DocumentCode :
3293107
Title :
Generating human-like social motion in a human-looking humanoid robot: The biomimetic approach
Author :
Rahman, S. M. Mizanoor
Author_Institution :
Dept. of Mech. Eng., Vrije Univ. Brussel (VUB), Brussels, Belgium
fYear :
2013
fDate :
12-14 Dec. 2013
Firstpage :
1377
Lastpage :
1383
Abstract :
The objective was to generate human-like social motion in a human-looking humanoid robot following the biomimetic approach. We thought that the robot´s motion should be human-like for their social acceptance and it could be done if and only if (i) the robot appearance is made human-like, and (ii) human kinematics is implemented in the robot to generate the motion. To address the above, we introduced a very new custom-made human-looking humanoid robot including our in-house software and communication platform to animate the robot. We then studied and analyzed human´s hand waving motion (a representative social motion) using a 3D motion capture system and implemented the human´s kinematic and temporal information into the robot to generate the human-like hand waving. We then studied and analyzed the robot´s hand waving motion using the same motion capture system and compared them with that of the human. The subjects also subjectively evaluated the robot´s hand waving motion for its human-likeness. The results showed that the robot´s motion, which was inspired by the human, was almost same as the human motion. The findings will help develop socially accepted humanoid robots with a proper combination of human-like performances and appearances.
Keywords :
humanoid robots; image motion analysis; robot kinematics; 3D motion capture system; biomimetic approach; custom-made human-looking humanoid robot; hand waving motion; human kinematics; human-like social motion; kinematic information; robot motion; temporal information; Elbow; Humanoid robots; Joints; Kinematics; Shoulder; Wrist;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Robotics and Biomimetics (ROBIO), 2013 IEEE International Conference on
Conference_Location :
Shenzhen
Type :
conf
DOI :
10.1109/ROBIO.2013.6739657
Filename :
6739657
Link To Document :
بازگشت