Limited-Information Event-Triggered Observer-Based Control of Nonlinear Systems

Abstract

In this note, we propose a framework for the event-based semiglobal practical stabilization of nonlinear continuous-time systems based on limited amount of information. Following an emulation approach, it is required the availability of a convergent state observer and of a global asymptotic stabilizer in continuous time, generally depending on the estimated state and on the measured output. We then introduce time sampling, quantizations on the input, observed state and output signals and a periodic event-triggered mechanism to update the control law only when necessary. The proposed methodology guarantees practical stabilization of the system with state convergence to an arbitrarily small neighborhood of the origin by using finite data and a reduced number of control updates, which is desirable in digital and networked contexts with limited bandwidth. Numerical simulations on an actuated inverted pendulum show the potential of the approach.