Brain information output apparatus, robot, and brain information output method

摘要

A brain information output apparatus includes an intention determination information storage unit in which two or more pieces of intention determination information can be stored, with each intention determination information including a pair of an intention identifier, and a learning feature amount group including one or more feature amounts extracted from second learning data that is obtained by converting first learning data into intracerebral brain activity data, the first leaning data being acquired from the outside of the cranium of a user when the user performs a trial according to one intention; a first brain activity data acquiring unit that acquires first brain activity data from the outside of the cranium of a user; a second brain activity data acquiring unit that converts the first brain activity data to intracerebral brain activity data, and acquires second brain activity data; a feature amount group acquiring unit that acquires, from the second brain activity data, an input feature amount group including one or more feature amounts; an intention identifier acquiring unit that acquires an intention identifier corresponding to the input feature amount group based on the two or more pieces of intention determination information; and an intention identifier output unit that outputs the intention identifier.

主权项

1. A brain information output apparatus, comprising:
an intention determination information storage unit which stores two or more pieces of intention determination information, each piece of intention determination information including i) an intention identifier that is information for identifying an intention of a user, and ii) two or more types of learning feature amount groups that respectively derive from two or more types of first learning data, each type of the first learning data being data of a brain activity acquired from the outside of the cranium of a user when the user performs a trial, which is a series of actions, according to an intention identified by the intention identifier, wherein each learning feature amount group includes one or more feature amounts extracted from second learning data that is obtained by converting the first learning data into intracerebral brain activity data; a first brain activity data acquiring unit, which acquires two or more types of first brain activity data, which is data of a brain activity, from the outside of the cranium of a user; a second brain activity data acquiring unit, which converts each of the two or more types of first brain activity data acquired from the outside of the cranium of a user to intracerebral brain activity data, and acquires two or more types of second brain activity data a feature amount group acquiring unit, which performs signal processing on each of the two or more types of second brain activity data, and acquires input feature amount groups including one or more feature amounts; and an intention identifier acquiring unit, which comprises: a probability calculation section, which, for each type of the learning feature amount groups included in the intention determination information, and for each intention identifier, calculates a probability that the input feature amount group acquired by the feature amount group acquiring unit corresponds to each of two or more intention identifiers included in the two or more pieces of intention determination information in the intention determination information storage unit, using the two or more types of learning feature amount groups, for each of the two or more intention identifiers, and which also calculates an ultimate probability for each intention identifier by using the probabilities calculated; and an intention identifier acquiring section, which acquires an intention identifier that corresponds to a largest probability of the ultimate probabilities obtained for each of the intention identifiers calculated by the probability calculation section.