DocumentCode :
565648
Title :
Incremental learning of gestures by imitation in a humanoid robot
Author :
Calinon, Sylvain ; Billard, Aude
Author_Institution :
LASA Lab., Ecole Polytech. Fed. de Lausanne (EPFL), Lausanne, Switzerland
fYear :
2007
fDate :
9-11 March 2007
Firstpage :
255
Lastpage :
262
Abstract :
We present an approach to teach incrementally human gestures to a humanoid robot. By using active teaching methods that puts the human teacher “in the loop” of the robot´s learning, we show that the essential characteristics of a gesture can be efficiently transferred by interacting socially with the robot. In a first phase, the robot observes the user demonstrating the skill while wearing motion sensors. The motion of his/her two arms and head are recorded by the robot, projected in a latent space of motion and encoded probabilistically in a Gaussian Mixture Model (GMM). In a second phase, the user helps the robot refine its gesture by kinesthetic teaching, i.e. by grabbing and moving its arms throughout the movement to provide the appropriate scaffolds. To update the model of the gesture, we compare the performance of two incremental training procedures against a batch training procedure. We present experiments to show that different modalities can be combined efficiently to teach incrementally basketball officials´ signals to a HOAP-3 humanoid robot.
Keywords :
Gaussian processes; gesture recognition; humanoid robots; learning (artificial intelligence); sensors; GMM; Gaussian mixture model; HOAP-3 humanoid robot; active teaching methods; batch training procedure; imitation gestures; incremental learning; kinesthetic teaching; motion sensors; robots learning; Abstracts; Actuators; Manipulators; RNA; Gaussian Mixture Model; Imitation Learning; Incremental learning; Programming by Demonstration;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Human-Robot Interaction (HRI), 2007 2nd ACM/IEEE International Conference on
Conference_Location :
Arlington, VA
ISSN :
2167-2121
Print_ISBN :
978-1-59593-617-2
Type :
conf
Filename :
6251697
Link To Document :
بازگشت