Proceedings of the Tenth International Conference on Epigenetic Robotics , Lund, İsveç, 5 - 07 Kasım 2010, ss.1-8
In this paper, we used the affordance formalization framework (Sahin et al; Adaptive Behavior, 15(4), 447-472, 2007) to link the concepts represented by verbs and nouns in language to the affordance relations that a robot
acquires through its interaction with the environment. First, we argued that a verb is
linked not to a specific behavior of the robot,
but to a specific effect that different behaviors
may generate. We showed how demonstrations made by a human can allow the robot to
learn an effect prototype in its own sensorymotor space, and use this to command the
robot to achieve desired goals. Second, we
investigated how the affordance relations of
objects can be used to create noun concepts
(to name objects) and how multi-task learning
can utilize the learned object concepts to facilitate the learning of new affordances. The results are demonstrated on an iCub humanoid
robot platform.