Mechanical implementation of kinematic synergy for continual grasping generation of anthropomorphic hand

Chen, Wenbin, Xiong, Caihua and Yue, Shigang (2015) Mechanical implementation of kinematic synergy for continual grasping generation of anthropomorphic hand. IEEE/ASME Transactions on Mechatronics, 20 (3). pp. 1249-1263. ISSN 1083-4435

Full content URL: https://doi.org/10.1109/TMECH.2014.2329006

Documents
Mechanical implementation of kinematic synergy for continual grasping generation of anthropomorphic hand
17879 ChenWXiongCYueS 2014 mechanical implementation of kinematic synergy for continual grasping generation IEEEASME TM-06843988.pdf
[img]
[Download]
[img] PDF
2015 ChenW&XiongC&YueS 2014 mechanical implementation of kinematic synergy for continual grasping generation IEEE&ASME-downloaded.pdf - Whole Document
Restricted to Repository staff only

1MB
[img]
Preview
PDF
17879 ChenWXiongCYueS 2014 mechanical implementation of kinematic synergy for continual grasping generation IEEEASME TM-06843988.pdf - Whole Document

1MB
Item Type:Article
Item Status:Live Archive

Abstract

The synergy-based motion generation of current anthropomorphic hands generally employ the static posture synergy, which is extracted from quantities of joint trajectory, to design the mechanism or control strategy. Under this framework, the temporal weight sequences of each synergy from pregrasp phase to grasp phase are required for reproducing any grasping task. Moreover, the zero-offset posture has to be preset before starting any grasp. Thus, the whole grasp phase appears to be unlike natural human grasp. Up until now, no work in the literature addresses these issues toward simplifying the continual grasp by only inputting the grasp pattern. In this paper, the kinematic synergies observed in angular velocity profile are employed to design the motion generation mechanism. The kinematic synergy extracted from quantities of grasp tasks is implemented by the proposed eigen cam group in tendon space. The completely continual grasp from the fully extending posture only require averagely rotating the two eigen cam groups one cycle. The change of grasp pattern only depends on respecifying transmission ratio pair for the two eigen cam groups. An illustrated hand prototype is developed based on the proposed design principle and the grasping experiments demonstrate the feasibility of the design method. The potential applications include the prosthetic hand that is controlled by the classified pattern from the bio-signal.

Keywords:angular velocity control, grippers, manipulator kinematics, motion control, trajectory control, bmjgoldcheck, NotOAChecked
Subjects:H Engineering > H671 Robotics
Divisions:College of Science > School of Computer Science
Related URLs:
ID Code:17879
Deposited On:17 Jul 2015 08:05

Repository Staff Only: item control page