Chaos-Entropy Analysis and Acquisition of Individuality and Proficiency of Human Operator's Skill Using a Neural Controller

Abstract

The emergence of intelligence in an autonomous robot exists in the dexterity of humans or creatures as complex systems and research and development procedures along this approach seems necessary for realization of an intellectual robot. However, although strict judgment is required during stabilizing control of an unstable system, such as an inverted pendulum on a cart by human operators, it is assumed that human operators exhibit complex behavior intermittently. A previous paper investigated the skill of a human operator and investigated the formation of a complex system in the learning process of human operators with objects difficult to control. It also considered the mechanism of robustness of human operators against such a disturbance. The current paper shows that the neural network controller identified from time series data of each trial of several operators exhibits the human-generated decision-making characteristics with the chaos and a large amount of disorder. It also confirms that the estimated degrees of freedom of motion increases and the estimated amount of disorder decreases with an increase of proficiency. In addision, this paper shows that the agreement between the neural control simulation and the experimental results of neural control for the degrees of freedom of motion and the entropy ratio is particularly good when the simulated wave form and the measured wave form are similar in appearance.