摘要

In this paper, we propose a robot behavior generation scheme that generates behaviors based on the senses of “curiosity”and “boredom”, to create a robot named SELF with which humans do not become bored. The scheme was developed using a simple but unique simulation environment. A red ball and a blue ball are displayed on the computer, and the human subject moves the red ball with an analog joystick. The behavior patterns under natural interhuman interaction are classified by having another human remotely controlling the blue ball from a networked PC. Next, a behavior generator is designed based on these behavior patterns, utilizing conditional probability. The autonomous learning capability of SELF is realized by updating this conditional probability. Our previous studies have revealed that interaction capability and animal-like speed are required for a companion robot. In addition, this paper shows how SELF can have a sense of curiosity and boredom as seen in animals, including humans, by using the developed scheme. Lastly, impressions of the robot behavior generated by the proposed scheme are examined using the simulationenvironment where the blue ball is controlled by the proposed scheme.