Details
| Titel in Übersetzung | Modelheuristics for efficient forward model learning |
|---|---|
| Originalsprache | Deutsch |
| Seiten (von - bis) | 848-857 |
| Seitenumfang | 10 |
| Fachzeitschrift | At-Automatisierungstechnik |
| Jahrgang | 69 |
| Ausgabenummer | 10 |
| Frühes Online-Datum | 26 Okt. 2021 |
| Publikationsstatus | Veröffentlicht - 26 Okt. 2021 |
| Extern publiziert | Ja |
Abstract
Forward model learning, i. e., learning forward models from data, finds application in prediction-based control. This involves observing inputs and outputs of the system to build a transition model and make predictions about future time steps. In particular, complex state spaces require the use of specialized search and model building techniques. In this work, we present abstraction heuristics for high-dimensional state spaces, which allow to reduce the model complexity and, in many cases, yield an interpretable result. In the context of two case studies, we demonstrate the effectiveness of the presented procedure in the context of artificial intelligence in games and motion control scenarios. The transfer of these methods enables promising applications in automation engineering.
Schlagwörter
- autonomous control, decomposition of forward models, forward model learning, local forward model, object-based forward model
ASJC Scopus Sachgebiete
- Ingenieurwesen (insg.)
- Steuerungs- und Systemtechnik
- Informatik (insg.)
- Angewandte Informatik
- Ingenieurwesen (insg.)
- Elektrotechnik und Elektronik
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
in: At-Automatisierungstechnik, Jahrgang 69, Nr. 10, 26.10.2021, S. 848-857.
Publikation: Beitrag in Fachzeitschrift › Artikel › Forschung › Peer-Review
}
TY - JOUR
T1 - Modellheuristiken für effizientes forward model learning
AU - Dockhorn, Alexander
AU - Kruse, Rudolf
PY - 2021/10/26
Y1 - 2021/10/26
N2 - Forward model learning, i. e., learning forward models from data, finds application in prediction-based control. This involves observing inputs and outputs of the system to build a transition model and make predictions about future time steps. In particular, complex state spaces require the use of specialized search and model building techniques. In this work, we present abstraction heuristics for high-dimensional state spaces, which allow to reduce the model complexity and, in many cases, yield an interpretable result. In the context of two case studies, we demonstrate the effectiveness of the presented procedure in the context of artificial intelligence in games and motion control scenarios. The transfer of these methods enables promising applications in automation engineering.
AB - Forward model learning, i. e., learning forward models from data, finds application in prediction-based control. This involves observing inputs and outputs of the system to build a transition model and make predictions about future time steps. In particular, complex state spaces require the use of specialized search and model building techniques. In this work, we present abstraction heuristics for high-dimensional state spaces, which allow to reduce the model complexity and, in many cases, yield an interpretable result. In the context of two case studies, we demonstrate the effectiveness of the presented procedure in the context of artificial intelligence in games and motion control scenarios. The transfer of these methods enables promising applications in automation engineering.
KW - autonomous control
KW - decomposition of forward models
KW - forward model learning
KW - local forward model
KW - object-based forward model
UR - http://www.scopus.com/inward/record.url?scp=85117839499&partnerID=8YFLogxK
U2 - 10.1515/auto-2021-0037
DO - 10.1515/auto-2021-0037
M3 - Artikel
AN - SCOPUS:85117839499
VL - 69
SP - 848
EP - 857
JO - At-Automatisierungstechnik
JF - At-Automatisierungstechnik
SN - 0178-2312
IS - 10
ER -