Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

Discrete-time differentiators: design and comparative analysis

Abstract : This work deals with the problem of online differentiation of noisy signals. In this context, several types of differentiators including linear, sliding-mode based, adaptive, Kalman, and ALIEN differentiators are studied through mathematical analysis and numerical experiments. To resolve the drawbacks of the exact differentiators, new implicit and semi-implicit discretization schemes are proposed in this work to suppress the digital chattering caused by the wrong time-discretization of set-valued functions as well as providing some useful properties, e.g., finite-time convergence, invariant sliding-surface, exactness. A complete comparative analysis is presented in the manuscript to investigate the behavior of the discrete-time differentiators in the presence of several types of noises, including white noise, sinusoidal noise, and bell-shaped noise. Many details such as quantization effect and realistic sampling times are taken into account to provide useful information based on practical conditions. Many comments are provided to help the engineers to tune the parameters of the differentiators.
Complete list of metadatas

Cited literature [96 references]  Display  Hide  Download

https://hal.inria.fr/hal-02960923
Contributor : Bernard Brogliato <>
Submitted on : Thursday, October 8, 2020 - 9:28:25 AM
Last modification on : Monday, October 12, 2020 - 11:16:02 AM

File

Report_open_loop.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02960923, version 1

Collections

Citation

Mohammad Rasool Mojallizadeh, Bernard Brogliato, Vincent Acary. Discrete-time differentiators: design and comparative analysis. 2020. ⟨hal-02960923⟩

Share

Metrics

Record views

35

Files downloads

116