Saved in:
| Main Author: | |
|---|---|
| Format: | Recurso digital |
| Language: | |
| Published: |
Zenodo
2026
|
| Online Access: | https://doi.org/10.5281/zenodo.19637903 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Table of Contents:
- <p><span>This paper proposes a novel robust matrix constraint modeling method that unifies the linear matrix inequality (LMI) structure with Box uncertainty sets, constructing a unified expression for describing the positive semidefiniteness of matrices under parameter perturbations. Traditional LMI methods primarily handle deterministic or affine parameter systems, while Box uncertainty sets can characterize the worst-case scenario where parameters vary independently within interval boundaries. By introducing "vertex matrix decomposition mapping" and "worst-case boundary projection operator," this paper derives an equivalent verifiable form of the robust linear matrix inequality (RBLMI), thus transforming the infinite constraint problem into a set of finite positive semidefinite matrix constraints. This method has potential applications in robust control, optimization theory, and stability analysis of machine learning models.</span></p>