Research paper
Updating Robot Safety Representations Online from Natural Language Feedback
Vision-language interpretation converts user feedback and images into updated safety representations, then warm-starts a Hamilton–Jacobi controller. Simulation and hardware examples examine adaptation to contextual constraints. Incorrect language grounding and unrecognized hazards remain important limits of this interface.
Reviewed 2026-11-05 · Source and bibliography checked
- Original title
- Updating Robot Safety Representations Online from Natural Language Feedback
- Publisher
- arXiv
- Authors
- Santos, Leonardo, Li, Zirui, Peters, Lasse, Bansal, Somil, Bajcsy, Andrea
- DOI
- 10.48550/arxiv.2409.14580
- First archive posting
- 2024-09-22
- Source language
- en
- Source country
- US
- Edition
- v1
- Access
- Open archive metadata and manuscript
- Publication status
- Preprint
Related resources and engineering notes
Source overview
Vision-language interpretation converts user feedback and images into updated safety representations, then warm-starts a Hamilton–Jacobi controller. Simulation and hardware examples examine adaptation to contextual constraints. Incorrect language grounding and unrecognized hazards remain important limits of this interface.
Use and review
Original summary of the primary archive abstract and bibliographic record. Specialist technical and native-language review remain pending. The reported method does not establish a deployed safety function or achieved PL.
Sources & editions
Updating Robot Safety Representations Online from Natural Language Feedback ↗
arXiv · v1
arXiv abstract, authors, submission history and DOI links; first posting 2024-09-22, abstract reviewed in v1. Publisher dates are separate.