國立虎尾科技大學 |

Object Detection and Pose Estimation from Rgb and Depth Data for Real-Time, Adaptive Robotic Grasping.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Object Detection and Pose Estimation from Rgb and Depth Data for Real-Time, Adaptive Robotic Grasping./
作者:	Paul, Shuvo Kumar.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2020,
面頁冊數:	64 p.
附註:	Source: Masters Abstracts International, Volume: 82-01.
Contained By:	Masters Abstracts International82-01.
標題:	Robotics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=27959084
ISBN:	9798641332925

Object Detection and Pose Estimation from Rgb and Depth Data for Real-Time, Adaptive Robotic Grasping.
Paul, Shuvo Kumar.

Object Detection and Pose Estimation from Rgb and Depth Data for Real-Time, Adaptive Robotic Grasping. - Ann Arbor : ProQuest Dissertations & Theses, 2020 - 64 p.

Source: Masters Abstracts International, Volume: 82-01.

Thesis (M.S.)--University of Nevada, Reno, 2020.

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

In recent times, object detection and pose estimation have gained significant attention in the context of robotic vision applications. Both the identification of objects of interest as well as the estimation of their pose remain important capabilities in order for robots to provide effective assistance for numerous robotic applications ranging from household tasks to industrial manipulation. This problem is particularly challenging because of the heterogeneity of objects having different and potentially complex shapes, and the difficulties arising due to background clutter and partial occlusions between objects. As the main contribution of this work, we propose a system that performs real-time object detection and pose estimation, for the purpose of dynamic robot grasping. The robot has been pre-trained to perform a small set of canonical grasps from a few fixed poses for each object. When presented with an unknown object in an arbitrary pose, the proposed approach allows the robot to detect the object identity and its actual pose, and then adapt a canonical grasp in order to be used with the new pose. For training, the system defines a canonical grasp by capturing the relative pose of an object with respect to the gripper attached to the robot’s wrist. During testing, once a new pose is detected, a canonical grasp for the object is identified and then dynamically adapted by adjusting the robot arm's joint angles, so that the gripper can grasp the object in its new pose. We conducted experiments using a humanoid PR2 robot and showed that the proposed framework can detect well-textured objects, and provide accurate pose estimation in the presence of tolerable amounts of out-of-plane rotation. The performance is also illustrated by the robot successfully grasping objects from a wide range of arbitrary poses.

ISBN: 9798641332925Subjects--Topical Terms:

561941
Robotics.
Subjects--Index Terms:

Adaptive grasping

Object Detection and Pose Estimation from Rgb and Depth Data for Real-Time, Adaptive Robotic Grasping.
LDR:02971nam a2200361 4500 001 1037969
005 20210910100644.5
008 211029s2020 ||||||||||||||||| ||eng d
020 $a 9798641332925
035 $a (MiAaPQ)AAI27959084
035 $a AAI27959084
040 $a MiAaPQ $c MiAaPQ
100 1 $a Paul, Shuvo Kumar. $3 1335282
245 1 0 $a Object Detection and Pose Estimation from Rgb and Depth Data for Real-Time, Adaptive Robotic Grasping.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2020
300 $a 64 p.
500 $a Source: Masters Abstracts International, Volume: 82-01.
500 $a Advisor: Nicolescu, Mircea.
502 $a Thesis (M.S.)--University of Nevada, Reno, 2020.
506 $a This item must not be sold to any third party vendors.
520 $a In recent times, object detection and pose estimation have gained significant attention in the context of robotic vision applications. Both the identification of objects of interest as well as the estimation of their pose remain important capabilities in order for robots to provide effective assistance for numerous robotic applications ranging from household tasks to industrial manipulation. This problem is particularly challenging because of the heterogeneity of objects having different and potentially complex shapes, and the difficulties arising due to background clutter and partial occlusions between objects. As the main contribution of this work, we propose a system that performs real-time object detection and pose estimation, for the purpose of dynamic robot grasping. The robot has been pre-trained to perform a small set of canonical grasps from a few fixed poses for each object. When presented with an unknown object in an arbitrary pose, the proposed approach allows the robot to detect the object identity and its actual pose, and then adapt a canonical grasp in order to be used with the new pose. For training, the system defines a canonical grasp by capturing the relative pose of an object with respect to the gripper attached to the robot’s wrist. During testing, once a new pose is detected, a canonical grasp for the object is identified and then dynamically adapted by adjusting the robot arm's joint angles, so that the gripper can grasp the object in its new pose. We conducted experiments using a humanoid PR2 robot and showed that the proposed framework can detect well-textured objects, and provide accurate pose estimation in the presence of tolerable amounts of out-of-plane rotation. The performance is also illustrated by the robot successfully grasping objects from a wide range of arbitrary poses.
590 $a School code: 0139.
650 4 $a Robotics. $3 561941
650 4 $a Computer science. $3 573171
653 $a Adaptive grasping
653 $a Homography
653 $a Planar pose
653 $a Pose estimation
653 $a Robotic grasp
690 $a 0984
690 $a 0771
710 2 $a University of Nevada, Reno. $b Computer Science. $3 845722
773 0 $t Masters Abstracts International $g 82-01.
790 $a 0139
791 $a M.S.
792 $a 2020
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=27959084