Anisotropy-Based Controller Design for Linear Discrete-Time Systems with Multiplicative Noise
- Authors: Yurchenkov A.V.1,2
-
Affiliations:
- Bauman Moscow State Technical University
- Trapeznikov Institute of Control Sciences, Russian Academy of Sciences
- Issue: Vol 57, No 6 (2018)
- Pages: 864-873
- Section: Control in Stochastic Systems and Under Uncertainty Conditions
- URL: https://journals.rcsi.science/1064-2307/article/view/220238
- DOI: https://doi.org/10.1134/S1064230718060138
- ID: 220238
Cite item
Abstract
This paper extends anisotropy-based control theory methods to the class of linear discrete-time systems with multiplicative noise, which involves an informational criterion to describe the stochastic uncertainty of an external disturbance. An approximate calculation method for the anisotropic norm of such systems using decomposition into linear subsystems without multiplicative noise is suggested. The cases of perfect and imperfect state measurements are considered. The existence criterion for anisotropy-based controllers in form of special matrix inequalities is established.
About the authors
A. V. Yurchenkov
Bauman Moscow State Technical University; Trapeznikov Institute of Control Sciences, Russian Academy of Sciences
Author for correspondence.
Email: alexander.yurchenkov@yandex.ru
Russian Federation, Moscow; Moscow