JP2003181784A - Automatic answering robot - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an automatic answering robot which is able to increase and maintain an intimate feeling between a user and the robot by giving variety to a manner for changing feelings in the robot. <P>SOLUTION: The automatic answering robot is provided with user information inputting means for inputting user information, storage means for storing information about user's gesture as a pattern of a gesture of the robot and also for storing a unique pattern of a gesture due to internal information, movement control means for processing the patterns by comparing the respective gesture patterns stored in the storage means and the patterns corresponding to the user information inputted in the user information inputting means, and regeneration means for regenerating any of gesture patterns based on the output from the movement control means. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic response robot, and in particular, for example, to enter a living space of a human (user) and respond with a behavior like a pet to the user to give a feeling of familiarity to the user. The present invention relates to an automatic response robot.

[0002]

2. Description of the Related Art Conventionally, in a robot having an emotion, for example, by changing an emotion in the robot with respect to an external input from a touch switch (pressure sensor), a microphone, a camera, etc. By changing the response, you can express more richly and solve the problem that you get bored as a result of motion becoming monotonous, and improve the expression power to increase the familiarity of humans (users) to the robot. Was effective.

However, if the essential emotional changes are centrally determined by the external input as described above, the response of the robot becomes monotonous as a result, and again, there is a problem that the user gets bored immediately. there were.

[0004]

SUMMARY OF THE INVENTION Therefore, the main object of the present invention is to make the way of changing emotions in a robot more varied, so that the user and the robot become more familiar. The goal is to provide an automatic response robot that never gets tired of being.

[0005]

The present invention relates to a user information input means for inputting user information, a storage means for storing user gesture information as a robot gesture pattern, and a unique gesture pattern based on internal information. An operation control means for comparing and processing each gesture pattern stored in the storage means with a pattern corresponding to the user information input from the user information input means, and one of the gestures based on the output from this operation control means. It is an automatic response robot provided with a reproducing means for reproducing a pattern.

[0006]

According to the gesture pattern of the robot, even if the same external signal (user information signal) is input, different results can be obtained depending on the state, for example, the length of the touch time by the pressure sensor, and the response output from the robot is also obtained. It will be different.

[0007]

According to the present invention, it is possible to realize an automatic response robot in which the user and the robot are more intimate and the user is not bored.

Further, it is possible to eliminate discomfort when the robot enters the human living space, and it is possible to smoothly introduce and operate the robot.

The above-mentioned objects, other objects, features and advantages of the present invention will become more apparent by the detailed description of the embodiments given below with reference to the drawings.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Referring to FIG. 1, an automatic response robot (hereinafter, simply referred to as "robot") 1 according to an embodiment of the present invention.
0 is a body portion 14 having a head portion 12, and the body portion 14
Arm portions 16 and 18 provided on both sides of the body portion, and movable leg portions 20 and 22 formed by wheels or the like provided on the bottom portion of the body portion 14. The external shape of the robot 10 may be a stuffed animal such as a dog.

A face portion 24 having eyes, a nose, and a mouth is provided on the front surface of the head portion 12, and ear portions 26 and 2 each having a microphone 38 for inputting a voice to be described later on both sides of the face portion 24.
Have eight. In addition, the mouth of the face portion 24 has a built-in speaker 48 that outputs a sound to be described later. Further, although not shown, for example, a pressure sensor 40 as a touch switch for sensing the pressure of the user's stroking or tapping described later is provided on the upper surface of the head 12, for example.

The head 12 is swung back and forth, the left and right arms 16 and 18 are swung appropriately, and the movable leg 2 is used.
A battery (not shown) and a drive motor driven by the battery are mounted to drive 0 and 22 to perform a walking motion.

The hardware configuration shown in FIG. 2 will now be described. The robot 10 has an external world signal detecting section 30 for detecting input user information and an external world for processing an analog signal detected by the external world signal detecting section 30. Signal processing unit 3
2. An operation control unit 34 including a microcomputer (hereinafter, simply referred to as “microcomputer”) that performs operation processing based on a digital signal digitally processed by the external signal processing unit 32, and an output of the operation control unit 34 An output response unit 36 that operates the robot 10 based on

That is, the operation control unit 34 determines the behavior of the robot 10 based on the obtained external information (user information) and internal information described later, and the output response unit 36 determines the emotion of the robot 10 (for example, "lonely. "·"pleasant"·
Three emotions such as "anger" and actions are transmitted to the user.

The external signal detector 30 includes, for example, a microphone 38 provided on the ears 26 and 28 or a pressure sensor 40 provided on the upper surface of the head 12.
The voice information of the user input from the microphone 38 is processed into a digital voice signal by a voice processing recognition device 42 including a voice processing microcomputer and a voice recognition microcomputer and recognized.
The user pressure information input from the pressure sensor 40 is converted into a digital signal by the A / D converter 44.

Further, the microcomputer of the operation control unit 34 stores a ROM storing an operation program, a RAM storing user gesture information as a robot gesture pattern and a unique gesture pattern based on internal information, and user information and internal information. It includes a CPU or the like that calculates a corresponding pattern operation based on information. Then, the result of the arithmetic processing by the operation control unit 34 is output as voice from the voice output circuit 46 of the output response unit 36 to the speaker 48, or the head 12, arm 16, and the like of the robot 10 via the actuator drive circuit 50. The actuator 52 such as 18 or the movable legs 20 and 22 is driven by a drive motor. As the gesture pattern of the robot 10, for example, changes in the three emotional states (lonely, angry, and enjoyable) described above depending on the difference in touch time (contact time) of the user with respect to the head 12 of the robot 10 are output by the response unit 36. Can be expressed as

Here, the basic operation of the robot 10 will be described with reference to the flowchart shown in FIG.

First, after starting, the user information is input to the external signal detection unit 30 in step S1, the user information input to the operation control unit 34 through the external signal processing unit 32 is detected in step S3, and further steps are performed. In S5, the gesture of the user information detected in step S3 is compared with a plurality of gesture patterns stored in the RAM in advance.

Then, in step S7, the operation controller 34
Determines whether the user's gesture based on the user information matches the robot's gesture pattern, and as a result,
If "YES", the gesture pattern motion is reproduced in step S9, and the reproduction output is output to the output response unit 36 of the robot 10 in step S11 to end the series of operations.

In the case of "NO" in which no gesture pattern matching the gesture corresponding to the user information is present in step S7, the unique gesture pattern based on the internal information stored in advance in the RAM of the microcomputer in step S13. The operation is reproduced, and the reproduction output is output to the output response unit 36 of the robot 10 in step S11 to end the series of operations.

For example, the operation control unit 34 determines the action of the robot 10 based on each gesture pattern corresponding to the user information (external information) and internal information stored in the microcomputer, and the action expression device, that is, the speaker. Voice by 48 and head 12, arm 16 as actuator 52,
The emotions and actions of the robot 10 can be expressed by the movements of 18 or the movable legs 20 and 22.

Further, a microphone 38 for detecting a user's voice is provided as an auditory information input device, and a pressure sensor for detecting a user's stroking or hitting of the head 12 of the robot 10 is provided as a tactile information input device. The pressure sensor 40 of FIG. Also, as internal information, besides emotions,
Although not shown, a battery fuel gauge for detecting starvation information of the robot 10 and a motor thermometer for detecting fatigue information of the robot 10 may be provided.

The operation based on concrete external information is as follows.
For example, when the user returns home, saying "I'm home" at the front door, this user information is input from the microphone 38 mounted on the robot 10, and the operation control unit 34
As a result, the robot 10 outputs "Welcome back" and a synthesized voice from the speaker 48 based on the gesture pattern, or the user strokes the head 12 of the robot 10 lightly or taps it. User information is input from the pressure sensor 40 and is similarly processed by the motion control unit 34. As a result, the robot 10 swings the head 12 back and forth, for example, to express a "pleasant" motion. Further, when the remaining battery level of each of the driven parts, that is, the head 12, the arms 16 and 18, and the drive motors that drive the movable legs 20 and 22 becomes small, for example, the robot 10 may detect the head based on the internal information. 12 is used to express hunger, or the motion of the arms 16 and 18 and the movable legs 20 and 22 is stopped to express fatigue.

Further, as another application example of the robot 10, an embodiment in which life support for the elderly is performed at home will be described.

First, in order to support the life of the elderly by the microcomputer of the operation control unit 34, for example, an operating program for waking up and going to bed at a time preset by the elderly and a simple conversation "Good morning" and "Good night"
The ROM stores an operation program for exchanging information, and an operation program for taking a walk.

Further, the operation control unit 34 has an operation variable X
There are [n], and the numerical values of n = 1 to 5 have the following meanings, for example, so that the value of X [n] represents the intention of the robot 10 to operate.

That is, X [1] = want to occur, X [2] =
I have five states: I want to sleep, X [3] = I want to call, X [4] = I want to be called, X [5] = I want to walk. Then, X [n] takes a value of 0 or 1, and when this value is 1, it indicates that the behavior is desired.

In the initial state, the robot 10 has X [n] =
0 (n = 1, ... 5). The robot 10 is woken up by an elderly person in advance at 7 am and
It is assumed that the operation program is set in the ROM so as to go to bed at 0 o'clock.

In this set state, the robot 1 is started in step S15 starting from the operation flowchart of FIG.
0 determines whether or not it is 7 o'clock in the morning, which is the time (one of the internal information) set by the timer built in the microcomputer of the operation control unit 34, and if the result is “YES”,
In step S17, the wakeup X [1] = 1 is changed to step S17.
At 9, the gesture pattern motion of the robot is reproduced. As a result, in step S11, the speaker 48 mounted on the mouth of the face portion 24 of the robot 10 outputs, for example, a voice saying "It's morning, please wake up." The audio output can be set not only once but repeatedly several times. If the determination result in step S15 is "NO", the operation in step S15 is repeated.

Similarly, at 10 pm, the speaker 48
A voice saying "It's night, good night" is output.

In this way, the robot 10 can assist the elderly in getting up and going to bed. Therefore,
The living support for the elderly living alone can be easily performed with a sense of familiarity and familiarity.

Further, another concrete example corresponding to the change of the emotion of the robot according to the present invention will be described below.

Concrete Example 1: In this case, the processing method of the sensor input (pressure sensor) 40 is changed depending on the emotional state. However, E is a parameter indicating the emotional state of the robot 10,
It can take three values: "lonely", "happy" and "angry". Further, t represents the time when the touch switch (pressure sensor) is pressed.

In the first specific example, the emotional state of the robot 10 is changed only by the switch input. Then, the difference in the time during which the touch switch (pressure sensor) 40 is continuously pressed is recognized as the difference in the pressed manner. FIG. 4 shows the flow of processing for changing the emotional state in this case.

In FIG. 4, after the power is turned on, step S1
Then, it is assumed that the robot 10 is in a "lonely" emotional state. Next, in step S3, the touch switch (pressure sensor) 40
Is read, and the processing is branched into three depending on the emotional state at that time.

That is, in step S5, the operation control unit 34
Determines if you are lonely and the result is "YES"
If so, the process proceeds to step S7, the touch switch input process 1 described later is executed, and the process returns to step S3. If “NO” in the step S5, it is determined in the step S9 by the operation control section 34 whether or not the operation is fun, and the result is “YE”.
If "S", the process proceeds to step S11, the touch switch input process 2 described later is executed, and the process returns to step S3. If the determination result of step S9 is "NO", the process proceeds to step S13, which will be described later. The touch switch input process 3 is executed, and the process returns to step S3.

As described above, after the operation control unit 34 has executed the processing for each emotional state, the touch switch input is read again and the processing according to the emotional state is repeated.

Next, a concrete operation flow of each of the input processes 1 to 3 in FIG. 4 will be described with reference to FIGS.

First, in FIG. 5, when the vehicle is in a "lonely" state after the start (start), t <100 in step S1.
ms? If the result is "YES", the robot 10 remains "lonely" in step S3 (this means that the robot 10 is suddenly hit and feels uncomfortable).

Then, in step S1, t <100 ms?
If the result is “NO”, step S5
To change to a "happy" state (this means leaving the lonely state due to contact with a person).

Next, in FIG. 6, when the vehicle is in a "fun" state after starting (start), in step S1, t <
100ms? If the result is “YES”,
In step S3, the state is changed to "angry" (this means that the player was struck and angry). On the other hand, if the determination result in step S1 is "NO", the process proceeds to step S5, where t <500 ms? If the result is “YES”, the state is changed to a “lonely” state in step S7 (this means that the person is fooled and becomes lonely). Then, in step S5, if t ≧ 500 ms and “NO”, step S5
Proceed to 9 and remain in a “fun” state (this means that you have been stroked and you remain in a good mood).

Further, in FIG. 7, when starting (starting) and in an "angry" state, t <1000 in step S1.
ms? If the result is "YES", it is not changed in an angry state in step S3 (this means that in an angry state, when touched, the player is in a bad mood).
Then, if “NO” in the step S1, the step S5
And change it to a "fun" state here (this is
It means that he was stroked, which means he was in a good mood).

As described above, by changing the way of processing the sensor input depending on the emotional state and changing the response of the robot 10, it is possible to solve the problem that the way of the response becomes monotonous.

Concrete Example 2: The emotional state of the robot is changed according to the response of the user to the response of the robot 10.

In the second specific example, after the robot 10 outputs a voice from the speaker 48, the emotional state of the robot 10 is changed according to the reaction from the user.

The flow of processing at this time is shown in FIG. Figure 8
In step S <b> 1, the robot 10 outputs a voice as “Nanadashi” from the speaker 48 based on the internal information. This is a response prompting the user to be stroked. Next, in step S3, the touch switch input is read, and in step S5, is there touch switch input? To judge. As a result, if there is no switch input with "NO", the state is changed to "lonely" in step S7 (this means that it is lonely because it was left alone). For example, the robot 10 is in a state where the head 12 is drooping.

On the other hand, if "YES" in step 5, the process proceeds to step S9, t <200 ms? If the result is “YES”, the state is changed to “angry” in step S11. (This means that he was angry because he was hit even though he wanted to stroke.) . As the state of “anger”, for example, “anger” can be expressed without the robot 10 performing a particular motion.

Then, in step S9, t ≧ 200.
In the case of "NO", which is judged to be ms, the state is changed to "fun" in step S13 (this means that I was in good mood because I was stroked as expected). As the operation of the robot 10 in this case, for example, the head 1
It can be expressed by shaking 2 back and forth.

Here, an example is shown in which the robot 10 wants to be stroked and outputs a voice from the speaker 48, saying, "Pandice," but the robot gives a response that expects a person (user) to make another action. Similarly, even when 10 outputs a voice, by changing the emotional state according to the reaction from the person, the emotion can be changed by the interaction with the person, and as a result, the response of the robot 10 changes. .

As a result, a person can feel a sense of familiarity when the robot 10 recognizes his / her existence and the person responds in response to the action from the robot 10.

[Brief description of drawings]

FIG. 1 is an external view showing an embodiment of an automatic response robot according to the present invention.

FIG. 2 is a hardware configuration diagram for controlling the operation of the robot shown in FIG.

FIG. 3 is an operation flowchart of a robot having the hardware configuration shown in FIG.

FIG. 4 is a main flowchart of touch switch input processing in another embodiment of the automatic response robot according to the present invention.

5 is an operation flow chart of touch switch input processing 1 in FIG. 4. FIG.

6 is an operation flowchart of touch switch input processing 2 in FIG.

FIG. 7 is an operation flowchart of touch switch input processing 3 in FIG.

FIG. 8 is an operational flowchart of a process of changing an emotional state in response to a voice output in still another embodiment of the automatic response robot according to the present invention.

[Explanation of symbols]

10 ... Automatic response robot 12 ... Head 14 ... Body 16, 18 ... Left and right arm 20, 22 ... Moving legs 30 ... External signal detector (user information input means) 32 ... External signal processing unit 34 ... Operation control section (operation control means) 36 ... Output response unit (playback means)

   ─────────────────────────────────────────────────── ─── Continued front page    F-term (reference) 2C150 CA01 CA02 DA24 DA26 DA27                       DA28 DF03 DF04 DF06 DF33                       DK02 ED42 ED52 EF03 EF07                       EF16 EF17 EF22 EF23 EF28                       EF29 EF33 EF36                 3C007 AS34 AS36 CS08 KS31 KS39                       MT14 WA03 WA04 WA16 WB16                       WB19 WB22 WB25 WC07

Claims (5)

[Claims] 1. A user information input means for inputting user information, a storage means for storing user's gesture information as a robot gesture pattern and for storing internal information as a unique gesture pattern, each of the memory elements stored in the storage means. Operation control means for comparing and processing a gesture pattern and a pattern corresponding to the user information input from the user information input means, and reproduction for reproducing any one of the gesture patterns based on the output from the operation control means. An automatic response robot having means. 2. The automatic response robot according to claim 1, wherein the internal information includes three emotional states of loneliness, fun and anger. 3. The automatic response according to claim 1, wherein the unique gesture pattern according to the emotional state is output from the reproducing means based on a command from the operation control means when the user information does not exist. robot. 4. The automatic response robot according to claim 1, wherein the user information input unit includes at least one of a voice input unit and a tactile input unit. 5. The automatic response robot according to claim 1, wherein the reproducing means includes at least one of a voice output means and an actuator means.