國立虎尾科技大學 |

Grasp Force Measurement via Fingernail Imaging with Multiple Robotic Arms.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Grasp Force Measurement via Fingernail Imaging with Multiple Robotic Arms./
作者:	Fallahinia, Navid.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2022,
面頁冊數:	91 p.
附註:	Source: Dissertations Abstracts International, Volume: 84-05, Section: B.
Contained By:	Dissertations Abstracts International84-05B.
標題:	Robotics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28966940
ISBN:	9798357570192

Grasp Force Measurement via Fingernail Imaging with Multiple Robotic Arms.
Fallahinia, Navid.

Grasp Force Measurement via Fingernail Imaging with Multiple Robotic Arms. - Ann Arbor : ProQuest Dissertations & Theses, 2022 - 91 p.

Source: Dissertations Abstracts International, Volume: 84-05, Section: B.

Thesis (Ph.D.)--The University of Utah, 2022.

This item must not be sold to any third party vendors.

The objective of this dissertation is to develop an entirely vision-based system of multiple robotic arms for interaction sensing during Human-Robot Interaction (HRI) applications by dynamically tracking human fingers in a workspace and estimating 3D grasp forces via fingernail imaging. Fingernail imaging has been proven to be an effective method for measuring grasp forces on multiple fingers in a totally unconstrained manner, i.e. without restricting the human’s haptic senses or constraining how the human grasps an object. However, this method is limited by the ability to maintain a proper view of the human fingernails with one or more cameras as the fingernails translate and rotate within a workspace during a dynamic grasping task. To overcome this problem, a system of multiple robotic arms with cameras mounted on their end-effectors has been developed. Furthermore, a novel approach for the real-time estimation of 3D tactile forces via Deep Neural Networks (DNN) has been introduced, which is entirely monocular vision-based and does not require any physical force sensor. Unlike prior models, the new DNN model does not need to be calibrated to an individual human subject. This research is the first to attempt unconstrained dynamic force estimation during object grasping via fingernail imaging using multiple robots for tracking human fingers. Distinct technical contributions include 1) A visual servoing algorithm that enables one or more robots to detect and dynamically track the fingers based on fingernail features, 2) A DNN model for fingernail image alignment and estimating 3D grasp forces in real-time using imaging, 3) Sensitivity analysis characterizing how the accuracy of force estimation depends on contact surface curvature 4) A never-before-seen comparison of constrained versus unconstrained finger placement and grasp force distributions during dynamic grasping tasks, and 5) An experimental characterization of the performance of the system for real-time estimation of grasp forces. The multiple robotic system enables grasp force measurement via fingernail imaging without constraining the contact locations of the fingers, and without constraining the position of the hand as it moves in the workspace.

ISBN: 9798357570192Subjects--Topical Terms:

561941
Robotics.
Subjects--Index Terms:

Robotic arms

Grasp Force Measurement via Fingernail Imaging with Multiple Robotic Arms.
LDR:03310nam a2200337 4500 001 1104575
005 20230619080052.5
006 m o d
007 cr#unu||||||||
008 230907s2022 ||||||||||||||||| ||eng d
020 $a 9798357570192
035 $a (MiAaPQ)AAI28966940
035 $a AAI28966940
040 $a MiAaPQ $c MiAaPQ
100 1 $a Fallahinia, Navid. $3 1413435
245 1 0 $a Grasp Force Measurement via Fingernail Imaging with Multiple Robotic Arms.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2022
300 $a 91 p.
500 $a Source: Dissertations Abstracts International, Volume: 84-05, Section: B.
500 $a Advisor: Mascaro, Stephen A.
502 $a Thesis (Ph.D.)--The University of Utah, 2022.
506 $a This item must not be sold to any third party vendors.
520 $a The objective of this dissertation is to develop an entirely vision-based system of multiple robotic arms for interaction sensing during Human-Robot Interaction (HRI) applications by dynamically tracking human fingers in a workspace and estimating 3D grasp forces via fingernail imaging. Fingernail imaging has been proven to be an effective method for measuring grasp forces on multiple fingers in a totally unconstrained manner, i.e. without restricting the human’s haptic senses or constraining how the human grasps an object. However, this method is limited by the ability to maintain a proper view of the human fingernails with one or more cameras as the fingernails translate and rotate within a workspace during a dynamic grasping task. To overcome this problem, a system of multiple robotic arms with cameras mounted on their end-effectors has been developed. Furthermore, a novel approach for the real-time estimation of 3D tactile forces via Deep Neural Networks (DNN) has been introduced, which is entirely monocular vision-based and does not require any physical force sensor. Unlike prior models, the new DNN model does not need to be calibrated to an individual human subject. This research is the first to attempt unconstrained dynamic force estimation during object grasping via fingernail imaging using multiple robots for tracking human fingers. Distinct technical contributions include 1) A visual servoing algorithm that enables one or more robots to detect and dynamically track the fingers based on fingernail features, 2) A DNN model for fingernail image alignment and estimating 3D grasp forces in real-time using imaging, 3) Sensitivity analysis characterizing how the accuracy of force estimation depends on contact surface curvature 4) A never-before-seen comparison of constrained versus unconstrained finger placement and grasp force distributions during dynamic grasping tasks, and 5) An experimental characterization of the performance of the system for real-time estimation of grasp forces. The multiple robotic system enables grasp force measurement via fingernail imaging without constraining the contact locations of the fingers, and without constraining the position of the hand as it moves in the workspace.
590 $a School code: 0240.
650 4 $a Robotics. $3 561941
650 4 $a Mechanical engineering. $3 557493
653 $a Robotic arms
690 $a 0548
690 $a 0771
710 2 $a The University of Utah. $b Mechanical Engineering. $3 1180268
773 0 $t Dissertations Abstracts International $g 84-05B.
790 $a 0240
791 $a Ph.D.
792 $a 2022
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28966940