Updating Robot Safety Representations Online from Natural Language Feedback

Robots must operate safely when deployed in novel and human-centered environments, like homes. Current safe control approaches typically assume that the safety constraints are known a priori, and thus, the robot can pre-compute a corresponding safety controller. While this may make sense for some sa...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Santos, Leonardo, Li, Zirui, Peters, Lasse, Bansal, Somil, Bajcsy, Andrea
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Santos, Leonardo
Li, Zirui
Peters, Lasse
Bansal, Somil
Bajcsy, Andrea
description Robots must operate safely when deployed in novel and human-centered environments, like homes. Current safe control approaches typically assume that the safety constraints are known a priori, and thus, the robot can pre-compute a corresponding safety controller. While this may make sense for some safety constraints (e.g., avoiding collision with walls by analyzing a floor plan), other constraints are more complex (e.g., spills), inherently personal, context-dependent, and can only be identified at deployment time when the robot is interacting in a specific environment and with a specific person (e.g., fragile objects, expensive rugs). Here, language provides a flexible mechanism to communicate these evolving safety constraints to the robot. In this work, we use vision language models (VLMs) to interpret language feedback and the robot's image observations to continuously update the robot's representation of safety constraints. With these inferred constraints, we update a Hamilton-Jacobi reachability safety controller online via efficient warm-starting techniques. Through simulation and hardware experiments, we demonstrate the robot's ability to infer and respect language-based safety constraints with the proposed approach.
doi_str_mv 10.48550/arxiv.2409.14580
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2409_14580</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2409_14580</sourcerecordid><originalsourceid>FETCH-arxiv_primary_2409_145803</originalsourceid><addsrcrecordid>eNpjYJA0NNAzsTA1NdBPLKrILNMzMjGw1DM0MbUw4GTwDi1ISSzJzEtXCMpPyi9RCE5MSy2pVAhKLShKLU7NKwHK5ecVK_jn5WTmpSqkFeXnKvgllpQWJeYo-CTmpZcmpqcquKWmpiQlJmfzMLCmJeYUp_JCaW4GeTfXEGcPXbC18QVFmbmJRZXxIOvjwdYbE1YBAMyJOu8</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Updating Robot Safety Representations Online from Natural Language Feedback</title><source>arXiv.org</source><creator>Santos, Leonardo ; Li, Zirui ; Peters, Lasse ; Bansal, Somil ; Bajcsy, Andrea</creator><creatorcontrib>Santos, Leonardo ; Li, Zirui ; Peters, Lasse ; Bansal, Somil ; Bajcsy, Andrea</creatorcontrib><description>Robots must operate safely when deployed in novel and human-centered environments, like homes. Current safe control approaches typically assume that the safety constraints are known a priori, and thus, the robot can pre-compute a corresponding safety controller. While this may make sense for some safety constraints (e.g., avoiding collision with walls by analyzing a floor plan), other constraints are more complex (e.g., spills), inherently personal, context-dependent, and can only be identified at deployment time when the robot is interacting in a specific environment and with a specific person (e.g., fragile objects, expensive rugs). Here, language provides a flexible mechanism to communicate these evolving safety constraints to the robot. In this work, we use vision language models (VLMs) to interpret language feedback and the robot's image observations to continuously update the robot's representation of safety constraints. With these inferred constraints, we update a Hamilton-Jacobi reachability safety controller online via efficient warm-starting techniques. Through simulation and hardware experiments, we demonstrate the robot's ability to infer and respect language-based safety constraints with the proposed approach.</description><identifier>DOI: 10.48550/arxiv.2409.14580</identifier><language>eng</language><subject>Computer Science - Robotics</subject><creationdate>2024-09</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,781,886</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2409.14580$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2409.14580$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Santos, Leonardo</creatorcontrib><creatorcontrib>Li, Zirui</creatorcontrib><creatorcontrib>Peters, Lasse</creatorcontrib><creatorcontrib>Bansal, Somil</creatorcontrib><creatorcontrib>Bajcsy, Andrea</creatorcontrib><title>Updating Robot Safety Representations Online from Natural Language Feedback</title><description>Robots must operate safely when deployed in novel and human-centered environments, like homes. Current safe control approaches typically assume that the safety constraints are known a priori, and thus, the robot can pre-compute a corresponding safety controller. While this may make sense for some safety constraints (e.g., avoiding collision with walls by analyzing a floor plan), other constraints are more complex (e.g., spills), inherently personal, context-dependent, and can only be identified at deployment time when the robot is interacting in a specific environment and with a specific person (e.g., fragile objects, expensive rugs). Here, language provides a flexible mechanism to communicate these evolving safety constraints to the robot. In this work, we use vision language models (VLMs) to interpret language feedback and the robot's image observations to continuously update the robot's representation of safety constraints. With these inferred constraints, we update a Hamilton-Jacobi reachability safety controller online via efficient warm-starting techniques. Through simulation and hardware experiments, we demonstrate the robot's ability to infer and respect language-based safety constraints with the proposed approach.</description><subject>Computer Science - Robotics</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNpjYJA0NNAzsTA1NdBPLKrILNMzMjGw1DM0MbUw4GTwDi1ISSzJzEtXCMpPyi9RCE5MSy2pVAhKLShKLU7NKwHK5ecVK_jn5WTmpSqkFeXnKvgllpQWJeYo-CTmpZcmpqcquKWmpiQlJmfzMLCmJeYUp_JCaW4GeTfXEGcPXbC18QVFmbmJRZXxIOvjwdYbE1YBAMyJOu8</recordid><startdate>20240922</startdate><enddate>20240922</enddate><creator>Santos, Leonardo</creator><creator>Li, Zirui</creator><creator>Peters, Lasse</creator><creator>Bansal, Somil</creator><creator>Bajcsy, Andrea</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20240922</creationdate><title>Updating Robot Safety Representations Online from Natural Language Feedback</title><author>Santos, Leonardo ; Li, Zirui ; Peters, Lasse ; Bansal, Somil ; Bajcsy, Andrea</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-arxiv_primary_2409_145803</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computer Science - Robotics</topic><toplevel>online_resources</toplevel><creatorcontrib>Santos, Leonardo</creatorcontrib><creatorcontrib>Li, Zirui</creatorcontrib><creatorcontrib>Peters, Lasse</creatorcontrib><creatorcontrib>Bansal, Somil</creatorcontrib><creatorcontrib>Bajcsy, Andrea</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Santos, Leonardo</au><au>Li, Zirui</au><au>Peters, Lasse</au><au>Bansal, Somil</au><au>Bajcsy, Andrea</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Updating Robot Safety Representations Online from Natural Language Feedback</atitle><date>2024-09-22</date><risdate>2024</risdate><abstract>Robots must operate safely when deployed in novel and human-centered environments, like homes. Current safe control approaches typically assume that the safety constraints are known a priori, and thus, the robot can pre-compute a corresponding safety controller. While this may make sense for some safety constraints (e.g., avoiding collision with walls by analyzing a floor plan), other constraints are more complex (e.g., spills), inherently personal, context-dependent, and can only be identified at deployment time when the robot is interacting in a specific environment and with a specific person (e.g., fragile objects, expensive rugs). Here, language provides a flexible mechanism to communicate these evolving safety constraints to the robot. In this work, we use vision language models (VLMs) to interpret language feedback and the robot's image observations to continuously update the robot's representation of safety constraints. With these inferred constraints, we update a Hamilton-Jacobi reachability safety controller online via efficient warm-starting techniques. Through simulation and hardware experiments, we demonstrate the robot's ability to infer and respect language-based safety constraints with the proposed approach.</abstract><doi>10.48550/arxiv.2409.14580</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2409.14580
ispartof
issn
language eng
recordid cdi_arxiv_primary_2409_14580
source arXiv.org
subjects Computer Science - Robotics
title Updating Robot Safety Representations Online from Natural Language Feedback
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-18T14%3A23%3A00IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Updating%20Robot%20Safety%20Representations%20Online%20from%20Natural%20Language%20Feedback&rft.au=Santos,%20Leonardo&rft.date=2024-09-22&rft_id=info:doi/10.48550/arxiv.2409.14580&rft_dat=%3Carxiv_GOX%3E2409_14580%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true