Reduced-Attention Output Feedback Control of Linear Systems with Bounded Disturbances

One of the main challenges in Networked Control Systems is to make better use of shared communication and computing resources while preserving closed-loop stability and good performance of the controlled process. Herein, we present an output feedback control strategy that reduces the communication overhead of the networked control system by increasing the time between control and measurement samples. The control design is based on self-triggered control and ensures closed-loop stability and ultimate boundedness of the state vector despite the presence of bounded input and output disturbances. The current work assumes that not all state variables are available for measurement and employs a reconstruction technique to estimate the state vector. Since the sampling is time-varying, this paper also presents results on the preservation of reconstructability and observability of linear systems under aperiodic sampling. Two numerical examples are presented to illustrate the performance of the state estimation and reduced-attention control strategy.