Abstract
In this note, we propose a robust recursive least squares approach for the adaptive estimation of parameters in dynamic regressor systems. The approach provides linear dynamics of the covariance matrix. It is shown that, by choosing the tuning gains of the parameter update and the covariance updates, exponential convergence of the parameters is achieved while the covariance matrix vanishes to the origin at a slower exponential rate. The analysis also demonstrates that exponential convergence of the parameter estimates is preserved in the presence of bounded measurement error. A simulation study is presented to highlight the properties of the proposed technique.
| Original language | English |
|---|---|
| Title of host publication | 2026 European Control Conference (ECC): Proceedings |
| Publisher | European Control Association (EUCA) |
| Pages | 3783-3788 |
| Number of pages | 6 |
| ISBN (Electronic) | 9783907144138 |
| Publication status | Published - Jul 2026 |
| Event | 2026 European Control Conference - Reykjavík, Iceland Duration: 7 Jul 2026 → 10 Jul 2026 |
Conference
| Conference | 2026 European Control Conference |
|---|---|
| Abbreviated title | ECC 2026 |
| Country/Territory | Iceland |
| City | Reykjavík |
| Period | 7/07/26 → 10/07/26 |
Fingerprint
Dive into the research topics of 'A robust recursive least squares adaptive estimation approach'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver