Robot and control method

• US 20020049515A1
• Filed: 12/19/2001
• Published: 04/25/2002
• Est. Priority Date: 05/10/1999
• Status: Active Grant

• Current Assignee: Hiroshi Osawa
• Original Assignee: Hiroshi Osawa
• Inventors: Osawa, Hiroshi
• Granted Patent: US 6,512,965 B2
• Time in Patent Office: Days
• US Class Current: 700/245
• CPC Class Codes: A63H 2200/00 Computerized interactive toys, e.g. dolls
• CPC Class Codes: A63H 3/28 Arrangements of sound-producing means in dolls; Means in dolls for producing sounds
• CPC Class Codes: G06N 3/008 based on physical entities controlled by simulated intelligence so as to replicate intelligent life forms, e.g. based on robots replicating pets or humans in their appearance or behaviour

Abstract

With a robot and a control method for it, information is read in from the outside, based on which a particular object is detected, and it is judged upon detecting the object whether or not the object satisfies given conditions, based on the judgment results of which a robot generates predetermined actions. Thus, a robot can be realized that acts naturally like a living thing. Consequently, its entertaining quality for human beings can be increased greatly.

Claims

1. A robot comprising and characterized by:

an external information capturing means for reading in information from the outside,

a detection means for detecting a particular object based on said information captured by said external information capturing means,

a judging means for judging whether or not the object satisfies given conditions when said object is detected by said detection means,

an action generation means for generating a given action based on the results obtained by said judging unit.

2. A robot as defined in the claim 1 characterized by:

said action generating means for generating said action of approaching toward said object detected by means of said object detecting means when said object satisfies said conditions.

3. A robot as defined in the claim 2, wherein:

said action generation means comprising, a condition recognition unit for recognizing said external condition based on said information captured by said external information capturing unit,

an emotion modeling unit for keeping a parameter value of the intensity of each emotion of at least ‘

Joy’ and

‘

Dislike’ and

for varying said parameter value of said emotions based on said judgment results at said condition recognition unit,

an action generation unit for generating said action based on said parameter value indicating judgment results obtained at said condition recognition unit and the intensity of each of said emotion at said emotion modeling unit,

an altering unit for altering setting of said emotion modeling unit, if need be, based on the judgment results at said judgment unit, and

said alteration means alters a setting of said emotion modeling means, in case said object satisfies said conditions so that it is easier to increase said parameter value of said intensity of an emotion ‘

joy’

, and alters a setting of said emotion modeling means, in case said object does not satisfy said conditions so that it is easier to increase said parameter value of said intensity of an emotion ‘

Dislike’

.

4. A robot as defined in the claim 2, wherein:

said generating means, comprising;

a condition recognizing unit for recognizing said external conditions based on said information captured by said external information capturing unit,

an action determining unit for determining said subsequent action with probability based on the recognition results obtained at said condition recognition unit,

an action generation unit for generating said action determined by said action determining unit, and,

an alteration unit for altering said setting of the action determining unit as required, based on said judgment results obtained by said judgment means, and

said alteration unit for altering the setting of said action determining unit to increase the selection probability for said action expressing ‘

Joy’

if said object satisfies said conditions, and for altering the setting of said action determining unit to increase the selection probability for said action expressing ‘

Dislike’

if said object does not satisfy said conditions.

5. Control method of a robot comprising:

the first step for capturing external information and detecting a particular object based on said information,

the second step for making judgment on whether or not said object satisfies given conditions when said object is detected, and

the third step for making a robot which generates given actions based on said judgment results.

6. Control means for a robot as defined in the claim 5 wherein;

said third step generates said action to approach said object if said object satisfies said conditions, and generates said action to go away from said object if said object does not satisfies said conditions.

7. Control method of a robot as defined in the claim 6 wherein:

said robot comprising;

a condition recognizing unit for recognizing said external condition based on said information read in by said external information capturing unit,

an emotion modeling unit for retaining the intensity of each emotion of at least ‘

Joy’ and

‘

Dislikes’

as parameter value and for varying said parameter values of said emotions based on recognition results at said condition recognition unit, and an action generation unit for generating said actions based on said parameters indicating the recognition results at said condition recognition unit and the intensity of each of said emotions at said emotion modeling unit, and

said third step;

alters the setting of said emotion modeling unit to make it easier to increase said parameter value indicating the intensity of said emotion ‘

Joy’

, if said object satisfies said conditions while, if said object does not satisfy said conditions, alters the setting of said emotion modeling unit to make it easier to increase said parameter value indicating the intensity of said emotion ‘

Dislike’

.

8. Control method of a robot system as defined in the claim 6, wherein:

said robot comprising;

a condition recognition unit for recognizing said external condition based on said information captured by means of said external information read-in means;

an action determining unit for deciding upon the subsequent action with probability, based on the recognition results obtained at said condition recognition unit, and

an action generation means for generating said action determined by said action determining unit, and

the third step;

alters the setting of said action determining to increase the selection probability of said action expressing ‘

joy’

if said object satisfies said conditions, whereas, if said object does not satisfy said conditions, alters the setting of said action determining means to increase the selection probability expressing ‘

Dislike’

.