SpotLight: Robotic Scene Understanding through Interaction and Affordance Detection

Tim Engelbracht, René Zurbrügg, Marc Pollefeys, Hermann Blum, Zuria Bauer
{"title":"SpotLight: Robotic Scene Understanding through Interaction and Affordance Detection","authors":"Tim Engelbracht, René Zurbrügg, Marc Pollefeys, Hermann Blum, Zuria Bauer","doi":"arxiv-2409.11870","DOIUrl":null,"url":null,"abstract":"Despite increasing research efforts on household robotics, robots intended\nfor deployment in domestic settings still struggle with more complex tasks such\nas interacting with functional elements like drawers or light switches, largely\ndue to limited task-specific understanding and interaction capabilities. These\ntasks require not only detection and pose estimation but also an understanding\nof the affordances these elements provide. To address these challenges and\nenhance robotic scene understanding, we introduce SpotLight: A comprehensive\nframework for robotic interaction with functional elements, specifically light\nswitches. Furthermore, this framework enables robots to improve their\nenvironmental understanding through interaction. Leveraging VLM-based\naffordance prediction to estimate motion primitives for light switch\ninteraction, we achieve up to 84% operation success in real world experiments.\nWe further introduce a specialized dataset containing 715 images as well as a\ncustom detection model for light switch detection. We demonstrate how the\nframework can facilitate robot learning through physical interaction by having\nthe robot explore the environment and discover previously unknown relationships\nin a scene graph representation. Lastly, we propose an extension to the\nframework to accommodate other functional interactions such as swing doors,\nshowcasing its flexibility. Videos and Code:\ntimengelbracht.github.io/SpotLight/","PeriodicalId":501031,"journal":{"name":"arXiv - CS - Robotics","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Robotics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.11870","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Despite increasing research efforts on household robotics, robots intended for deployment in domestic settings still struggle with more complex tasks such as interacting with functional elements like drawers or light switches, largely due to limited task-specific understanding and interaction capabilities. These tasks require not only detection and pose estimation but also an understanding of the affordances these elements provide. To address these challenges and enhance robotic scene understanding, we introduce SpotLight: A comprehensive framework for robotic interaction with functional elements, specifically light switches. Furthermore, this framework enables robots to improve their environmental understanding through interaction. Leveraging VLM-based affordance prediction to estimate motion primitives for light switch interaction, we achieve up to 84% operation success in real world experiments. We further introduce a specialized dataset containing 715 images as well as a custom detection model for light switch detection. We demonstrate how the framework can facilitate robot learning through physical interaction by having the robot explore the environment and discover previously unknown relationships in a scene graph representation. Lastly, we propose an extension to the framework to accommodate other functional interactions such as swing doors, showcasing its flexibility. Videos and Code: timengelbracht.github.io/SpotLight/
SpotLight:通过交互和亲和力检测理解机器人场景
尽管对家用机器人技术的研究日益增多,但用于家庭环境中的机器人仍然难以完成更复杂的任务,例如与抽屉或电灯开关等功能元件进行交互,这主要是由于对特定任务的理解和交互能力有限。这些任务不仅需要检测和姿势估计,还需要了解这些元素提供的能力。为了应对这些挑战并增强机器人对场景的理解,我们引入了 SpotLight:一个用于机器人与功能元素(特别是灯光开关)进行交互的综合框架。此外,该框架还能让机器人通过交互提高对环境的理解。我们进一步介绍了包含 715 幅图像的专用数据集,以及用于检测灯开关的自定义检测模型。我们展示了该框架如何通过物理交互促进机器人学习,让机器人探索环境并发现场景图表示中之前未知的关系。最后,我们提出了对该框架的扩展,以适应其他功能的交互,如旋转门,从而展示了该框架的灵活性。视频和代码:timengelbracht.github.io/SpotLight/
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信