Publication
Dynamically consistent probabilistic model for robot motion learning
Conference Article
Conference
IROS Workshop on Learning and Interaction in Haptic Robots (IROS LIHR)
Edition
2012
Pages
1-2
Doc link
http://www.iit.it/images/stories/advanced-robotics/Workshops/Iros_2012/Pardo-IROS2012WS.pdf
File
Authors
Projects associated
Abstract
This work presents a probabilistic model for learning robot tasks from human demonstrations using kinesthetic teaching. The difference with respect to previous works is that a complete state of the robot is used to obtain a consistent representation of the dynamics of the task. The learning framework is based on hidden Markov models and Gaussian mixture regression, used for coding and reproducing the skills. Benefits of the proposed approach are shown in the execution of a simple self-crossing trajectory by a 7-DoF manipulator.
Categories
learning (artificial intelligence), manipulators, robot dynamics.
Author keywords
learning from demonstration, kinesthetic teaching, Gaussian mixture regression (GMR), hidden Markov models (HMM)
Scientific reference
D. Pardo, L. Rozo, G. Alenyà and C. Torras. Dynamically consistent probabilistic model for robot motion learning, 2012 IROS Workshop on Learning and Interaction in Haptic Robots, 2012, Algarve, Portugal, pp. 1-2.
Follow us!