies is proposed, whose thresholds vary according to arithmetic or geometric progressions. The validity, in terms of robustness, of tuning methods presented in literature for other sampling schemes is demonstrated to hold for the presented sampling strategies. In addition, a study regarding the event generation is conducted, focusing on the fundamental trade-off between events and performance of event-based control.