Salvato in:
Dettagli Bibliografici
Autori principali: van Esch, Stijn, Bonassi, Fabio, Schön, Thomas B.
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2408.12266
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914920754315264
author van Esch, Stijn
Bonassi, Fabio
Schön, Thomas B.
author_facet van Esch, Stijn
Bonassi, Fabio
Schön, Thomas B.
contents In this report we investigate the use of the Tustin neural network architecture (Tustin-Net) for the identification of a physical rotary inverse pendulum. This physics-based architecture is of particular interest as it builds on the known relationship between velocities and positions. We here aim at discussing the advantages, limitations and performance of Tustin-Nets compared to first-principles grey-box models on a real physical apparatus, showing how, with a standard training procedure, the former can hardly achieve the same accuracy as the latter. To address this limitation, we present a training strategy based on transfer learning that yields Tustin-Nets that are competitive with the first-principles model, without requiring extensive knowledge of the setup as the latter.
format Preprint
id arxiv_https___arxiv_org_abs_2408_12266
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Accounts of using the Tustin-Net architecture on a rotary inverted pendulum
van Esch, Stijn
Bonassi, Fabio
Schön, Thomas B.
Systems and Control
Machine Learning
In this report we investigate the use of the Tustin neural network architecture (Tustin-Net) for the identification of a physical rotary inverse pendulum. This physics-based architecture is of particular interest as it builds on the known relationship between velocities and positions. We here aim at discussing the advantages, limitations and performance of Tustin-Nets compared to first-principles grey-box models on a real physical apparatus, showing how, with a standard training procedure, the former can hardly achieve the same accuracy as the latter. To address this limitation, we present a training strategy based on transfer learning that yields Tustin-Nets that are competitive with the first-principles model, without requiring extensive knowledge of the setup as the latter.
title Accounts of using the Tustin-Net architecture on a rotary inverted pendulum
topic Systems and Control
Machine Learning
url https://arxiv.org/abs/2408.12266