Data processing: generic control systems or specific application – Specific application – apparatus or process – Robot control
Reexamination Certificate
2002-03-13
2003-11-18
Cuchlinski, Jr., William A. (Department: 3661)
Data processing: generic control systems or specific application
Specific application, apparatus or process
Robot control
C700S246000, C700S248000, C700S250000, C700S259000, C701S023000, C701S301000, C318S568120, C901S001000
Reexamination Certificate
active
06650965
ABSTRACT:
TECHNICAL FIELD
The present invention generally relates to a robot apparatus behavior deciding method and a robot apparatus, and more particularly to an autonomous robot apparatus and a method for deciding the behavior of the robot apparatus.
BACKGROUND ART
Recently, there have been proposed robot apparatuses shaped each like an animal, namely, so-called pet robots. Each of such robot apparatuses is shaped like a dog or cat kept in a common family, and autonomously behaves in response to the action such as “hitting” or “patting” by the user (owner) and adaptively to its surrounding environment. For example, its autonomous behavior includes “yelping”, “mewing”, “sleeping”, etc. similar to the behavior of an actual animal.
If a robot apparatus could behave more similarly to an actual pet animal, it has an improved likeness to a living thing or a creature and the user will feel more familiar and satisfactory with such a robot apparatus. The robot apparatus will amuse the user thereof more than ever.
To make the robot apparatus behave like an actual animal, it has been proposed to use an ethological approach for decision of the robot apparatus behavior.
For example, as a result of the behavior study with the ethological approach, a state of motivation space representation was disclosed in the paper by Sibly, Mcfarland et al. (ethologists) in 1975. Also, Ludlow disclosed competitive models of behavior in 1976. These results were argued in the “Old Tricks, New Dogs: Ethology and Interactive Creatures” (April, 1997) by Bruce Mitchell Blumberg (Bruce of Arts, Amherst College, 1977; Master of Sciences, Sloan School of Management, MIT, 1981). Bruce Mitchell Blumberg applied the above-mentioned theories to dogs created by 3D CG (computer graphics) and proved the above theories to be a behavior selection mechanism.
It should be reminded that Bruce Mitchell Blumberg verified the behavior selection mechanism of the animals using CG, not by applying the mechanism to any robot apparatuses existing in the substantial space.
For a computer-graphically created virtual creature displayed on a display screen of a computer system, it is possible to make a direct coupling between selection and apparition of behavior (behavior selection=behavior apparition) and so feed back the behavior to its motivation by the selection. For an actual robot apparatus, however, the selection and apparition of behavior cannot always be coupled directly to each other (namely, the behavior selection is not always equal to the behavior apparition) for the following reasons.
Selected behavior is possibly canceled by behavior effected irrespective of schemed behavior such as reflexive one.
Without an input from a sensor, it cannot be known whether behavior could have really been done.
An example to which the reason described just above is applicable will be given below. That is, even when action to “kick the ball with the foot” is selected when the robot has reached a distance at which it can kick the ball and a behavior command is output (given to the robot), the robot apparatus cannot kick the ball in some cases, for example, if the ball lies on a slope. A result that “the ball could successfully be kicked”, of the action to “kick the ball with the foot”, can only be recognized when it has been recognized that the robot apparatus has touched the ball and the ball has been moved forward. Namely, for this recognition, it is necessary to evaluate the behavior based on information supplied from a sensor included in the robot apparatus and change the internal state of the robot apparatus according to the result of the evaluation.
As seen from the above, the technique proposed by Bruce Mitchell Blumberg is not enough to decide the behavior of a robot apparatus existing in the substantial space.
DISCLOSURE OF THE INVENTION
Accordingly, the present invention has an object to overcome the above-mentioned drawbacks of the prior art by providing a robot apparatus having an improved likeness to a living thing or a creature and a method for deciding the behavior of the robot apparatus.
The above object can be attained by providing a robot apparatus whose moving parts are controlled to make the robot apparatus behave expressively, the device including:
means for detecting external or internal information;
means for acquiring a cause factor influencing the behavior from the external or internal information detected by the information detecting-means;
means for acquiring an occurrence tendency of the cause factor-influenced behavior based on the cause factor acquired by the cause factor-acquiring-means;
means for making a comparison among occurrence tendencies of two or more behaviors, acquired by the occurrence tendency acquiring means and belonging to the same group;
means for selecting one of the behaviors based on the result of the occurrence tendency comparison made by the occurrence tendency comparing means; and
means for controlling the moving parts based on the behavior selected by the behavior selecting means to have the robot apparatus express the selected behavior;
the occurrence tendency of the behavior selected by the behavior selecting means being varied adaptively to the cause factor which is variable due to the actual occurrence of the behavior.
In the robot apparatus constructed as above, external or internal information is detected by the information detecting-means, a cause factor influencing the behavior is acquired by the cause factor acquiring means from the external or internal information detected by the information detecting-means, and an occurrence tendency of the cause factor-influenced behavior is acquired by the occurrence tendency acquiring means based on the cause factor acquired by the cause factor acquiring means.
A comparison is made by the occurrence tendency comparing means among occurrence tendencies of two or more behaviors, acquired by the occurrence tendency acquiring means and belonging to the same group, one of the behaviors is selected by the behavior selecting means based on the result of the occurrence tendency comparison made by the occurrence tendency comparing means, and the moving parts are controlled by the moving part controlling means based on the behavior selected by the behavior selecting means to have the robot apparatus express the selected behavior. The occurrence tendency of the behavior selected by the behavior selecting means is varied adaptively to the cause factor which is variable due to the actual occurrence of the behavior.
The above robot apparatus selects one of the behaviors through a comparison between occurrence tendencies decided under the influence of the cause factor, and expresses the behavior as an ethological approach.
Also the above object can be attained by providing a method for deciding the behavior of a robot apparatus whose moving parts are controlled to have the robot apparatus behave expressively, the method including the steps of:
detecting external or internal information of the robot by an information detecting-means;
acquiring a cause factor influencing the behavior from the external or internal information detected in the information detecting step;
acquiring an occurrence tendency of the cause factor-influenced behavior based on the cause factor acquired in the cause factor acquiring step;
making a comparison among occurrence tendencies of two or more behaviors, acquired in the occurrence tendency acquiring step and belonging to the same group;
selecting one of the behaviors based on the result of the occurrence tendency comparison made in the occurrence tendency comparing step; and
controlling the moving parts based on the behavior selected in the behavior selecting step to have the robot apparatus express the selected behavior;
the occurrence tendency of the behavior selected in the behavior selecting step being varied adaptively to the cause factor which is variable due to the actual occurrence of the behavior.
In the above robot apparatus behavior deciding method, external or internal information is detected in the information detecting
Arkin Craig Ronald
Fujita Masahiro
Hasegawa Rika
Sabe Kotaro
Takagi Tsuyoshi
Cuchlinski Jr. William A.
Frommer William S.
Frommer & Lawrence & Haug LLP
Marc McDieunel
Smid Dennis M.
LandOfFree
Robot apparatus and behavior deciding method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Robot apparatus and behavior deciding method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Robot apparatus and behavior deciding method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3135177