Perceptual Control Based on Prediction for Natural Communication of a Partner Robot.

This paper discusses a perceptual system for natural communication between a partner robot and a human. The prediction is very important to reduce the computational cost and to extract the perceptual information for the natural communication with a human in the future. First, we propose a prediction-based perceptual control system based on spiking neurons. The proposed method is composed of four layers, namely: 1) input layer; 2) clustering layer; 3) prediction layer; and 4) perceptual module selection layer. Next, we propose an unsupervised learning method to perform the clustering of human behavior patterns. Furthermore, the robot selects perceptual modules used in the next perception according to the predicted perceptual mode. Furthermore, we show several experimental results of the communication between a partner robot and a human based on our proposed method. [ABSTRACT FROM AUTHOR]