Emergence and Categorization of Coordinated Visual Behavior Through Embodied Interaction
Abstract
This paper discusses the emergence of sensorimotor coordination for ESCHeR, a 4DOF redundant foveated rob ot-head, by interaction with its environment. A feedback-error-learning(FEL)-based distributed control provides the system with explorative abilities with reflexes constraining the learning space. A Kohonen network, trained at run-time, categorizes the sensorimotor patterns obtained over ESCHeR's interaction with its environment, enables the reinforcement of frequently executed actions, thus stabilizing the learning activity over time. We explain how the development of ESCHeR's visual abilities (namely gaze fixation and saccadic motion), from a context-free reflex-based control process to a context-dependent, pattern-based sensorimotor coordination can be related to the Piagetian ‘stage theory’.
Cite
Text
Berthouze and Kuniyoshi. "Emergence and Categorization of Coordinated Visual Behavior Through Embodied Interaction." Machine Learning, 1998. doi:10.1023/A:1007453010407Markdown
[Berthouze and Kuniyoshi. "Emergence and Categorization of Coordinated Visual Behavior Through Embodied Interaction." Machine Learning, 1998.](https://mlanthology.org/mlj/1998/berthouze1998mlj-emergence/) doi:10.1023/A:1007453010407BibTeX
@article{berthouze1998mlj-emergence,
title = {{Emergence and Categorization of Coordinated Visual Behavior Through Embodied Interaction}},
author = {Berthouze, Luc and Kuniyoshi, Yasuo},
journal = {Machine Learning},
year = {1998},
pages = {187-200},
doi = {10.1023/A:1007453010407},
volume = {31},
url = {https://mlanthology.org/mlj/1998/berthouze1998mlj-emergence/}
}