Abstract:
An apparatus and a method for creating combined artificial feelings to which different basic feelings like a human being's combined feeling are reflected. The apparatus for creating an artificial feeling comprises a feeling value group creation part which is configured to create a feeling value group having a feeling value which gradually decreased as it gets farther from a basic feeling assigned to a machine apparatus on an internal state coordinate system of a machine apparatus; and a feeling creation part which is configured to create, as a feeling of a machine apparatus, a group of a feeling value of each feeling value group that a coordinate of an internal state input value of the machine apparatus indicates on the internal state coordinate system.
Abstract:
An apparatus for selecting a motion signifying artificial feeling is provided. The apparatus includes: an feeling expression setting unit configured to set probabilities of each feeling expression behavior performed for each expression element of a robot for each predetermined feeling; a behavior combination generation unit configured to generate at least one behavior combination combined by randomly extracting the feeling expression behaviors in each expression element one by one; and a behavior combination selection unit configured to calculate an average for the probabilities of the feeling expression behaviors included in each behavior combination for each feeling of a robot and select behavior combinations in which the average of the probabilities of the feeling expression behaviors most approximates the predetermined feeling value of a robot from each behavior combination.
Abstract:
An optimal behavior combination of a robot in consideration of the robot's previous feeling is generated. In other words, the robot may express its feelings by combinations of its behavioral express elements which reflect its previous feelings. In particular, it has a continuity determination unit for determining whether a behavior to be expressed should be same as or different from a currently expressed behavior.
Abstract:
An apparatus for the learning emotion of a robot includes an internal state calculation unit for detecting external environment information and calculating an internal state vector value for one point of an internal state coordinate system; an emotion decision processing unit for generating an emotion value within each emotion value group, indicated by the coordinates of the internal state input value of the robot on the coordinate system, as an emotion of the robot; a behavior generation unit for deciding the behavior of the robot based on the decided motion value and making an emotional expression; and an emotion learning unit for generating feedback information on the emotion value, corresponding to one point of the coordinate system, based on the decided emotion value and information on the type of emotion from a user input unit and changing an emotion probability distribution of the coordinate system using the feedback information.
Abstract:
An apparatus for selecting a motion signifying artificial feeling is provided. The apparatus includes: an feeling expression setting unit configured to set probabilities of each feeling expression behavior performed for each expression element of a robot for each predetermined feeling; a behavior combination generation unit configured to generate at least one behavior combination combined by randomly extracting the feeling expression behaviors in each expression element one by one; and a behavior combination selection unit configured to calculate an average for the probabilities of the feeling expression behaviors included in each behavior combination for each feeling of a robot and select behavior combinations in which the average of the probabilities of the feeling expression behaviors most approximates the predetermined feeling value of a robot from each behavior combination.
Abstract:
An apparatus and a method for creating combined artificial feelings to which different basic feelings like a human being's combined feeling are reflected. The apparatus for creating an artificial feeling comprises a feeling value group creation part which is configured to create a feeling value group having a feeling value which gradually decreased as it gets farther from a basic feeling assigned to a machine apparatus on an internal state coordinate system of a machine apparatus; and a feeling creation part which is configured to create, as a feeling of a machine apparatus, a group of a feeling value of each feeling value group that a coordinate of an internal state input value of the machine apparatus indicates on the internal state coordinate system.
Abstract:
An apparatus for the learning emotion of a robot includes an internal state calculation unit for detecting external environment information and calculating an internal state vector value for one point of an internal state coordinate system; an emotion decision processing unit for generating an emotion value within each emotion value group, indicated by the coordinates of the internal state input value of the robot on the coordinate system, as an emotion of the robot; a behavior generation unit for deciding the behavior of the robot based on the decided motion value and making an emotional expression; and an emotion learning unit for generating feedback information on the emotion value, corresponding to one point of the coordinate system, based on the decided emotion value and information on the type of emotion from a user input unit and changing an emotion probability distribution of the coordinate system using the feedback information.