In many applications, it is important to be able to estimate online some number of derivatives of a given (differentiable) signal. Some famous algorithms solving the problem comprise linear high-gain observers and Levant’s exact differentiators, that is discontinuous. They are both homogeneous, as are many other ones. A disadvantage of continuous algorithms is that they are able to calculate exactly the derivatives only for a very small class of (polynomial) time signals. The discontinuous Levant’s differentiator, in contrast, can calculate in finite-time and exactly the derivatives of Lipschitz signals, which is a much larger class. However, it has the drawback that its convergence time increases very strongly with the size of the initial conditions. Thus, a combination of both algorithms seems advantageous, and this has been proposed recently by the author in [38]. In this work, some techniques used to design differentiators are reviewed and it is shown how the combination of two different homogeneous algorithms can be realized and that it leads to interesting properties. A novelty is the derivation of a very simple realization of the family of bi-homogeneous differentiatorsBi-homogeneous differentiator proposed in [38]. The methodological framework is based on the use of smooth Lyapunov functions to carry out their performance and convergence analysis.
Bi-homogeneous Differentiators
Studies in Systems, Decision and Control
01.11.2023
26 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
New Homogeneous Controllers and Differentiators
Springer Verlag | 2020
|Discrete-Time Implementations of Differentiators Homogeneous in the Bi-Limit
Springer Verlag | 2023
|Technology Report - ELECTRONICS - User interfaces becoming big differentiators
Online Contents | 2013
Silicon-Based Integrated Tunable Fractional Order Photonic Temporal Differentiators
British Library Online Contents | 2017
|Subpixel phase-based image registration using Savitzky–Golay differentiators in gradient-correlation
British Library Online Contents | 2018
|