Long Short-Term Memory for Affordances Learning

Sergio Roa; Geert-Jan Kruijff

In: Lola Cañamero; Pierre-Yves Oudeyer; Christian Balkenius (Hrsg.). Proceedings of the 9th International Conference on Epigenetic Robotics. International Conference on Epigenetic Robotics (EpiRob-09), Modeling Cognitive Development in Robotic Systems, November 12-14, Venice, Italy, Pages 235-236, Lund University Cognitive Studies, No. 146, ISBN 978-91-977-380-7-1, 11/2009.


This paper addresses the problem of sensorimotor learning from the perspective of affordances learning of simple objects. We are developing a scenario where a robotic arm interacts with a polyflap, a simple 3-dimensional geometrical object. We perform experiments with a simulated arm using a physics simulator, but we plan to use also a real arm. The robot interacts with the object by pushing it in different ways. We use Recurrent Neural Networks to predict the arm and object poses during this interaction, given a discrete set of random actions that the robot can produce.


RoaKruijffEPIROB09.pdf (pdf, 153 KB )

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence