DocumentCode :
2194704
Title :
A Bundle Trust Region Method for Optimal Grasp Planning of Multi-fingered Robot Hand
Author :
Yin, Yingjie ; Tamura, Kazuya ; Hosoe, Shigeyuki
Author_Institution :
Bio-mimetic Control Res. Center, Inst. of Phys. & Chem. Res., Saitama
fYear :
2006
fDate :
17-20 Dec. 2006
Firstpage :
884
Lastpage :
889
Abstract :
How to plan a proper set of contact points on an object so as to maximize the grasp capability is a common but generally difficult problem because of its nondifferentiable complexity. In current literature, subgradient method is commonly used in grasp planning. However, it converges slowly and requires much iteration to reach an optimum. In this study, we adapt the bundle trust region (BTR) method to the synthesis problem for more fast convergence and short computation time. Numerical tests validate the effectiveness of this new trial of BTR to the grasp planning.
Keywords :
manipulators; planning; bundle trust region method; grasp planning; multifingered robot hand; optimal grasp planning; subgradient method; Biomimetics; Convergence; Fingers; Grasping; Humans; Optimal control; Optimization methods; Robotics and automation; Robots; Shape; Grasp planning; bundle trust region method; nondifferentiable optimization;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Robotics and Biomimetics, 2006. ROBIO '06. IEEE International Conference on
Conference_Location :
Kunming
Print_ISBN :
1-4244-0570-X
Electronic_ISBN :
1-4244-0571-8
Type :
conf
DOI :
10.1109/ROBIO.2006.340337
Filename :
4141982
Link To Document :
بازگشت