Toward Designing a Robot that Learns Actions from Parental Demonstrations

Y. Nagai, C. Muhl, K. Rohlfing, in: The 2008 IEEE International Conference on Robotics and Automation, 2008, pp. 3545–3550.

Download
No fulltext has been uploaded.
Conference Paper | English
Author
; ;
Abstract
How to teach actions to a robot as well as how a robot learns actions is an important issue to be discussed in designing robot learning systems. Inspired by human parent-infant interaction, we hypothesize that a robot equipped with infant-like abilities can take advantage of parental proper teaching. Parents are known to significantly alter their infant-directed actions versus adult-directed ones, e.g. make more pauses between movements, which is assumed to aid the infants’ understanding of the actions. As a first step, we analyzed parental actions using a primal attention model. The model based on visual saliency can detect likely important locations in a scene without employing any knowledge about the actions or the environment. Our statistical analysis revealed that the model was able to extract meaningful structures of the actions, e.g. the initial and final state of the actions and the significant state changes in them, which were highlighted by parental action modifications. We further discuss the issue of designing an infant-like robot that can induce parent-like teaching, and present a human-robot interaction experiment evaluating our robot simulation equipped with the saliency model.
Keywords
Publishing Year
Proceedings Title
The 2008 IEEE International Conference on Robotics and Automation
Page
3545-3550
LibreCat-ID

Cite this

Nagai Y, Muhl C, Rohlfing K. Toward Designing a Robot that Learns Actions from Parental Demonstrations. In: The 2008 IEEE International Conference on Robotics and Automation. ; 2008:3545-3550. doi:10.1109/robot.2008.4543753
Nagai, Y., Muhl, C., & Rohlfing, K. (2008). Toward Designing a Robot that Learns Actions from Parental Demonstrations. In The 2008 IEEE International Conference on Robotics and Automation (pp. 3545–3550). https://doi.org/10.1109/robot.2008.4543753
@inproceedings{Nagai_Muhl_Rohlfing_2008, title={Toward Designing a Robot that Learns Actions from Parental Demonstrations}, DOI={10.1109/robot.2008.4543753}, booktitle={The 2008 IEEE International Conference on Robotics and Automation}, author={Nagai, Yukie and Muhl, Claudia and Rohlfing, Katharina}, year={2008}, pages={3545–3550} }
Nagai, Yukie, Claudia Muhl, and Katharina Rohlfing. “Toward Designing a Robot That Learns Actions from Parental Demonstrations.” In The 2008 IEEE International Conference on Robotics and Automation, 3545–50, 2008. https://doi.org/10.1109/robot.2008.4543753.
Y. Nagai, C. Muhl, and K. Rohlfing, “Toward Designing a Robot that Learns Actions from Parental Demonstrations,” in The 2008 IEEE International Conference on Robotics and Automation, 2008, pp. 3545–3550.
Nagai, Yukie, et al. “Toward Designing a Robot That Learns Actions from Parental Demonstrations.” The 2008 IEEE International Conference on Robotics and Automation, 2008, pp. 3545–50, doi:10.1109/robot.2008.4543753.

Export

Marked Publications

Open Data LibreCat

Search this title in

Google Scholar