2026-09-19T16:43:47Zhttps://dspace.mit.edu/server/oai/request

oai:dspace.mit.edu:1721.1/1536812024-03-14T03:01:04Zcom_1721.1_7582com_1721.1_7581col_1721.1_131022

Adelson, Edward H. Liu, Sandra Q. Massachusetts Institute of Technology. Department of Mechanical Engineering 2024-03-13T13:26:10Z 2024-03-13T13:26:10Z 2024-02 2024-02-15T21:16:07.636Z https://hdl.handle.net/1721.1/153681 When we look to the future of soft robotics and manipulation, we begin to look towards sensory-rich and compliant grasping mechanisms. Not only do we want to capitalize on the significant advantages in safety and adaptability that soft robots have, we also want to incorporate high-resolution tactile sensors, which will allow soft robots to perform more tasks. One such system is the GelSight sensor, which is low-cost, effective, and high-resolution. However, the integration of these camera-based sensors into compliant manipulators is difficult due to the rigidity of the sensor backing. This thesis explores the design of multiple different compliant high-resolution tactile manipulators, along with some examples of their real-world uses. The first such design incorporates a simple camera-based tactile sensor into an exoskeleton-covered soft robot with vision-based proprioception. A later design integrates full camera-based tactile sensing capabilities into a flexible Fin Ray structure. Finally, the designs culminate in a novel soft-rigid human-inspired robotic hand with continuous tactile sensing which is capable of grasping heavier objects and safely interacting with humans. The incorporation of high-resolution tactile sensors into soft, compliant robots brings us closer to developing new manipulators that could someday match or exceed the capabilities of human hands. Ph.D. Massachusetts Institute of Technology Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) Copyright retained by author(s) https://creativecommons.org/licenses/by-nc-nd/4.0/ Soft, Compliant Tactile Robotic Manipulators Thesis application/pdf Doctoral Doctor of Philosophy Publication unknown unknown <Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="f937c71b-7844-45fb-aaea-640ebc3e61eb"> <Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type> <Title>Soft, Compliant Tactile Robotic Manipulators</Title> <PublishedIn> <Publication> </Publication> </PublishedIn> <PublicationDate>2024-02</PublicationDate> <Authors> <Author> <DisplayName>Liu, Sandra Q.</DisplayName> <Affiliation> <OrgUnit> </OrgUnit> </Affiliation> </Author> </Authors> <Editors> </Editors> <Publishers> <Publisher> <DisplayName>Massachusetts Institute of Technology</DisplayName> <OrgUnit /> </Publisher> </Publishers> <License>https://creativecommons.org/licenses/by-nc-nd/4.0/</License> <Abstract>When we look to the future of soft robotics and manipulation, we begin to look towards sensory-rich and compliant grasping mechanisms. Not only do we want to capitalize on the significant advantages in safety and adaptability that soft robots have, we also want to incorporate high-resolution tactile sensors, which will allow soft robots to perform more tasks. One such system is the GelSight sensor, which is low-cost, effective, and high-resolution. However, the integration of these camera-based sensors into compliant manipulators is difficult due to the rigidity of the sensor backing. This thesis explores the design of multiple different compliant high-resolution tactile manipulators, along with some examples of their real-world uses. The first such design incorporates a simple camera-based tactile sensor into an exoskeleton-covered soft robot with vision-based proprioception. A later design integrates full camera-based tactile sensing capabilities into a flexible Fin Ray structure. Finally, the designs culminate in a novel soft-rigid human-inspired robotic hand with continuous tactile sensing which is capable of grasping heavier objects and safely interacting with humans. The incorporation of high-resolution tactile sensors into soft, compliant robots brings us closer to developing new manipulators that could someday match or exceed the capabilities of human hands.</Abstract> <Access xmlns="http://purl.org/coar/access_right" > </Access> </Publication>