Emotional and Conditional Model for Pet Robot based on Neural Network

Abstract

Recently more and more pet robot products are launched, showing the increasing market demand for pet robots. A pet robot has a healing effect for suitably developing companionship with people. A plenty of researches pointed out that the interaction between human and robot is important. This paper introduces the concentration and conditioning to model the emotion and behavior of a pet robot. For the overall system of this pet robot, a condition matrix is properly designed to support the pet robot for the adaption of several kinds of stimuli and response about behaviors. The behavior controller is achieved by the means of a neural network with learning mechanism, which can be trained for the behavioral control of voice, expression and action. All these mechanisms are established by psychological principles, accordingly the pet robot can behave naturally.