2026-09-19T01:16:40Zhttps://dspace.mit.edu/server/oai/requestoai:dspace.mit.edu:1721.1/1390482022-01-15T03:25:01Zcom_1721.1_7582com_1721.1_7581col_1721.1_131023
Isola, Phillip J.
Lin, Yen-Chen
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
2022-01-14T14:46:45Z
2022-01-14T14:46:45Z
2021-06
2021-06-24T19:25:32.642Z
https://hdl.handle.net/1721.1/139048
Humans are remarkable at manipulating unfamiliar objects. For the past decades of robotics, tremendous efforts have been dedicated to endow robot manipulation systems with such capabilities. As classic solutions typically require prior knowledge of the objects (e.g., 3D CAD models) which are not available in the unstructured environments, data-driven solutions that learn from robot-environment interactions (e.g., trial and error) have emerged as a promising approach for autonomously acquiring complex skills for manipulation. For data-driven methods, the ability to do more with less data is incredibly important, since data collection through physical interaction between the robots and the environment can be both time consuming and expensive. In this thesis, we develop transfer learning algorithms for robotic manipulation in order to reduce the amount of robot-environment interactions needed to adapt to different environments. With real robot hardware, we show that our algorithms enable robots to learn to pick and grasp arbitrary objects with 10 minutes of trial and error, and help robots learn to push unfamiliar objects with 5 interactions.
S.M.
Massachusetts Institute of Technology
In Copyright - Educational Use Permitted
Copyright MIT
http://rightsstatements.org/page/InC-EDU/1.0/
Visual Transfer Learning for Robotic Manipulation
Thesis
application/pdf
Master
Master of Science in Electrical Engineering and Computer Science
Publication
unknown
unknown
<Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="a4d75f6e-5ce6-45fd-91a6-a5cb87ea98ad">
<Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type>
<Title>Visual Transfer Learning for Robotic Manipulation</Title>
<PublishedIn>
<Publication>
</Publication>
</PublishedIn>
<PublicationDate>2021-06</PublicationDate>
<Authors>
<Author>
<DisplayName>Lin, Yen-Chen</DisplayName>
<Affiliation>
<OrgUnit>
</OrgUnit>
</Affiliation>
</Author>
</Authors>
<Editors>
</Editors>
<Publishers>
<Publisher>
<DisplayName>Massachusetts Institute of Technology</DisplayName>
<OrgUnit />
</Publisher>
</Publishers>
<License>http://rightsstatements.org/page/InC-EDU/1.0/</License>
<Abstract>Humans are remarkable at manipulating unfamiliar objects. For the past decades of robotics, tremendous efforts have been dedicated to endow robot manipulation systems with such capabilities. As classic solutions typically require prior knowledge of the objects (e.g., 3D CAD models) which are not available in the unstructured environments, data-driven solutions that learn from robot-environment interactions (e.g., trial and error) have emerged as a promising approach for autonomously acquiring complex skills for manipulation. For data-driven methods, the ability to do more with less data is incredibly important, since data collection through physical interaction between the robots and the environment can be both time consuming and expensive. In this thesis, we develop transfer learning algorithms for robotic manipulation in order to reduce the amount of robot-environment interactions needed to adapt to different environments. With real robot hardware, we show that our algorithms enable robots to learn to pick and grasp arbitrary objects with 10 minutes of trial and error, and help robots learn to push unfamiliar objects with 5 interactions.</Abstract>
<Access xmlns="http://purl.org/coar/access_right"
>
</Access>
</Publication>