KIT | KIT-Bibliothek | Impressum | Datenschutz

Hidden Parameter Recurrent State Space Models For Changing Dynamics Scenarios

Shaj Kumar, Vaisakh 1; Büchler, D.; Sonker, R.; Becker, Philipp 1; Neumann, Gerhard 1
1 Institut für Anthropomatik und Robotik (IAR), Karlsruher Institut für Technologie (KIT)

Abstract (englisch):

Recurrent State-space models (RSSMs) are highly expressive models for learning patterns in time series data and system identification. However, these models assume that the dynamics are fixed and unchanging, which is rarely the case in real-world scenarios. Many control applications often exhibit tasks with similar but not identical dynamics which can be modeled as a latent variable. We introduce the Hidden Parameter Recurrent State Space Models (HiP-RSSMs), a framework that parametrizes a family of related dynamical systems with a low-dimensional set of latent factors. We present a simple and effective way of learning and performing inference over this Gaussian graphical model that avoids approximations like variational inference. We show that HiP-RSSMs outperforms RSSMs and competing multi-task models on several challenging robotic benchmarks both on real-world systems and simulations.


Volltext §
DOI: 10.5445/IR/1000143406
Veröffentlicht am 31.08.2022
Cover der Publikation
Zugehörige Institution(en) am KIT Institut für Anthropomatik und Robotik (IAR)
Publikationstyp Forschungsbericht/Preprint
Publikationsdatum 20.01.2022
Sprache Englisch
Identifikator KITopen-ID: 1000143406
Vorab online veröffentlicht am 05.10.2021
Nachgewiesen in arXiv
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page