Functional contour-following via haptic perception and reinforcement learning

Simple item page
buir.contributor.authorTekin, Cem
dc.citation.epage72en_US
dc.citation.issueNumber1en_US
dc.citation.spage61en_US
dc.citation.volumeNumber11en_US
dc.contributor.authorHellman, R. B.en_US
dc.contributor.authorTekin, Cemen_US
dc.contributor.authorSchaar, M. V.en_US
dc.contributor.authorSantos, V. J.en_US
dc.date.accessioned2019-02-21T16:05:53Zen_US
dc.date.available2019-02-21T16:05:53Zen_US
dc.date.issued2018en_US
dc.departmentDepartment of Electrical and Electronics Engineeringen_US
dc.description.abstractMany tasks involve the fine manipulation of objects despite limited visual feedback. In such scenarios, tactile and proprioceptive feedback can be leveraged for task completion. We present an approach for real-time haptic perception and decision-making for a haptics-driven, functional contour-following task: The closure of a ziplock bag. This task is challenging for robots because the bag is deformable, transparent, and visually occluded by artificial fingertip sensors that are also compliant. A deep neural net classifier was trained to estimate the state of a zipper within a robot's pinch grasp. A Contextual Multi-Armed Bandit (C-MAB) reinforcement learning algorithm was implemented to maximize cumulative rewards by balancing exploration versus exploitation of the state-action space. The C-MAB learner outperformed a benchmark Q-learner by more efficiently exploring the state-action space while learning a hard-to-code task. The learned C-MAB policy was tested with novel ziplock bag scenarios and contours (wire, rope). Importantly, this work contributes to the development of reinforcement learning approaches that account for limited resources such as hardware life and researcher time. As robots are used to perform complex, physically interactive tasks in unstructured or unmodeled environments, it becomes important to develop methods that enable efficient and effective learning with physical testbeds.en_US
dc.description.provenanceMade available in DSpace on 2019-02-21T16:05:53Z (GMT). No. of bitstreams: 1 Bilkent-research-paper.pdf: 222869 bytes, checksum: 842af2b9bd649e7f548593affdbafbb3 (MD5) Previous issue date: 2018en
dc.description.sponsorshipThe authors wish to thank Peter Aspinall for assistance with the construction of the robot testbed. This work was supported in part by National Science Foundation Awards #1461547, #1463960, and #1533983, and the Office of Naval Research Award #N00014-16-1-2468.en_US
dc.identifier.doi10.1109/TOH.2017.2753233en_US
dc.identifier.eissn2329-4051
dc.identifier.issn1939-1412en_US
dc.identifier.urihttp://hdl.handle.net/11693/50279en_US
dc.language.isoEnglishen_US
dc.publisherInstitute of Electrical and Electronics Engineersen_US
dc.relation.isversionofhttps://doi.org/10.1109/TOH.2017.2753233en_US
dc.relation.project1461547 - 1463960 - 1533983 - Office of Naval Research, ONR: 00014-16-1-2468en_US
dc.source.titleIEEE Transactions on Hapticsen_US
dc.subjectActive touchen_US
dc.subjectContour-followingen_US
dc.subjectDecision makingen_US
dc.subjectHaptic perceptionen_US
dc.subjectManipulationen_US
dc.subjectReinforcement learningen_US
dc.titleFunctional contour-following via haptic perception and reinforcement learningen_US
dc.typeArticleen_US

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Functional_Contour_following_via_Haptic_Perception_and_Reinforcement_Learning.pdf
Size:
888.84 KB
Format:
Adobe Portable Document Format
Description:
Full printable version
Download

Collections

Department of Electrical and Electronics Engineering