Interactive Bayesian identification of kinematic mechanisms
This paper addresses the problem of identifying mechanisms based on data gathered while interacting with them. We present a decision-theoretic formulation of this problem, using Bayesian filtering techniques to maintain a distributional estimate of the mechanism type and parameters. In order to redu...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Institute of Electrical and Electronics Engineers (IEEE)
2016
|
| Online Access: | http://hdl.handle.net/1721.1/100723 https://orcid.org/0000-0003-4749-4979 https://orcid.org/0000-0002-8657-2450 https://orcid.org/0000-0001-6054-7145 |
| _version_ | 1826211492808097792 |
|---|---|
| author | Barragan, Patrick R. Lozano-Perez, Tomas Kaelbling, Leslie P. |
| author2 | Massachusetts Institute of Technology. Materials Processing Center |
| author_facet | Massachusetts Institute of Technology. Materials Processing Center Barragan, Patrick R. Lozano-Perez, Tomas Kaelbling, Leslie P. |
| author_sort | Barragan, Patrick R. |
| collection | MIT |
| description | This paper addresses the problem of identifying mechanisms based on data gathered while interacting with them. We present a decision-theoretic formulation of this problem, using Bayesian filtering techniques to maintain a distributional estimate of the mechanism type and parameters. In order to reduce the amount of interaction required to arrive at a confident identification, we select actions explicitly to reduce entropy in the current estimate. We demonstrate the approach on a domain with four primitive and two composite mechanisms. The results show that this approach can correctly identify complex mechanisms including mechanisms which are difficult to model analytically. The results also show that entropy-based action selection can significantly decrease the number of actions required to gather the same information. |
| first_indexed | 2024-09-23T15:06:48Z |
| format | Article |
| id | mit-1721.1/100723 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T15:06:48Z |
| publishDate | 2016 |
| publisher | Institute of Electrical and Electronics Engineers (IEEE) |
| record_format | dspace |
| spelling | mit-1721.1/1007232022-09-29T12:47:02Z Interactive Bayesian identification of kinematic mechanisms Barragan, Patrick R. Lozano-Perez, Tomas Kaelbling, Leslie P. Massachusetts Institute of Technology. Materials Processing Center Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Barragan, Patrick R. Kaelbling, Leslie P. Lozano-Perez, Tomas This paper addresses the problem of identifying mechanisms based on data gathered while interacting with them. We present a decision-theoretic formulation of this problem, using Bayesian filtering techniques to maintain a distributional estimate of the mechanism type and parameters. In order to reduce the amount of interaction required to arrive at a confident identification, we select actions explicitly to reduce entropy in the current estimate. We demonstrate the approach on a domain with four primitive and two composite mechanisms. The results show that this approach can correctly identify complex mechanisms including mechanisms which are difficult to model analytically. The results also show that entropy-based action selection can significantly decrease the number of actions required to gather the same information. National Science Foundation (U.S.) (Grant 1117325) United States. Office of Naval Research. Multidisciplinary University Research Initiative (Grant N00014-09-1-1051) United States. Air Force Office of Scientific Research (Grant FA2386-10-1-4135) Singapore. Ministry of Education (SUTD-MIT International Design Centre) 2016-01-06T16:21:52Z 2016-01-06T16:21:52Z 2014-05 Article http://purl.org/eprint/type/ConferencePaper 978-1-4799-3685-4 http://hdl.handle.net/1721.1/100723 Barragan, Patrick R., Leslie Pack Kaelbling, and Tomas Lozano-Perez. “Interactive Bayesian Identification of Kinematic Mechanisms.” 2014 IEEE International Conference on Robotics and Automation (ICRA) (May 2014). https://orcid.org/0000-0003-4749-4979 https://orcid.org/0000-0002-8657-2450 https://orcid.org/0000-0001-6054-7145 en_US http://dx.doi.org/10.1109/ICRA.2014.6907126 Proceedings of the 2014 IEEE International Conference on Robotics and Automation (ICRA) Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Institute of Electrical and Electronics Engineers (IEEE) MIT web domain |
| spellingShingle | Barragan, Patrick R. Lozano-Perez, Tomas Kaelbling, Leslie P. Interactive Bayesian identification of kinematic mechanisms |
| title | Interactive Bayesian identification of kinematic mechanisms |
| title_full | Interactive Bayesian identification of kinematic mechanisms |
| title_fullStr | Interactive Bayesian identification of kinematic mechanisms |
| title_full_unstemmed | Interactive Bayesian identification of kinematic mechanisms |
| title_short | Interactive Bayesian identification of kinematic mechanisms |
| title_sort | interactive bayesian identification of kinematic mechanisms |
| url | http://hdl.handle.net/1721.1/100723 https://orcid.org/0000-0003-4749-4979 https://orcid.org/0000-0002-8657-2450 https://orcid.org/0000-0001-6054-7145 |
| work_keys_str_mv | AT barraganpatrickr interactivebayesianidentificationofkinematicmechanisms AT lozanopereztomas interactivebayesianidentificationofkinematicmechanisms AT kaelblinglesliep interactivebayesianidentificationofkinematicmechanisms |