JP2001353679A - Robot for use as pet - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a robot for use as a pet that grows well enough for the robot user to have a stronger affection. SOLUTION: The user is verified by an iris verification system 4 and a behavioral pattern for the user is loaded from a user behavioral pattern memory 6; the robot behaves in accordance with the behavioral pattern and obtains information from sensors 2 and 3 so that a learning/growing system 5 varies the behavioral pattern for the user. The behavioral pattern for each user is saved in the memory 6.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a pet robot.

[0002]

2. Description of the Related Art In recent years, pet robots that imitate dogs, cats, and the like have been developed. The pet robot obtains information of the user's intention as information from the user via a sensor arranged on the body, for example, a touch sensor depending on the strength of the touch from the user, and information on the external world via a camera, for example, the color of the external world. It acquires information on shapes and shapes and learns / grows behavior patterns based on information obtained from sensors.

[0003]

In a conventional pet robot, even if it is the owner of the robot, the user is not authenticated. There is a problem in that the behavior pattern is shown, and especially the owner of the robot cannot feel that the growth is poor.

[0004] It is an object of the present invention to provide a pet robot that has a high growth potential and allows a user of the robot to have a stronger attachment.

[0005]

In order to achieve the above object, a pet robot according to the present invention is provided with personal authentication means for authenticating the user, and the user is authenticated by the personal authentication means and the authenticated user is used. An action pattern is obtained based on information on the user up to the previous time, and the action pattern for the user is changed based on information newly obtained from the sensor.

[0006]

Embodiments of the present invention will be described with reference to the drawings. Elements common to the drawings are denoted by the same reference numerals.

[0007] First Embodiment FIG. 1 a block diagram of a pet robot according to the first embodiment, FIG. 7 is an example view showing the arrangement of the iris authentication system. The pet robot 1 (hereinafter referred to as the robot 1) has a sensor 2, a sensor 3, an iris authentication system 4,
The system includes a learning / growth system 5, a user-specific behavior pattern memory 6, and a main system 7. The main system 7 controls the entire robot 1.

[0008] The sensor 2 receives information from the user 8 and outputs it to the main system 7. For example, by using a touch sensor as the sensor 2, the user 8 gives information to the main system 7 according to the strength of the touch.

When the output of the sensor 2 exceeds the threshold value, the main system 7 recognizes the information as malicious information (for example, rounding) for the robot. It is recognized as information that has a favorable impression (for example, stroking).

The sensor 3 acquires information on the outside world of the robot 1 and outputs the information to the main system 7. For example, by using a camera for the sensor 3, the main system 7 can acquire information on the color and shape of the outside world.

As shown in FIG. 7, the iris authentication system 4 extracts a characteristic part of the iris from an image of the eyes of the user 8 obtained through a photographing port 9a of a camera provided at the eyes, nose, etc. of the robot 1. Then, personal authentication is performed by comparing the registered data with the registered data, and output to the main system 7.

The learning / growth system is connected to the sensor 2 and the sensor 3 and the iris authentication system 4 via the main system 5.
Changes the behavior pattern of the robot based on the information obtained from.

The user-specific behavior pattern memory 6 inputs behavior pattern data from the main system 7 and stores the data for each user.

FIG. 2 is a block diagram of the iris authentication system. The video capture unit obtains an image of the eyes of the user 8 through the sensor 3 via the main system 7, digitizes the image, and outputs the digitized image to the image processing unit. The image processing unit cuts out an iris from the image input from the video capture unit.

The iris feature extraction unit extracts iris features from the iris image cut out by the image processing unit and outputs the iris features to the iris matching unit. The iris matching unit compares the registered iris data stored in the database with the registered iris data and outputs whether the user 8 has been authenticated (new user) to the main system 7 via the result output unit.

FIG. 3 is an operation flowchart of the pet robot shown in FIG. In step S1, the user 8 is authenticated through the iris authentication system 4. It is checked whether or not the user 8 has been authenticated. If the user 8 has been authenticated, the process branches to step S2. If not, the process branches to step S7. I do.

In step S2, an action pattern of the user 8 is loaded with reference to the user-specific action pattern memory 6, and an action corresponding to the action pattern is taken in step S3.

In step S4, information is acquired through the sensors 2 and 3, and in step S5, the behavior pattern is changed by the learning / growth system 5. In step S6, the changed behavior pattern is stored in the user-specific behavior pattern memory 6.
To save.

If the flow branches to step S7, since the user 8 is authenticated for the first time, a new user behavior pattern is created, and the flow proceeds to step S3.

According to the first embodiment, the provision of the iris authentication system enables the robot to take a behavior pattern for each user. For example, it is assumed that a robot imitates a dog, a user A gives information (for example, a stroke) that gives a favorable impression to the robot, and the behavior pattern becomes a favorable behavior pattern. This behavior pattern is stored in the user-specific behavior pattern memory together with the iris information of the user A. Next, when the robot authenticates the user as the user A, a favorable behavior pattern (eg, waving a tail) is shown.

On the other hand, it is assumed that a certain user B gives malicious information (for example, rounding) to the robot, and the behavior pattern becomes a malicious behavior pattern. This behavior pattern is stored in the user-specific behavior pattern memory together with the iris information of the user B. Next, when the robot authenticates the user as the user B, a malicious behavior pattern (eg, barking) is shown. As described above, by providing the iris authentication system, the robot can show a behavior pattern for each user.

The behavior pattern may be completely different for each user, or only a part of the behavior pattern may be different. For example, if it is completely different for each user, the degree of growth (for example, toddling to firm walking) also differs for each user. When the behavior pattern of one copy is different, the degree of growth is the same for each user, but the behavior (eg, wagging the tail, barking) is different.

Also, the user often faces the robot's face with his / her face, but by providing the camera with a camera opening for photographing the iris at the robot's eyes, nose, etc., the robot can take an unreasonable posture to the user. The user's iris can be photographed in a natural form without forcing.

Second Embodiment FIG. 4 is a block diagram of a pet robot according to a second embodiment . The difference from the pet robot according to the first embodiment is that the proximity / azimuth sensor 15 and the wide-angle sensor are different. The point is that a camera 16 and a user confirmation system 17 are added.

As shown in FIG. 5, the proximity / azimuth sensor 15 has proximity sensors 15a to 15d arranged at 90 ° intervals. The proximity of the user to the robot depends on which proximity sensor is responding. And its direction can be detected.

The wide-angle camera 16 acquires an image with a wide angle of view. If the sensor 3 is a camera, it may be shared with a wide-angle camera.

The user confirmation system 17 is a wide-angle camera 16
It is determined whether or not the user is reflected from the image acquired in step (1). One example of such determination is pattern matching.
The sensor 2 receives information from the user 8 and outputs the information to the main system 7. For example, by using a touch sensor as the sensor 2, the user 8 gives information to the main system 7 according to the strength of the touch.

The sensor 3 acquires information on the outside world of the robot 1 and outputs the information to the main system 7. For example, by using a camera for the sensor 3, the main system 7 can acquire information on the color and shape of the outside world.

The iris authentication system 4 extracts the feature of the iris from the image of the eyes of the user 8, performs personal authentication by comparing it with registered data, and outputs it to the main system 7.

The learning / growing system is connected to the sensor 2 and the sensor 3 and the iris authentication system 4
Changes the behavior pattern of the robot based on the information obtained from.

The user-specific behavior pattern memory 6 inputs behavior pattern data from the main system 7 and stores the data for each user.

The proximity / azimuth sensor 15 grasps that the user 8 has approached the robot 20 and its direction, and outputs it to the main system 7. The wide-angle camera 16 acquires an image with a wide angle of view and outputs the image to the main system 7. The user confirmation system 17 determines from the image acquired by the wide-angle camera 16 whether the user 8 is shown.

FIG. 6 is an operation flowchart of the pet robot shown in FIG. In step S10, the proximity of the user 8 is checked through the proximity / azimuth sensor 15, and if there is a response, the flow branches to step S11.

In step S11, the wide-angle camera 16 is moved
Point in the direction acquired from the direction sensor 15. Step S
In step 12, it is checked whether or not the user 8 exists through the user confirmation system 17. If the user 8 exists, the process branches to step S13. If not, the process branches to step S10.

In step S13, the iris authentication system 4
The authentication of the user 8 is performed through, and whether or not the user 8 has been authenticated is checked. If the user 8 has been authenticated, the process branches to step S14;
At this time, the image used for the iris authentication may be obtained by extracting the eyes from the image obtained by the wide-angle camera 16 or by calculating the eye position from the image of the wide-angle camera 16 and obtaining the eye image by another camera. good.

In step S14, the behavior pattern of the user 8 is loaded with reference to the behavior pattern memory 6 for each user,
In step S15, an action corresponding to the action pattern is taken.

In step S16, information is acquired through the sensors 2 and 3, and in step S17, the behavior pattern is changed by the learning / growth system 5, and in step S18.
Then, the changed behavior pattern is saved in the user-specific behavior pattern memory 6.

When the flow branches to step S19, since the user 8 is authenticated for the first time, a new user behavior pattern is created, and the flow proceeds to step S15.

According to the second embodiment, the proximity / azimuth sensor, the wide-angle camera, and the user confirmation system are newly provided, so that the robot can actively acquire the image of the eyes of the user. I can move. This eliminates the need for the user to perform special operations such as looking into the camera opening provided at the robot's eyes, nose, etc. in order to be authenticated. There is an effect that can be expected.

[0040]

The present invention is configured as described above and has the following effects.

A personal authentication means for authenticating a user is provided,
Authenticating the user by personal authentication means, taking action patterns based on the previous information for the authenticated user,
By changing the behavior pattern for the user based on the information newly obtained from the sensor, it is possible to provide a pet robot that has a high growth potential and allows the user of the robot to have a stronger attachment.

By using the iris authentication device for the personal authentication system, the authentication accuracy is higher than other authentication devices,
The iris can be actively acquired from the robot side without requiring the user to take any special action, and the user can be authenticated.

[Brief description of the drawings]

FIG. 1 is a block diagram of a pet robot according to a first embodiment.

FIG. 2 is a block diagram of an iris authentication system.

FIG. 3 is an operation flowchart of the pet robot shown in FIG. 1;

FIG. 4 is a block diagram of a pet robot according to a second embodiment.

FIG. 5 is a schematic diagram of a proximity / azimuth sensor.

6 is an operation flowchart of the pet robot shown in FIG. 4;

FIG. 7 is an example showing an arrangement of an iris authentication system.

[Explanation of symbols]

 1,20 Robot 4 Iris authentication system 15 Proximity / direction sensor

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 3F059 AA00 BA02 BB06 BC07 CA05 CA06 DA02 DA05 DA08 DB02 DB08 DB09 DC01 DC08 DD01 DD06 FA01 FA03 FA05 FA08 FA10 FB01 FC02 FC06 FC07 FC13 FC14 FC15 3F060 AA00 CA14 CA26 GA05 GA13 GD15 GD14 HA02 HA32 HA35

Claims (5)

[Claims] 1. A pet robot for learning / growing a behavior pattern by obtaining information from a sensor arranged on a body, comprising: a personal authentication unit for authenticating a user; A pet robot which takes an action pattern based on information up to the last time for an authenticated user and changes the action pattern for the user based on information newly obtained from the sensor. 2. The pet robot according to claim 1, wherein said personal authentication means is an iris authentication device. 3. The pet robot according to claim 2, wherein the iris authentication device has an imaging port of a camera provided in an eye of the robot. 4. The pet robot according to claim 2, wherein the iris authentication device has a photographing opening of a camera provided at a tip of the nose of the robot. 5. The pet robot according to claim 1, further comprising a proximity / azimuth sensor for detecting that the user is near and the direction of the user.