Evolutionary Simulations based on a Robotic Approach to Emotion

Abstract This paper reports on the current state of our efforts to synthesize emotion in robots from a selectionist perspective in order to explore the origin and the adaptations of emotion and to realize efficient robotic systems. We have defined a minimal model of emotion in robots based on a behavioristic theory on human emotions, and have shown by conducting a robotic experiment that human beings can identify a set of basic emotions in robots. In this paper, we conduct evolutionary simulations based on the definition of emotions to verify the evolutionary adaptivity with the scenario of the origin and the evolution of human emotions in mind. The simulations show that robots are evolved to move effectively in the environments with various physical conditions by changing their emotions corresponding to the type of the condition they encounter.