Building an Affective Model for Social Robots with Customizable Personality

Abstract

As robot technology advances, there is a progressively growing demand to empower robots with human-like emotions. Designing appropriate personality features for social robots can enhance user contact with them as well as help them complete their works more effectively. In this paper, an affective model for social robots with adjustable personality is proposed. This model describes the changing process between emotional states using emotion, mood, and personality. Our model can convincingly mimic the emotional fluctuations of individuals with various personalities by using the results of experiments on real humans. First, we proposed a bowl-shaped affective model with eight fundamental feelings in three dimensions of emotional space, and then we provide a way for modeling personality. Second, we use the emotional arousal methods to contrast and examine how various personality groups responded to the same emotional stimulus. Based on this, we build two models that can fully capture introverted and extroverted personality traits, and then we use the validation data set to assess the validity and reliability of the parameters in our model. The results show that the model can effectively simulate the emotional change process of different personality groups after being subjected to varied emotional stimuli.