2026-09-19T23:31:28Zhttps://dspace.mit.edu/server/oai/requestoai:dspace.mit.edu:1721.1/1624132025-08-22T03:07:01Zcom_1721.1_7582com_1721.1_7581col_1721.1_131024
Agrawal, Pulkit
Chang, Ethan
Massachusetts Institute of Technology. Department of Mechanical Engineering
2025-08-21T17:00:22Z
2025-08-21T17:00:22Z
2025-05
2025-06-17T16:10:42.221Z
https://hdl.handle.net/1721.1/162413
https://orcid.org/0009-0005-0166-7493
Robotic manipulation remains a complex and unsolved challenge due to the need for adaptability across diverse objects and tasks. In this work, we explore how to train effective manipulation policies using reinforcement learning in simulation for the Eyesight Hand: a novel, low-cost, tactile-enabled robotic hand. We implement a range of experiments in MuJoCo to evaluate the impact of controller types, observation spaces, reward formulations, and curriculum strategies on policy performance. Our findings highlight the benefits of delta position control, a carefully selected observation space including joint states, control vectors, object pose, and contact forces, and success-driven curriculum learning. Our study establishes baseline strategies for training robust, tactile-based policies on this in-house hardware.
S.B.
Massachusetts Institute of Technology
In Copyright - Educational Use Permitted
Copyright retained by author(s)
https://rightsstatements.org/page/InC-EDU/1.0/
Simulation-Based Reinforcement Learning Policy Optimization for
Tactile Manipulation: A Case Study on the Eyesight Hand
Thesis
application/pdf
Bachelor
Bachelor of Science in Mechanical Engineering
Publication
unknown
unknown
<Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="c83c1156-aede-48eb-a536-46204096df3e">
<Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type>
<Title>Simulation-Based Reinforcement Learning Policy Optimization for
Tactile Manipulation: A Case Study on the Eyesight Hand</Title>
<PublishedIn>
<Publication>
</Publication>
</PublishedIn>
<PublicationDate>2025-05</PublicationDate>
<Authors>
<Author>
<DisplayName>Chang, Ethan</DisplayName>
<Affiliation>
<OrgUnit>
</OrgUnit>
</Affiliation>
</Author>
</Authors>
<Editors>
</Editors>
<Publishers>
<Publisher>
<DisplayName>Massachusetts Institute of Technology</DisplayName>
<OrgUnit />
</Publisher>
</Publishers>
<License>https://rightsstatements.org/page/InC-EDU/1.0/</License>
<Abstract>Robotic manipulation remains a complex and unsolved challenge due to the need for adaptability across diverse objects and tasks. In this work, we explore how to train effective manipulation policies using reinforcement learning in simulation for the Eyesight Hand: a novel, low-cost, tactile-enabled robotic hand. We implement a range of experiments in MuJoCo to evaluate the impact of controller types, observation spaces, reward formulations, and curriculum strategies on policy performance. Our findings highlight the benefits of delta position control, a carefully selected observation space including joint states, control vectors, object pose, and contact forces, and success-driven curriculum learning. Our study establishes baseline strategies for training robust, tactile-based policies on this in-house hardware.</Abstract>
<Access xmlns="http://purl.org/coar/access_right"
>
</Access>
</Publication>