Observer‐based event‐triggered control co‐design for linear systems

Abstract

International audienceThe paper deals with an observer-based event-triggered control strategy for linear systems using only local (that is available) variables. Sufficient conditions based on linear matrix inequalities (LMI) associated to convex optimization problems are proposed to ensure the asymptotic stability of the closed loop and the output convergence to a constant reference in both emulation and co-design contexts. Indeed, the proposed approach allows either to design the event-triggering rules or co-design the event-triggering rule along with the controller gain

A Nonsmooth Hybrid Invariance Principle Applied to Robust Event-Triggered Design

“…where the parameter of the controller has been obtained using an optimization process provided in [1] A c := 1.0919 −1. x c , the whole dynamics described by (24) and (25) can be reformulated as system (8) with…”

“…Hence, it is interesting to address the design of event-triggered controllers by using only measured signals. Some works have addressed this challenge as, for example, [24] in which the dynamic controller is an observerbased one, [1] in which the co-design of the output feedback law and the event-triggering conditions is addressed by using the hybrid framework.…”

Adaptive Event Triggering Normed Observer Control of Nonlinear Time Delayed Systems

“…Different control techniques have been applied with event-triggering control strategy, namely in [8][9][10][11]. In [8], the authors have considered an event-triggered controller together with delay conditions in a networked system.…”

“…Again in the case of linear systems in [11], an observer-based approach for event triggering control of the system is studied.…”

Observer-based event-triggered control for discrete-time systems in the presence of sector-bounded nonlinear inputs

“…The event-trigger is a control strategy that updates the control signal only when necessary, based on a trigger criterion. Many trigger criteria have been proposed (Tabuada, 2007), but, in practical situations, it is necessary to use only the available information (measured or local signals) to define the trigger criterion (Groff et al, 2016a), (Tarbouriech et al, 2016). Chosen the trigger criterion, it is possible to design it and the control law gain simultaneously (co-design) (Abdelrahim et al, 2014), (Groff et al, 2016a) or just the event-triggering rule, with the control law previously computed (emulation design) (Wang and Lemmon, 2008).…”