Gu, Shangding ; Kshirsagar, Alap ; Du, Yali ; Chen, Guang ; Peters, Jan ; Knoll, Alois (2023)
A human-centered safe robot reinforcement learning framework with interactive behaviors.
In: Frontiers in Neurorobotics, 17
doi: 10.3389/fnbot.2023.1280341
Artikel, Bibliographie
Dies ist die neueste Version dieses Eintrags.
Kurzbeschreibung (Abstract)
Deployment of Reinforcement Learning (RL) algorithms for robotics applications in the real world requires ensuring the safety of the robot and its environment. Safe Robot RL (SRRL) is a crucial step toward achieving human-robot coexistence. In this paper, we envision a human-centered SRRL framework consisting of three stages: safe exploration, safety value alignment, and safe collaboration. We examine the research gaps in these areas and propose to leverage interactive behaviors for SRRL. Interactive behaviors enable bi-directional information transfer between humans and robots, such as conversational robot ChatGPT. We argue that interactive behaviors need further attention from the SRRL community. We discuss four open challenges related to the robustness, efficiency, transparency, and adaptability of SRRL with interactive behaviors.
Typ des Eintrags: | Artikel |
---|---|
Erschienen: | 2023 |
Autor(en): | Gu, Shangding ; Kshirsagar, Alap ; Du, Yali ; Chen, Guang ; Peters, Jan ; Knoll, Alois |
Art des Eintrags: | Bibliographie |
Titel: | A human-centered safe robot reinforcement learning framework with interactive behaviors |
Sprache: | Englisch |
Publikationsjahr: | 9 November 2023 |
Ort: | Lausanne |
Verlag: | Frontiers Media S.A. |
Titel der Zeitschrift, Zeitung oder Schriftenreihe: | Frontiers in Neurorobotics |
Jahrgang/Volume einer Zeitschrift: | 17 |
Kollation: | Artikel-ID: 1280341 |
DOI: | 10.3389/fnbot.2023.1280341 |
Zugehörige Links: | |
Kurzbeschreibung (Abstract): | Deployment of Reinforcement Learning (RL) algorithms for robotics applications in the real world requires ensuring the safety of the robot and its environment. Safe Robot RL (SRRL) is a crucial step toward achieving human-robot coexistence. In this paper, we envision a human-centered SRRL framework consisting of three stages: safe exploration, safety value alignment, and safe collaboration. We examine the research gaps in these areas and propose to leverage interactive behaviors for SRRL. Interactive behaviors enable bi-directional information transfer between humans and robots, such as conversational robot ChatGPT. We argue that interactive behaviors need further attention from the SRRL community. We discuss four open challenges related to the robustness, efficiency, transparency, and adaptability of SRRL with interactive behaviors. |
Freie Schlagworte: | interactive behaviors, safe exploration, value alignment, safe collaboration, bi-direction information |
Zusätzliche Informationen: | This article is part of the Research Topic: Insights in Neurorobotics: 2023-2024 |
Sachgruppe der Dewey Dezimalklassifikatin (DDC): | 000 Allgemeines, Informatik, Informationswissenschaft > 004 Informatik 600 Technik, Medizin, angewandte Wissenschaften > 621.3 Elektrotechnik, Elektronik |
Fachbereich(e)/-gebiet(e): | 20 Fachbereich Informatik 20 Fachbereich Informatik > Intelligente Autonome Systeme |
Hinterlegungsdatum: | 17 Jun 2024 12:27 |
Letzte Änderung: | 17 Jun 2024 12:27 |
PPN: | |
Export: | |
Suche nach Titel in: | TUfind oder in Google |
Verfügbare Versionen dieses Eintrags
-
A human-centered safe robot reinforcement learning framework with interactive behaviors. (deposited 11 Jun 2024 11:39)
- A human-centered safe robot reinforcement learning framework with interactive behaviors. (deposited 17 Jun 2024 12:27) [Gegenwärtig angezeigt]
![]() |
Frage zum Eintrag |
Optionen (nur für Redakteure)
![]() |
Redaktionelle Details anzeigen |