Title
Human-Like Behavior Generation Based on Head-Arms Model for Robot Tracking External Targets and Body Parts.
Abstract
Facing and pointing toward moving targets is a usual and natural behavior in daily life. Social robots should be able to display such coordinated behaviors in order to interact naturally with people. For instance, a robot should be able to point and look at specific objects. This is why, a scheme to generate coordinated head-arm motion for a humanoid robot with two degrees-of-freedom for the head and seven for each arm is proposed in this paper. Specifically, a virtual plane approach is employed to generate the analytical solution of the head motion. A quadratic program (QP)-based method is exploited to formulate the coordinated dual-arm motion. To obtain the optimal solution, a simplified recurrent neural network is used to solve the QP problem. The effectiveness of the proposed scheme is demonstrated using both computer simulation and physical experiments.
Year
DOI
Venue
2015
10.1109/TCYB.2014.2351416
IEEE transactions on cybernetics
Keywords
Field
DocType
head-arm motion generation,humanoid robot,redundancy resolution,tracking targets and body parts,robot kinematics,humanoid robots,face,biomedical research,bioinformatics
Social robot,Computer vision,Head movements,Recurrent neural network,Robot kinematics,Artificial intelligence,Quadratic programming,Robot,Mathematics,Robotics,Humanoid robot
Journal
Volume
Issue
ISSN
PP
99
2168-2275
Citations 
PageRank 
References 
13
0.64
21
Authors
3
Name
Order
Citations
PageRank
Zhijun Zhang1130.64
Aryel Beck2130.64
Nadia Magnenat-Thalmann35119659.15