| dc.description.abstract | The main purpose of this dissertation is to study the affective model of robot interaction. After simulating human interaction, the simulation model will further apply the development of the emotion-based robot by receiving the stimulus event and the emotional change after the stimulus event. In order to simulate rich and varied personality, this paper proposes to use the constellation as a basis for classification personality, plus the original Favorability, Attitude, Mood, Personality and Emotion (Emotion). The constellation complex emotion model (FAMPE-C Model). The model uses different personality as the core parameters. The three main emotions: mood, emotion, and attitude influence each other and are subject to different personality parameters. The main emotions are divided into more delicate branches of emotions, such as happy and sad. Branches are presented in a quantitative way for observation. Goodwill is an index of goodwill after interaction between the robot and the robot. After adding goodwill, you can observe the goodwill values of both parties after the interaction, and it will affect both the ups and downs of the likes and dislikes of both parties.
Regarding the verification of the constellation compound emotion model, the MATLAB R2017a was used as a simulation tool to verify the accuracy of the model, including pre-interaction emotions, post-interaction emotions, favorability simulations, and the simulation results showed that all three simulations were composited with the constellation. The definition of the emotional model is consistent. | en_US |