Clustering of Internal State of A Partner Robot based on Modular Neural Networks

抄録

Imitation is a powerful tool for behavior learning and human communication. Basically, imitative learning is composed of model observation and model reproduction. And this paper divides model reproduction into learnig and clustering of gestures. So this paper applies a steady-state genetic algorithm for model observation, modular neural networks and a steady-state genetic algorithm for learing of gestures, and the learning state of the modular neural networks for clustering of gestures. The proposed method is applied for a partner robot interacting with a human. Experimental results show that the proposed method enables a robot to learn behaviors through and can interact with a human efficiently.