A guided residual search for nonlinear state-space identification*
Merijn Floren, Jan Swevers · 2026
Read A guided residual search for nonlinear state-space identification* by Merijn Floren, Jan Swevers (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-22964.txt.
Description
Merijn Floren and one co-author make nonlinear state-space identification more reliable. Fitting such models from input-output data usually means solving a highly non-convex optimization problem, prone to slow convergence and poor local solutions. Their method splits the problem into a sequence of tractable subproblems.
Starting from a linear baseline model, nonlinear residual dynamics are first estimated with a guided residual search, then refined through multiple-shooting optimization. Tests on two benchmarks match state-of-the-art black-box methods while converging better than naive initialization, giving practitioners a steadier route to good fits.
Catalog record
| CREATOR | Merijn Floren, Jan Swevers |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 35 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-22964