*Enlace a Spanish version
Materials: [ Cód.: MuvsKalvsHinffilterENG.mlx ] [ PDF ]
This video is the conclusion of the case study of the design of an estimator of an
internal variable (observer) of a system with an uncertain model, which began in
the video [
Although formally the theory just pursues guarantees in the frequency domain, in a real application we will actually see signals in the time domain. Thus, this video discusses the performances when subject to step and sinusoidal inputs with random measurement noise. It is argued that, since hinfsyn or musyn do not optimize the variance of the output, it would be necessary to experimentally check or/and add high-frequency filtering to the limit templates to ensure that the variance when facing measurement noise is acceptable. The case in continuous time also has its peculiarities because white noise is an infinite variance signal... The objective of this video is not to go into those subtleties but at least approach how to craft a sensible simulation.
*Link to my [ whole collection] of videos in English. Link to larger [ Colección completa] in Spanish.