Simulating Emotion and Personality for Intelligent Agent

Abstract

Synthesizing the reasonable and natural emotion is important to achieve a more effective and believable human-computer interaction. The paper presents an emotion and personality model based on HMM. The emotion process is regarded as a double random process in the model. Different personality traits of intelligent agent can be built by adjusting the parameters of the model. At the same time, as emotion engine, the model can generate corresponding emotion states to changes of stimulus under the different personality circumstances and can be applied to emotion robots or other intelligent agent.