One-shot backpropagation for multi-step prediction in physics-based system identification -- EXTENDED VERSION
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2023
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908393710551040 |
|---|---|
| author | Donati, Cesare Mammarella, Martina Dabbene, Fabrizio Novara, Carlo Lagoa, Constantino |
| author_facet | Donati, Cesare Mammarella, Martina Dabbene, Fabrizio Novara, Carlo Lagoa, Constantino |
| contents | The aim of this paper is to present a novel physics-based framework for the identification of dynamical systems, in which the physical and structural insights are reflected directly into a backpropagation-based learning algorithm. The main result is a method to compute in closed form the gradient of a multi-step loss function, while enforcing physical properties and constraints. The derived algorithm has been exploited to identify the unknown inertia matrix of a space debris, and the results show the reliability of the method in capturing the physical adherence of the estimated parameters. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2310_20567 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | One-shot backpropagation for multi-step prediction in physics-based system identification -- EXTENDED VERSION Donati, Cesare Mammarella, Martina Dabbene, Fabrizio Novara, Carlo Lagoa, Constantino Systems and Control Machine Learning The aim of this paper is to present a novel physics-based framework for the identification of dynamical systems, in which the physical and structural insights are reflected directly into a backpropagation-based learning algorithm. The main result is a method to compute in closed form the gradient of a multi-step loss function, while enforcing physical properties and constraints. The derived algorithm has been exploited to identify the unknown inertia matrix of a space debris, and the results show the reliability of the method in capturing the physical adherence of the estimated parameters. |
| title | One-shot backpropagation for multi-step prediction in physics-based system identification -- EXTENDED VERSION |
| topic | Systems and Control Machine Learning |
| url | https://arxiv.org/abs/2310.20567 |