Observer Design for Continuous-Discrete Time State Affine Systems up to Output Injection

Abstract

By continuous-discrete time affine systems, we mean continuous dynamical systems with discrete-time output measurements. The observability of this class of systems depends on the system inputs and on the output sampling time. In this context, we give sufficient conditions on both output sampling time and system input, which allow to preserve the observability of the system and consequently, guarantee the convergence of the continuous-discrete time observer synthesized in this paper.In order to validate the theoretical results, the performance of the proposed observer is illustrated through an example dealing with biomass growth and bio-synthesis reactions.

Observador de Alta Ganancia Constante para una Clase de Sistema No Lineal de Forma Triangular

“…Otro tipo de observador continuo-discreto para sistemas no lineales con inyección de salida se muestra en (Nadri et al, 2004).…”

“…Sin embargo, existen procesos en los que el periodo de muestreo de las mediciones es grande y generalmente diferente al tiempo de integración usado para las ecuaciones diferenciales del modelo dinámico; para este caso una alternativa de solución es el diseño de Observadores Continuo-Discretos (OCD) (Nadri et al, 2004), cuya principal ventaja radica en emplear un tiempo de integración (para la etapa de predicción) diferente del tiempo de muestreo (para la etapa de corrección) del observador. Esta condición permite emplear un tiempo relativamente amplio durante la adquisición de las variables medibles usadas para la estimación.…”

Dynamical continuous high gain observer for sampled measurements systems

“…Most of this kind of observers is based on the extended Kalman filter techniques. Unlike the discrete time approach, this one provides in most cases global and asymptotic stability [7], [8], [9], [10].…”

Observer-based monitoring of heat exchangers

“…They are used to infer unknown parameters or unmeasured state variables from on-line and/or off-line measurements; see e.g. [7][8][9]. Much of the work done to develop observers is based on Kalman filters or extended Kalman filters estimators (EKF) [10,11].…”