Title :
Humanoid robot motion imitation using Kinect
Author :
Hsien-I Lin ; Chan-Ching Chou
Author_Institution :
Grad. Inst. of Autom. Technol., Nat. Taipei Univ. of Technol., Taipei, Taiwan
Abstract :
Humanoid robot motion imitation is a problem of controlling robot balance with given body motion. It is a simple method to program robot motion because many robot motion editing tools do not consider balance control. Thus, this paper proposes a concept that a RGB-D camera such as Kinect is used to capture human motion and apply the captured motion on a humanoid robot. A genetic algorithm is used to adjust the robot motion to maintain the balance and keep motion similar to the taught human motion. We improve the GA with less learning time and the results show that the robot effectively imitates human robot and does not fall down.
Keywords :
cameras; genetic algorithms; humanoid robots; image capture; intelligent robots; legged locomotion; motion estimation; robot vision; GA; Kinect; RGB-D camera; body motion; genetic algorithm; human motion capture; humanoid robot motion imitation; learning time; robot balance control; robot motion editing tools; robot motion programming; Convergence; Genetic algorithms; Humanoid robots; Joints; Robot motion; Standards; RGB-D camera; Robot motion imitation; balance; genetic algorithm;
Conference_Titel :
Advanced Robotics and Intelligent Systems (ARIS), 2015 International Conference on
Conference_Location :
Taipei
DOI :
10.1109/ARIS.2015.7158363