Recovering from failure by asking for help
Robots inevitably fail, often without the ability to recover autonomously. We demonstrate an approach for enabling a robot to recover from failures by communicating its need for specific help to a human partner using natural language. Our approach automatically detects failures, then generates targe...
| Main Authors: | , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Springer US
2016
|
| Online Access: | http://hdl.handle.net/1721.1/105525 https://orcid.org/0000-0002-8293-0492 https://orcid.org/0000-0001-5473-3566 |
| _version_ | 1811091653900894208 |
|---|---|
| author | Tellex, Stefanie Li, Adrian Knepper, Ross A. Roy, Nicholas Rus, Daniela L |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Tellex, Stefanie Li, Adrian Knepper, Ross A. Roy, Nicholas Rus, Daniela L |
| author_sort | Tellex, Stefanie |
| collection | MIT |
| description | Robots inevitably fail, often without the ability to recover autonomously. We demonstrate an approach for enabling a robot to recover from failures by communicating its need for specific help to a human partner using natural language. Our approach automatically detects failures, then generates targeted spoken-language requests for help such as “Please give me the white table leg that is on the black table.” Once the human partner has repaired the failure condition, the system resumes full autonomy. We present a novel inverse semantics algorithm for generating effective help requests. In contrast to forward semantic models that interpret natural language in terms of robot actions and perception, our inverse semantics algorithm generates requests by emulating the human’s ability to interpret a request using the Generalized Grounding Graph (G[superscript 3]) framework. To assess the effectiveness of our approach, we present a corpus-based online evaluation, as well as an end-to-end user study, demonstrating that our approach increases the effectiveness of human interventions compared to static requests for help. |
| first_indexed | 2024-09-23T15:05:48Z |
| format | Article |
| id | mit-1721.1/105525 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T15:05:48Z |
| publishDate | 2016 |
| publisher | Springer US |
| record_format | dspace |
| spelling | mit-1721.1/1055252022-10-02T00:35:13Z Recovering from failure by asking for help Tellex, Stefanie Li, Adrian Knepper, Ross A. Roy, Nicholas Rus, Daniela L Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology. Department of Aeronautics and Astronautics Roy, Nicholas Rus, Daniela L Robots inevitably fail, often without the ability to recover autonomously. We demonstrate an approach for enabling a robot to recover from failures by communicating its need for specific help to a human partner using natural language. Our approach automatically detects failures, then generates targeted spoken-language requests for help such as “Please give me the white table leg that is on the black table.” Once the human partner has repaired the failure condition, the system resumes full autonomy. We present a novel inverse semantics algorithm for generating effective help requests. In contrast to forward semantic models that interpret natural language in terms of robot actions and perception, our inverse semantics algorithm generates requests by emulating the human’s ability to interpret a request using the Generalized Grounding Graph (G[superscript 3]) framework. To assess the effectiveness of our approach, we present a corpus-based online evaluation, as well as an end-to-end user study, demonstrating that our approach increases the effectiveness of human interventions compared to static requests for help. Boeing Company U.S. Army Research Laboratory (Robotics Collaborative Technology Alliance) 2016-12-02T17:15:53Z 2016-12-02T17:15:53Z 2015-08 2014-12 2016-08-18T15:42:35Z Article http://purl.org/eprint/type/JournalArticle 0929-5593 1573-7527 http://hdl.handle.net/1721.1/105525 Knepper, Ross A., Stefanie Tellex, Adrian Li, Nicholas Roy, and Daniela Rus. “Recovering from Failure by Asking for Help.” Auton Robot 39, no. 3 (August 6, 2015): 347–362. https://orcid.org/0000-0002-8293-0492 https://orcid.org/0000-0001-5473-3566 en http://dx.doi.org/10.1007/s10514-015-9460-1 Autonomous Robots Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ Springer Science+Business Media New York application/pdf Springer US Springer US |
| spellingShingle | Tellex, Stefanie Li, Adrian Knepper, Ross A. Roy, Nicholas Rus, Daniela L Recovering from failure by asking for help |
| title | Recovering from failure by asking for help |
| title_full | Recovering from failure by asking for help |
| title_fullStr | Recovering from failure by asking for help |
| title_full_unstemmed | Recovering from failure by asking for help |
| title_short | Recovering from failure by asking for help |
| title_sort | recovering from failure by asking for help |
| url | http://hdl.handle.net/1721.1/105525 https://orcid.org/0000-0002-8293-0492 https://orcid.org/0000-0001-5473-3566 |
| work_keys_str_mv | AT tellexstefanie recoveringfromfailurebyaskingforhelp AT liadrian recoveringfromfailurebyaskingforhelp AT knepperrossa recoveringfromfailurebyaskingforhelp AT roynicholas recoveringfromfailurebyaskingforhelp AT rusdanielal recoveringfromfailurebyaskingforhelp |