Emergence and Categorization of Coordinated Visual Behavior ThroughEmbodied Interaction

Authors:
Luc Berthouze;Yasuo Kuniyoshi
Affiliations:
Electrotechnical Laboratory(ETL), AIST, MITI Intelligent Systems Division Umezono 1-1-4, Tsukuba, Ibaraki 305, Japan. E-mail: {berthouz,kuniyosh}@etl.go.jp;Electrotechnical Laboratory(ETL), AIST, MITI Intelligent Systems Division Umezono 1-1-4, Tsukuba, Ibaraki 305, Japan. E-mail: {berthouz,kuniyosh}@etl.go.jp
Venue:
Machine Learning - Special issue on learning in autonomous robots
Year:
1998

Citing 7
Cited 3

Action

Foundations of cognitive science
Real-time gaze holding in binocular robot vision

Real-time gaze holding in binocular robot vision
Neural network control for a closed-loop System using Feedback-error-learning

Neural Networks
Active and exploratory perception

CVGIP: Image Understanding - Special issue on purposive, qualitative, active vision
Driving saccade to pursuit using image motion

International Journal of Computer Vision
Active Stereo Vision System with Foveated Wide Angle Lenses

ACCV '95 Invited Session Papers from the Second Asian Conference on Computer Vision: Recent Developments in Computer Vision
Velocity and Disparity Cues for Robust Real-Time Binocular Tracking

CVPR '97 Proceedings of the 1997 Conference on Computer Vision and Pattern Recognition (CVPR '97)

A Neural Structure for Learning by Imitation

ECAL '99 Proceedings of the 5th European Conference on Advances in Artificial Life
Incremental learning of sequence patterns with a modular network model

Neurocomputing
A biologically constrained architecture for developmental learning of eye---head gaze control on a humanoid robot

Autonomous Robots

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper discusses the emergence of sensorimotor coordination forESCHeR, a 4DOF redundant foveated rob ot-head, by interaction with itsenvironment. A feedback-error-learning(FEL)-based distributed controlprovides the system with explorative abilities with reflexes constrainingthe learning space. A Kohonen network, trained at run-time, categorizes thesensorimotor patterns obtained over ESCHeR‘s interaction with itsenvironment, enables the reinforcement of frequently executed actions, thusstabilizing the learning activity over time. We explain how the developmentof 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’.