Motor learning on a heaving plate via improved-SNR algorithms
Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2009.
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | eng |
| Published: |
Massachusetts Institute of Technology
2009
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/46638 |
| _version_ | 1826200802716286976 |
|---|---|
| author | Roberts, John W., Ph. D. Massachusetts Institute of Technology |
| author2 | Russ Tedrake. |
| author_facet | Russ Tedrake. Roberts, John W., Ph. D. Massachusetts Institute of Technology |
| author_sort | Roberts, John W., Ph. D. Massachusetts Institute of Technology |
| collection | MIT |
| description | Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2009. |
| first_indexed | 2024-09-23T11:42:00Z |
| format | Thesis |
| id | mit-1721.1/46638 |
| institution | Massachusetts Institute of Technology |
| language | eng |
| last_indexed | 2024-09-23T11:42:00Z |
| publishDate | 2009 |
| publisher | Massachusetts Institute of Technology |
| record_format | dspace |
| spelling | mit-1721.1/466382019-04-12T20:27:40Z Motor learning on a heaving plate via improved-SNR algorithms Motor learning on a heaving plate via improved-Signal-to-Noise Ratio algorithms Roberts, John W., Ph. D. Massachusetts Institute of Technology Russ Tedrake. Massachusetts Institute of Technology. Dept. of Mechanical Engineering. Massachusetts Institute of Technology. Dept. of Mechanical Engineering. Mechanical Engineering. Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2009. Includes bibliographical references (p. 71-75). Creatures in nature have subtle and complicated interactions with their surrounding fluids, achieving levels of performance as yet unmatched by engineered solutions. Model-free reinforcement learning (MFRL) holds the promise of allowing man-made controllers to take advantage of the subtlety of fluid-body interactions solely using data gathered on the actual system to be controlled. In this thesis, improved MFRL algorithms, motivated by a novel Signal-to-Noise Ratio for policy gradient algorithms, are developed, and shown to provide more efficient learning in noisy environments. These algorithms are then demonstrated on a heaving foil, where it is shown to learn a flapping gait on an experimental system orders of magnitude faster than the dynamics can be simulated, suggesting broad applications both in controlling robots with complex dynamics and in the study of controlled fluid systems. by John W. Roberts. S.M. 2009-08-26T17:09:49Z 2009-08-26T17:09:49Z 2009 2009 Thesis http://hdl.handle.net/1721.1/46638 426489366 eng M.I.T. theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. See provided URL for inquiries about permission. http://dspace.mit.edu/handle/1721.1/7582 75 p. application/pdf Massachusetts Institute of Technology |
| spellingShingle | Mechanical Engineering. Roberts, John W., Ph. D. Massachusetts Institute of Technology Motor learning on a heaving plate via improved-SNR algorithms |
| title | Motor learning on a heaving plate via improved-SNR algorithms |
| title_full | Motor learning on a heaving plate via improved-SNR algorithms |
| title_fullStr | Motor learning on a heaving plate via improved-SNR algorithms |
| title_full_unstemmed | Motor learning on a heaving plate via improved-SNR algorithms |
| title_short | Motor learning on a heaving plate via improved-SNR algorithms |
| title_sort | motor learning on a heaving plate via improved snr algorithms |
| topic | Mechanical Engineering. |
| url | http://hdl.handle.net/1721.1/46638 |
| work_keys_str_mv | AT robertsjohnwphdmassachusettsinstituteoftechnology motorlearningonaheavingplateviaimprovedsnralgorithms AT robertsjohnwphdmassachusettsinstituteoftechnology motorlearningonaheavingplateviaimprovedsignaltonoiseratioalgorithms |