Objective

Humans anthropomorphize anything that they interact with. An interdisciplinary research framework in the intersection between engineering and science to reveal the principles of the intelligent machines that communicate with humans based on this human nature is called android science. These studies can be proceeded by evaluating the intelligence from the viewpoint of influences for the interaction with humans or for the human society. In other words, robots can be developed through being introduced into the society, establishing complex relationships in it, and being evaluated.

(a) Human mechanism for participating society is examined through implementing human-like functions into socially developing humanoid robot. For this purpose, we have developed child robot.

(b) Through experiments with human subjects and socially developmental humanoid robots that we built, how rich social relationships they can establish is tested. Using infant type socially developmental humanoid robot, contingency model is constructed and mechanisms for communication is tested. Especially, robot mechanism to participate society through sharing words, mechanisms to joint attention during social task and to establish social relationship should be considered. In addition to the approach concerned in perso-synergistic intelligence, we focused on the response of subjects (ex. impression, unconscious responsive movement, and so on) as performance measures.

(c) Hypotheses of the basic mechanism of human social behavior towards understanding human are derived through building socially developing humanoid robot that acquire sociality.

Socially developmental modeling is examined by simulating sociality formation (including cooperation, competition, and role-assignment) in multi-agents.

(i) Examining pattern generation and changes of team behavior using humanoid robots through learning and development of cooperative and competitive behavior in the team formation.

(ii) Analyzing patterns of sociality formation among multi agents in the computer simulation

(iii) Modeling mechanisms of generating intelligence in group to reveal developmental mechanism for sociality formation as the process of interaction among multi-agents through these experiments.