JP2002120180A - Robot device and control method for it - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a robot device capable of improving an entertainment property and a control method for it. SOLUTION: In this autonomous robot device and control method for it, the outside and/inside status is recognized according to the output of an external sensor and/or internal sensor, and according to the recognition result, the robot behaves. After working given from the outside is detected, according to a previously stored program, a specified behavior previously uniquely associated according to the detection result is manifested, whereby according to the manifested behavior to the working given from the outside, a user easily can judge whether hardware or software in the device main body gets out of order or not.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a robot apparatus and a control method therefor, and is suitably applied to, for example, a pet robot.

[0002]

2. Description of the Related Art In recent years, a pet robot of a four-legged walking type that performs an action in response to a command from a user or the surrounding environment has been proposed and developed by the present applicant. Such a pet robot has a shape similar to a dog or cat bred in a general household, and acts autonomously in response to a command from a user or a surrounding environment. In the following, on the basis of departure from an external instruction, it is defined and used as autonomous to act in accordance with its own norm while sometimes obeying the instruction as necessary.

[0003]

By the way, in such a pet robot, it is necessary to be able to express emotions such as "anger" and "joy" in response to a user's action such as "hitting" or "stroking". If possible, it is considered that communication between the user and the pet robot can be made smooth, and the user's attachment and curiosity to the pet robot can be improved accordingly, thereby further improving the entertainment property.

Conventionally, based on such an idea,
The pet robot is a CCD arranged at a predetermined position on the body
(Charge Coupled Device) Based on the output of a camera, a microphone, a touch sensor, and the like, the surroundings and the state of the inside, the presence or absence of a command from a user, the presence or absence of an action, and the like are recognized, and the action is performed based on the recognition result. .

However, since such a pet robot behaves autonomously as described above, it does not always react or react in a fixed manner to a user's action or the like. For this reason, in such a pet robot, for example, when it is originally in a motionless posture (for example, when it is sleeping or just sitting)
Also, in the operation mode in which the movement is not intentionally expressed, the user may give a misunderstanding that the pet robot has been broken.

In a conventional pet robot, software in which a program for expressing the above-described autonomous behavior is stored is implemented on hardware in the main body of the pet robot. If the behavior that should be manifested in mode is not manifested,
It has been extremely difficult to determine externally whether "the behavior is suppressed by the emotion value of the virtual pet or the like" or "it is in a state where it cannot be expressed due to a failure of hardware or the like".

[0007] Therefore, in such a pet robot, if the user can easily recognize whether or not the behavior is normally exhibited in the autonomous operation mode, unnecessary misunderstanding or worry is prevented from being given to the user. Therefore, it is considered that the entertainment property can be further improved.

The present invention has been made in view of the above points, and has as its object to propose a robot apparatus capable of improving the entertainment property and a control method thereof.

[0009]

According to the present invention, there is provided an autonomous type apparatus for recognizing external and / or internal conditions based on the output of an external sensor and / or an internal sensor and acting on the recognition result. In the robot apparatus, a detecting means for detecting an externally applied action, a storing means for storing a predetermined program, and a program uniquely read in advance in accordance with a detection result of the detecting means based on a program read from the storing means. An action expressing means for expressing the associated specific action is provided.

As a result, in this robot device, the user can easily determine whether or not a failure has occurred in the hardware or software in the device main body based on the manifestation behavior in response to an externally applied action. Even when hardware or software in the apparatus main body does not actually have a failure, it is possible to prevent a user from misunderstanding that the robot apparatus has been broken.

According to the present invention, there is provided a method for controlling an autonomous robot apparatus which recognizes an external and / or internal situation based on an output of an external sensor and / or an internal sensor, and acts based on the recognition result. After detecting the given action, a specific action uniquely associated in advance is developed according to the detection result based on a program stored in advance.

As a result, in this method of controlling the robot apparatus, the user can easily determine whether or not a failure has occurred in the hardware or software in the apparatus body based on the manifestation of the action given from the outside. Thus, even if the hardware or software in the apparatus main body does not actually fail, the user can be prevented from misunderstanding that the robot apparatus has been broken.

Further, in the present invention, the storage means in which the program is stored is provided in the main body of the robot device. As a result, in the robot apparatus and the method thereof, when a failure occurs in the robot apparatus, whether the failed part is hardware or software in the apparatus body or a storage medium loaded from outside. Can be easily determined by the user.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below in detail with reference to the drawings.

(1) Configuration of Pet Robot According to the Present Embodiment In FIG. 1, reference numeral 1 denotes a pet robot according to the present embodiment as a whole, and leg units 3A to 3D are connected to the front, rear, left and right of body unit 2, respectively. The head unit 4 and the tail unit 5 are connected to the front and rear and rear ends of the body unit 2, respectively.

In this case, as shown in FIG. 2, a controller 10 for controlling the operation of the pet robot 1 as a whole, a battery 11 as a power source of the pet robot 1, a battery sensor 12, Internal sensor unit 15 including heat sensor 13, acceleration sensor 14, and the like
And a touch sensor (hereinafter referred to as a back touch sensor) 16 disposed on the back.

The head unit 4 has a microphone 17 corresponding to the "ears" of the pet robot 1 and a CCD (Charge Coupled De) corresponding to the "eyes".
device), a camera 18 and a touch sensor (hereinafter referred to as a head touch sensor) 19 disposed above, and a plurality of LEDs (Lig) having a function as an external "eye".
HT Emitting Diode, a light emitting diode (LED) 20, a speaker 21 corresponding to a “mouth”, a touch sensor 22 corresponding to a “chin” (hereinafter referred to as a chin touch sensor) 22, and the like. It is arranged in.

In this case, the back touch sensor 16 of the body unit 2 and the microphones 17 and C of the head unit 4
An external sensor unit 23 is configured by the CD camera 18, the head touch sensor 19, and the chin touch sensor 22.

Further, the tail unit 5 is provided with a tail 5A so as to be freely driven, and the tail 5A can emit blue or orange light for displaying the mental state of the pet robot 1 at that time. A simple LED (hereinafter, referred to as a mental state display LED) 5AL is provided.

Further, the joints of the leg units 3A to 3D, the leg units 3A to 3D and the trunk unit 2
Actuators 24 _{1 to} 24 _n corresponding to the respective degrees of freedom are arranged in each of the connecting portions, the connecting portion of the head unit 4 and the body unit 2, and the connecting portion of the tail unit 5 and the body unit 2. Has been established.

The microphone 1 of the external sensor unit 23
Reference numeral 7 collects command sounds such as “walk”, “down” or “follow the ball” given as a musical scale by a user via a sound commander (not shown), and sends the obtained audio signal S1A to the controller 10. . Further, the CCD camera 18 captures an image of the surrounding situation and obtains an image signal S.
1B is sent to the controller 10.

Further, the back touch sensor 16, the head touch sensor 19, and the chin touch sensor 22 are provided on the back of the body unit 2, the upper part of the head unit 4, and the chin, respectively, as apparent from FIG. And detects the pressure received by a physical action such as “stroke” or “hit” from the user, and returns the detection result to a pressure detection signal S.
1C, S1D, and S1E, respectively.
To send to.

The battery sensor 1 of the internal sensor section 15
2 detects the remaining energy of the battery 11, and sends the detection result to the controller 10 as a remaining battery detection signal S2A. The temperature sensor 13 detects the temperature inside the pet robot 1 and outputs the detection result to a temperature detection signal S2.
B is sent to the controller 10. Further, the acceleration sensor 14 detects accelerations in three axial directions (X-axis direction, Y-axis direction, and Z-axis direction), and outputs a detection result as an acceleration detection signal S
It is sent to the controller 10 as 2C.

The controller 10 includes a voice signal S1A, an image signal S1B and pressure detection signals S1C, S1D, S1E (hereinafter collectively referred to as an external information signal S1) provided from the external sensor unit 22, and an internal sensor unit 15 Remaining signal S2A and temperature detection signal S
Based on the 2B and the acceleration detection signal S2C and the like (hereinafter, these are collectively referred to as an internal information signal S2), the external and internal states, the presence or absence of a command from the user, and the presence or absence of an action are determined.

The controller 10 determines a subsequent action based on the result of this determination and a control program stored in advance in an external memory 25 which is removably mounted in the body unit 2. By driving the necessary actuators 24 _{1 to} 24 _n based on the above, the head unit 4 can be swung up and down, left and right, the tail 5A of the tail unit 5 can be moved, and each leg unit 3 can be moved.
A to 3D are performed to perform an action or action such as walking.

At this time, the controller 10 generates an audio signal S3 as necessary and gives it to the speaker 20, so that the audio based on the audio signal S3 is output to the outside, and as an external "eye". LE 20
By outputting the D drive signal S5, it is caused to emit light in a predetermined light emission pattern according to the determination result, and / or the LE 5AL is displayed on the psychological state display LED 5AL of the tail unit 5.
By transmitting the D drive signal S6, it emits light in a light emission pattern according to the mental state at that time.

As described above, in the pet robot 1, when the external memory 25 is loaded, an operation mode (hereinafter, referred to as an autonomous mode) having autonomy based on the control program read from the external memory 25. (Referred to as a mode), it is possible to autonomously act according to the external and internal states, the command from the user and the presence or absence of the user's action.

FIG. 3 shows a specific configuration of the LED section 20 having a function as an "eye" of the pet robot 1 in appearance.
Shown in As is apparent from FIG.
Is a pair of first red LEDs 20R each of which emits red light as an emotion expressing LED for expressing an emotion.
₁₁ , 20R ₁₂ and a pair of second red LEDs 20R
₂₁ and 20R ₂₂ and a pair of blue-green LEDs 20BG ₁ and 20GB ₂ respectively emitting blue-green light.

In this case, each first red LED 20R ₁₁ ,
Each of the 20Rs ₁₂ has a linear shape with a predetermined length of the light-emitting portion, and the front and rear of the head unit 4 are so arranged as to have a constricted positional relationship toward the front of the head unit 4 as indicated by an arrow a. It is arranged almost in the middle of the direction.

Each second red LED 20R ₂₁ , 20R
R ₂₂ has a light-emitting portion having a linear shape of a predetermined length, and has a positional relationship of expanding toward the front of the head unit 4.
The head unit 4 is disposed in the middle of the head unit 4 so as to have a substantially radial positional relationship with the 0R ₁₁ and 20R ₁₂ .

Thus, in the pet robot 1, by turning on the first red LEDs 20R ₁₁ and 20R ₁₂ at the same time, the expression of "anger" and the disgust as if the eyes were raised and angry were raised. such as facial expressions of "disgust" as if they feel able to express, by the each of the second red LED20R _21, 20R ₂₂ at the same time to light, such as facial expression if it were sad in "sadness" as if the And each of the first and second red LEDs 20R ₁₁ , 20R
_By turning on ₁₂ , 20R ₂₁ and 20R ₂₂ at the same time, it is possible to express the expression of “fear” as if feeling a fear, the expression of “surprise” as if being surprised, and the like. It has been done.

On the other hand, each blue-green LED 20BG ₁ ,
20BG ₂ are each light emitting portion has a bow shape of a predetermined length, the position immediately before the first red _LED20R 11, _{20R 12} corresponding in each head unit 4, a bow-shaped inner front (arrow a ).

Thus, in the pet robot 1, by turning on the blue-green LEDs 20BG ₁ and 20BG ₂ at the same time, “joy” as if they are smiling.
The expression can be expressed.

In the bot robot 1, the first and second red LEDs L are provided on the upper portion of the housing from near the front end of the head unit 4 to immediately before the head touch sensor 19.
_{ED20R 11, 20R 12, 20R 21} , 20R 22
And blue-green LED20BG _1, so as to cover the 20BG _2, for example, synthetic made of a resin material black translucent cover 26 (FIG. 1) is disposed.

Thus, in the pet robot 1, the first and second red LEDs 20R ₁₁ , 20R ₁₂ ,
20R ₂₁ , 20R ₂₂ and blue-green LED 20BG ₁ , 2
When 0BG ₂ is not lit, these are not visible from the outside, whereas the first and second red LEDs 20
When R ₁₁ , 20R ₁₂ , 20R ₂₁ , 20R ₂₂ and the blue-green LEDs 20BG ₁ , 20BG ₂ are lit, they can be reliably seen from the outside, and thus three types of “eyes” It is possible to effectively prevent a sense of incongruity caused by the presence of the.

In addition to the above configuration, this pet robot 1
In the case of, the LED section 20 is provided with a green LED 20G for displaying system information, which is driven to blink when the system of the pet robot 1 is in a special state, as described later.

In this case, the green LED 20G is an LED capable of emitting green light whose light emitting portion has a linear shape with a predetermined length.
And the first red LED 20 in the head unit 4
Translucent cover 2 slightly above R ₁₁ and 20R ₁₂
6 so as to be covered and hidden.

Thus, in the pet robot 1, the user can easily recognize the system state of the pet robot 1 based on the blinking state of the green LED 20G seen through the translucent cover 26. .

(2) Processing of Controller in Clinic Mode In this embodiment, when the external memory 26 is not loaded in the pet robot 1, a failure occurs based on the control program read from the internal memory 10A in the controller 10. By executing an operation mode (hereinafter, referred to as a clinic mode) for judging the presence or absence of the external information signal S provided from the external sensor unit 22
Specific actions associated in advance are expressed in accordance with the surrounding situation based on No. 1, the presence or absence of a command from the user, and the presence or absence of an action.

In the clinic mode, the controller 10 controls the head sensor 19, the back sensor 16, the chin sensor 22, and the CCD camera 18 of the external sensor unit 23 (hereinafter referred to as input sensors). ), A sensor input is given from any one of the input sensors, and in response to the sensor input,
LED unit 20 of head unit 4, LED 5AL for displaying mental state of tail unit 5, leg units 3A to 3D
And each output system of the speaker 21 of the head unit 4 (hereinafter, referred to as
(This is referred to as an output system device) is output (light emission, operation, sound emission) from at least one output system device.

For example, in the clinic mode, when the pet robot 1 captures an image by capturing a “ball” within the imaging range of the CCD camera 18, the controller 10 receives the imaging result and responds to each of the leg units 3A to 3D. LED 5A for displaying the psychological state of the tail unit 5 while driving the actuator 21 that performs
At the same time as lighting L in blue, a predetermined sound (scream) is emitted via the speaker 21 of the head unit 4.

In the clinic mode, when the controller 10 confirms that the back touch sensor 16 of the body unit 2 is pressed by the user in the pet robot 1, a predetermined sound (the speaker 21) of the head unit 4 is obtained. While emitting a vocal sound such as a reply), the actuator 24 corresponding to each of the leg users 3A to 3D is driven to proceed by five steps.

Further, in the clinic mode, when the controller 10 confirms that the pet robot 1 has pressed the head touch sensor 19 of the head user 4 by the user, the first and second LED units 20 of the head unit 4 are checked. 2
Red LEDs 20R ₁₁ , 20R ₁₂ , 20R ₂₁ , 2
The 0R _{22 and the} blue-green LEDs 20BG ₁ and 20BG ₂ emit light appropriately and selectively in accordance with the degree of the pressing, and sound with the intensity and rhythm according to the degree of the pressing via the speaker 21 of the head unit 4. Emit sound.

Specifically, when the user touches the head touch sensor 19
When the controller is stroked (Click), the controller 10
Emits a predetermined sound (cry) through the speaker 21 of the head unit 4.

When the user presses the head touch sensor 19 strongly (Hit), the controller 10 emits a predetermined sound (anger) through the speaker 21 of the head unit 4, 4 LED section 2
By turning on each of the first red LEDs 20R ₁₁ and 20R ₁₂ simultaneously, the expression of “anger” is expressed as if the eyes were raised and angry.

Further, when the user presses the head touch sensor 19 lightly (Pat), the controller 10 emits a predetermined sound (laughter) through the speaker 21 of the head unit 4 and simultaneously outputs the predetermined sound (laughter). LED
By turning on each of the blue-green LEDs 20BG ₁ and 20BG ₂ of the unit 20 at the same time, an expression of “joy” as if laughing is expressed.

Further, the user sets the head touch sensor 19 to three.
When pressed for longer than a second (Long Pat), the controller 10 emits a predetermined sound (cheers) via the speaker 21 of the head unit 4 and the first of the LED units 20 of the head unit 4. Red LED20R
₁₂ , blue-green LED 20BG ₁ , second red LED 20
R ₂₁ , second red LED 20R ₂₂ , blue-green LED 2
0BG By sequentially blinks one ₂ and in turn clockwise of the first red LED20R _11, though representing the expression of "delight" as if presenting lightning roulette user.

Further, in the clinic mode, when the controller 10 confirms that the pet robot 1 has pressed the chin touch sensor 22 of the head unit 4 by the user, the speaker 21 of the head user 4 is turned on regardless of the degree of the pressing. A predetermined sound (squealing sound) is emitted through the keyboard.

In this clinic mode, if no input operation is given by the user for a predetermined time in any of the input system sensors 16, 18, 19, and 22, the controller 10 sets each leg user 3A. By driving the actuator 24 corresponding to 3D, 5
Every 5 seconds, there are three types of “geishaku” expressed in different types of prone postures, “seated art” expressed in five different sitting postures, and “standing art” expressed in five different standing postures. ”Are sequentially and repeatedly expressed to express“ boring tricks ”as if they were bored.

As described above, in the pet robot 1, in the clinic mode, the input system sensors 16, 18, 1
9, any one of the input system sensors 16, 1
When a sensor input is provided from 8, 19, and 22, the controller 10 determines at least one of a plurality of output system devices that is set in advance corresponding to the sensor input, based on the control program read from the internal memory 10A.
The above output devices 3A to 3D, 5AL, 20, 21
And the respective output system devices 3A to 3D, 5AL,
By outputting (light emission, operation, sound emission) from the devices 20 and 21, it is possible to express a specific action associated in advance in accordance with the surrounding situation, a command from the user, the presence or absence of an action, and the like. .

(3) Processing of Controller 10 Next, specific processing of the controller 10 in the pet robot 1 will be described.

When the processing contents of the controller 10 are functionally classified, as shown in FIG. 4, a state recognition mechanism unit 30 for recognizing external and internal states, and emotions and emotions based on the recognition results of the state recognition mechanism unit 30 are obtained. Emotions that determine the state of instinct
An instinct model unit 31, an action determining mechanism unit 32 that determines a subsequent action or action based on the recognition result of the state recognition mechanism unit 30 and the output of the emotion / instinct model unit 31, and an action determined by the action determining mechanism unit 32 Transition mechanism unit 33 that makes a series of motion plans of the pet robot 1 for performing motion and motion
When, it can be divided into a device control unit 34 for controlling the actuator 24 ₁ to 24 _n based on the operation plan erected by the posture transition mechanism section 33.

Hereinafter, these state recognition mechanism units 30,
The instinct model unit 31, the action determination mechanism unit 32, the posture transition mechanism unit 33, and the device control mechanism unit 34 will be described in detail.

(3-1) Configuration of the state recognition mechanism 30 The state recognition mechanism 30 includes an external information signal S1 supplied from the external sensor unit 23 (FIG. 2) and an internal information signal S2 supplied from the internal sensor unit 15. And recognizes the specific information based on the above, and notifies the emotion / instinct model unit 31 and the action determination mechanism unit 32 of the recognition result as state recognition information S10.

In practice, the state recognition mechanism section 30 constantly monitors the sound collection signal S1A given from the microphone 17 (FIG. 2) of the external sensor section 23, and “walks” or “downs” as the spectrum of the sound collection signal S1A. ", When it detects a spectrum of the same scale as the command sound output from the sound commander in response to a command such as" follow the ball ", recognizes that the command has been given, and recognizes the recognition result as the emotion / instinct model unit 31 and The action decision mechanism 32 is notified.

The state recognizing mechanism 30 constantly monitors the image signal S1B provided from the CCD camera 18 (FIG. 2), and includes, for example, a “red round object” or a “red circle” in the image based on the image signal S1B. When a “vertical plane having a height equal to or more than a predetermined height” is detected, it is recognized that “there is a ball” and “there is a wall”, and the recognition result is notified to the emotion / instinct model unit 31 and the action determination mechanism unit 32.

Further, the state recognition mechanism 30 is provided with a pressure detection signal S1D given from the head touch sensor 19 (FIG. 2).
Is constantly monitored, and when a pressure that is equal to or more than a predetermined threshold value and is detected for a short time (for example, less than 2 seconds) is detected based on the pressure detection signal S1C, it is recognized as “hit” (scored), and When a pressure of less than a long time (for example, 2 seconds or more) is detected, it is recognized as “stroke (praised)”, and the recognition result is notified to the emotion / instinct model unit 31 and the action determination mechanism unit 32.

On the other hand, the state recognition mechanism section 30 constantly monitors the acceleration detection signal S2C given from the acceleration sensor 14 (FIG. 2) of the internal sensor section 15, and
For example, when an acceleration equal to or higher than a predetermined level detected based on 2C is detected, it is recognized that “a large impact has been received”. On the other hand, when an acceleration about a gravitational acceleration larger than this is detected, “(from a desk or the like)” is detected. And the recognition result is notified to the emotion / instinct model unit 31 and the action determination mechanism unit 32.

The state recognition mechanism 30 is provided with a temperature sensor 1
3 (FIG. 2) is constantly monitored, and when a temperature equal to or higher than a predetermined value is detected based on the temperature detection signal S2B, it is recognized that "the internal temperature has risen" and the recognition result is expressed as emotion / instinct. Notify the model section 31 and the action determination mechanism section 32.

(3-2) Processing of Emotion / Instinct Model Unit 31 As shown in FIG. 5, the emotion / instinct model unit 31 performs “joy”, “sadness”, “surprise”, “fear”, and “disgust”. And a basic emotion group 40 composed of emotion units 40A to 40F as emotion models provided corresponding to the six emotions of “anger” and “anger”, “appetite”, “love affection”, “search desire”, and “exercise desire”. Unit 41A-4 as a desire model provided corresponding to each of the four needs
1D basic desire group 41 and each emotion unit 40A
To 40F and the intensity increasing / decreasing functions 42A to 42J provided corresponding to the desire units 41A to 41D, respectively.

Each of the emotion units 40A to 40F
The degree of the corresponding emotion is represented by, for example, an intensity from 0 to 100 level, and the intensity is represented by intensity information S11A given from the corresponding intensity increase / decrease functions 42A to 42F.
１１S11F to change every moment.

Similarly to the emotion units 40A to 40F, each of the desire units 41A to 41D expresses the degree of the corresponding desire by the intensity from 0 to 100 level, and the intensity is expressed by the corresponding intensity increase / decrease functions 42G to 42J. It changes every moment based on the given intensity information S12G to S12J.

The emotion / instinct model 31 determines the state of emotion by combining the intensities of the emotion units 40A to 40F,
The state of the instinct is determined by combining the intensities of 1A to 41D, and the determined emotion and the state of the instinct are expressed as emotion /
The information is output to the action determining mechanism 32 as the instinct state information S12.

The intensity increasing / decreasing functions 42A to 42J correspond to state recognition information S10 given from the state recognition mechanism 30 and
Behavior information S representing the current or past behavior of the pet robot 1 itself given from the behavior determining mechanism 32 described later.
13, and generates and outputs intensity information S11A to S11J for increasing / decreasing the intensity of each of the emotion units 40A to 40F and each of the desire units 41A to 41D according to the parameters set in advance as described above. Function.

Thus, in the pet robot 1, the parameters of the intensity increasing / decreasing functions 42A to 42J are set to the respective behavior and action models (Baby1, Child1, Chi1).
ld2, Young 1 to Young 3, Adult
By setting different values for each of 1 to 4), the pet robot 1 can be given a character such as "irritable" or "quiet".

(3-3) Processing of the Action Determination Mechanism 32 The action determination mechanism 32 has a plurality of action models in the external memory 25. Then, the action determining mechanism unit 32 includes: the state recognition information 10 given from the state recognition mechanism unit 30;
Each emotion unit 40A-40 of the emotion / instinct model unit 31
F and the intensity of each of the desire units 41A to 41D and the corresponding behavior model, determine the next behavior or action, and use the determination result as the behavior determination information S14 as the posture transition mechanism unit 33.
And output to the growth control mechanism 35.

In this case, the action determining mechanism unit 32 determines the next action or operation from one node (state) ND _A0 as shown in FIG. 6 or any other node ND _{A0 to} ND _An as shown in FIG. To each node ND _A0 to
An algorithm called probability automaton for determining probabilistically based on the transition probability _P 0 to P _n which is set respectively arc _AR A0 _{to Ar An} for connecting the ND _An employed.

More specifically, the external memory 25 stores a behavior model as shown in FIG. 7 for each of the nodes NDA _{0 to} ND _An .
Is stored, and the action determining mechanism 32 determines the next action or operation based on the state transition table 50.

In the state transition table 50, input events (recognition results) as transition conditions at the nodes ND _{A0 to} ND _An are listed in a row of “input event” in order of priority, and further conditions for the transition conditions are further described. Are described in the corresponding columns in the rows of “data name” and “data range”.

Therefore, at the node ND ₁₀₀ defined in the state transition table 50 of FIG.
L) "is given, the" size (SIZ) "of the ball given together with the recognition result is given.
E) is in the range of 0 to 1000 (0,1000), or when a recognition result of “obstacle detected (OBSTABLE)” is given, the obstacle is given together with the recognition result. "Distance (DISTANC
"E)" is in the range of "0 to 100 (0, 100)", which is a condition for transition to another node.

In the node ND ₁₀₀ , even when there is no input of the recognition result, each emotion unit 4 of the emotion / instinct model unit 31 periodically referred to by the action determination mechanism unit.
0A to 40F and the intensity of each desire unit 41A to 41D, “joy” and “surprise”
E) ”or“ SUDNESS ”emotion unit 40A to 40F has a strength of“ 50 to 100 ”.
Is within the range (50, 100) ”, it is possible to transit to another node.

In the state transition table 50, in the column of “transition destination node” in the column of “transition probability to another node”, the names of nodes that can transition from the nodes ND _{A0 to} ND _An are listed, and Other nodes ND _{A0 to} ND _An to which a transition can be made when all the conditions described in each row of “event name”, “data value”, and “data range” are met
Is described in the row of “output action” in the column of “transition probability to another node”. Note that the sum of the transition probabilities of each row in the column of “transition probability to another node” is 100
[%].

Therefore, in the node NODE _{100 of} this example, for example, “ball is detected (BALL)”, and it is recognized that the “size” of the ball is in the range of (0, 1000) (0, 1000). If the result is given, the “node NODE” has a probability of “30 [%]”.
₁₂₀ (node 120) ", and the action or action of" ACTION 1 "is output at that time.

Each of the behavior models corresponds to the nodes ND _{A0 to} ND A described in the state transition table 50, respectively.
The configuration is such that a number of D _An are connected.

[0075] Thus the action determining unit 32 is stored, and when the state recognition information through S10 state recognition mechanism unit 30 is given to such last time a predetermined time from the expressing action has elapsed, the memory _{10 A} Using the state transition table 50 of the corresponding nodes ND _A0 to ND _{An in} the corresponding behavior model, the next behavior or behavior (the behavior or behavior described in the row of “output behavior”) is probabilistically determined. And outputs the result of the determination to the posture transition mechanism 33 and the growth control mechanism 35 as the action command information S14.

(3-4) Processing of Posture Transition Mechanism 33 When the behavior decision information S14 is given from the behavior decision mechanism 32, the posture transition mechanism 33 receives the action decision information S14.
A series of operation plans of the pet robot 1 for performing an action or an operation based on the operation plan are made, and operation command information S15 based on the operation plan is output to the control mechanism unit 34.

In this case, the posture transition mechanism unit 33 uses the nodes ND _{B0 to} ND _{B2 to} determine the postures that the pet robot 1 can take as shown in FIG.
And then, connected by directed arcs _AR B0 _{to Ar B2} representing the operation between possible transition nodes _ND B0 _{to ND B2,} and one node _ND B0 self operating arc operation to complete between _{~ND B2 AR} C0 _{~AR C2} Is used.

For this reason, in the memory 10A, a file in which the starting point posture and the ending posture of all the movements that can be expressed by the pet robot 1 as a source of such a directed graph are stored in a database (hereinafter, referred to as a network definition file) ) Is stored, and the posture transition mechanism unit 3
3 is based on this network definition file, and FIG.
12 to generate directed graphs 60 to 63 for the whole body, for the head, for the legs, and for the tail, respectively.

As is clear from FIGS. 9 to 12, in the pet robot 1, the posture is greatly “stand (oStanding)” and “stand (oSit)”.
"station", which is an attitude on a charging stand (not shown) for charging the battery 11 (FIG. 2), "oSleeping", and the battery 11 (FIG. 2).
n)). For each of these four postures, the basic posture (double circle) common to each "growth stage" and the one for "childhood", "childhood", "youth" and "adult" And one or more normal postures (circles).

For example, the portion surrounded by a broken line in FIGS. 9 to 12 represents the normal posture for “childhood”,
As is apparent from FIG. 9, the normal posture “down” in “childhood” is “oSleeping”.
b (baby) "," oSleeping b2 "~
There is “oSleeping b5”, and “oSitting b”, “oSit
ting b2 ".

Then, the posture transition mechanism unit 33 sends the “stand”, “walk”, “reach hand”,
When an action command such as "shake the head" or "move the tail" is given as the action command information S14, the command is designated from the current node using the corresponding directed graphs 60 to 63 while following the direction of the directed arc. A path to a node to which a posture is associated or a directed arc or a self-operation arc to which a specified operation is associated is searched, and the operations respectively associated with the directed arcs on the searched path are sequentially performed. Operation commands to be performed are sequentially output to the control mechanism unit 34 as operation command information S15.

For example, if the current node of the pet robot 1 is “oSitting” in the directed graph 60 for the whole body,
b ”, and the behavior determination mechanism unit 32 outputs“ oSleepi
If the action command of the operation to be expressed at a node ngb4 "(operation associated with the self-operating arc a ₁₎ is given to the posture transition mechanism unit, the posture transition mechanism section 33 on the directed graph 60 for the whole body" oSitting b) to search for a path from the node “o Sleeping b4” to the node “o Sleeping b4”, and an operation command to change the posture from the node “o Sleeping b 5” to the node “o Sleeping b 5”, and “o Sleeping b 5”
An operation command for changing the posture from the node of “oSleeping b3” to the node of “oSleeping b3”;
ng b3 ”node to“ oSleeping b
4 is sequentially output to the control mechanism unit 34 as the operation command information S15 as the operation command information S15, and finally, the self-operation associated with the operation specified by the node “oSleeping b4” arc _{a 1} through the "OSleeping b4" control mechanism unit 3 an operation command for returning to the node as operation command information S15 in
4 is output.

At this time, the “growth stage” of the pet robot 1
In order to change the movement (“rough” movement, “quiet” movement, etc.) according to the “personality” and the like, two nodes that can be transitioned may be connected by a plurality of directed arcs. In such a case, the posture transition mechanism 33
Selects a directed arc according to the “growth stage” and “character” of the pet robot 1 at that time based on the control of the growth control mechanism unit 35 described later as a path.

Similarly, a plurality of self-operation arcs returning from a certain node to the node may be provided in order to change the movement according to each “growth stage” and “character”. In such a case, the posture transition mechanism 3
3 selects a directed arc according to the “growth stage” and “character” of the pet robot 1 at that time as a path in the same manner as described above.

In the posture transition as described above, the time of staying in the middle position is almost “0”, and therefore, even during the transition of the posture, it passes through a node used in another “growth stage”. good. Therefore, when searching for a path from the current node to the target node or the directed arc or the self-acting arc, the posture transition mechanism unit 33 searches for the shortest path regardless of the current “growth stage”.

When an action command for the head, legs, or tail is given, the posture transition mechanism unit 33 changes the posture of the pet robot 1 based on the directed graph 60 for the whole body. Return to one of the basic postures (double circles), and then operate to change the posture of the head, leg, or tail using the directed graphs 61 to 63 of the corresponding head, leg, or tail. The command information S15 is output.

(3-5) Processing of the control mechanism unit 34 The control mechanism unit 34 generates a control signal S16 based on the operation command information S15 given from the posture transition mechanism unit 33, and generates each control signal based on the control signal S16. Actuator 2
4 by a ₁ to 24 _n for controlling the drive to perform the behavioral and operation specified in the pet robot 1.

(4) Operation and Effects of the Embodiment In the above configuration, the pet robot 1 starts up in the clinic mode when the power is turned on without the external memory 25 loaded in the body unit 2. At this time, a specific action uniquely associated in advance is expressed according to a sensor input such as a situation around the pet robot 1, a command from the user, and the presence or absence of an action.

Therefore, in the pet robot 1, the user can easily determine whether or not a failure has occurred in the hardware or software of the pet robot based on the output behavior of the pet robot in response to the input in the clinic mode. Can be. Thus, even if the hardware or software in the pet robot 1 has not actually failed, the user can operate the pet robot 1
It is possible to prevent a misunderstanding that the is broken.

Further, in the pet robot 1, the control program corresponding to the clinic mode is stored not in the external memory 25 but in the internal memory 10 A in the controller 10, so that a problem occurs in the pet robot 1. In this case, the part where the failure occurred is the pet robot 1
The user can easily determine whether it is hardware or software in the device or the external memory 25.

According to the above configuration, in the pet robot 1, when the power is turned on in a state where the external memory 25 is not loaded in the trunk unit 2, the pet robot 1 is started in the clinic mode, and the surroundings of the pet robot 1 and the user A specific action uniquely associated in advance in response to a command input from the user and a sensor input such as the presence or absence of an action is expressed, so that the user can easily perform the action based on the output action of the pet robot in response to the input in the clinic mode. Then, it can be determined whether or not the hardware or software of the pet robot has failed. Thus, even if the hardware or software in the pet robot 1 has not actually failed, the user can be prevented from misunderstanding that the pet robot 1 has been broken, Thus, the pet robot 1 that can improve the entertainment property can be realized.

(5) Other Embodiments In the above-described embodiment, the case where the present invention is applied to the four-legged walking type pet robot 1 configured as shown in FIG. 1 has been described. However, the present invention is not limited to this, and the point is that the present invention is widely applied to an autonomous robot apparatus that recognizes an external and / or internal situation based on the output of an external sensor and / or an internal sensor and acts based on the recognition result. Can be.

In the above-described embodiment, the input means sensors, that is, the head touch sensor 19, the back touch sensor 16, the chin Although the case where the touch sensor 22 and the CCD camera 18 are applied has been described, the present invention is not limited to this, and the microphone 16 other than the above-described input sensors of the external sensor unit 23 may be used. Various detection means can be widely applied.

Further, in the above-described embodiment, the case where the internal memory 10A in the controller 10 in which the control program corresponding to the clinic mode is stored is applied as the storage means for storing the predetermined program. As described above, the present invention is not limited to this, but the point is that any storage means provided inside the main body of the robot apparatus can be widely applied to various other structures.

Further, in the above-described embodiment, a specific action uniquely associated in advance in accordance with the sensor input of the input sensors 16, 18, 19, 22 based on the control program corresponding to the clinic mode. As an action manifesting means, the LED unit 20 of the head unit 4, the LED 5AL for displaying the psychological state of the tail unit 5, the respective output systems of the leg units 3A to 3D and the speaker 21 of the head unit 4 (hereinafter, referred to as This is referred to as an output device), but at least one of the output devices has been described. However, the present invention is not limited to this, and is widely applied to action expression means having various other configurations. Can be applied.

[0096]

As described above, according to the present invention, in an autonomous robot apparatus which recognizes an external and / or internal situation based on the output of an external sensor and / or an internal sensor, and acts based on the recognition result, Detection means for detecting an externally applied action, storage means for storing a predetermined program, and identification uniquely associated in advance in accordance with the detection result of the detection means based on the program read from the storage means And the behavior manifesting means for exhibiting the behavior of the device, the user can easily determine whether or not a failure has occurred in the hardware or software in the device main body based on the manifestation behavior in response to an externally applied action. This allows the user to determine whether the hardware or software in the main unit of the device has not actually failed. It is possible to avoid the misunderstanding that the robot apparatus is broken occurs in advance, thus the robot apparatus can be realized which can improve entertainment property.

Further, according to the present invention, in a control method of an autonomous robot apparatus which recognizes an external and / or internal situation based on an output of an external sensor and / or an internal sensor, and acts based on the recognition result, an externally provided After detecting a given action, based on a program stored in advance, by expressing a specific action uniquely associated in advance according to the detection result, it is possible to respond to an externally given action. Based on the manifestation behavior, the user can easily determine whether or not the hardware or software in the device body has failed, and as a result, the hardware or software in the device body has not actually failed. Even in this case, it is possible to prevent the user from misunderstanding that the robot device has been broken beforehand. Control method for a robot apparatus capable of improving entertainment property Te can be realized.

[Brief description of the drawings]

FIG. 1 is a perspective view showing an external configuration of a pet robot to which the present invention is applied.

FIG. 2 is a block diagram illustrating a circuit configuration of the pet robot.

FIG. 3 is a partial cross-sectional view illustrating a configuration of an LED unit.

FIG. 4 is a block diagram for explaining processing of a controller;

FIG. 5 is a conceptual diagram serving to explain data processing in an emotion / instinct model unit.

FIG. 6 is a conceptual diagram illustrating a stochastic automaton.

FIG. 7 is a conceptual diagram showing a state transition table.

FIG. 8 is a conceptual diagram serving to explain a directed graph.

FIG. 9 is a conceptual diagram showing a directed graph for the whole body.

FIG. 10 is a conceptual diagram showing a directed graph for the head.

FIG. 11 is a conceptual diagram showing a directed graph for a leg.

FIG. 12 is a conceptual diagram showing a directed graph for the tail.

[Explanation of symbols]

1 ... pet robot, 2 ... body unit, 3A ~
3D ... leg unit, 4 ... head unit, 5 ... tail unit, 5AL ... psychological state display LED, 10
... Controller, 10A ... Memory, 15 ... Internal sensor, 16 ... Back touch sensor, 17 ... Microphone, 18 ... CCD camera, 19 ... Head touch sensor, 20 ... LED section, 21 ... Speaker, 22: jaw touch sensor, 23: external sensor, 24 _{1 to} 24 _n ...
An actuator 25, an external memory 30, a state recognition mechanism 31, an emotion / instinct model 32, an action determining mechanism 33, a posture transition mechanism 34, a control mechanism 34.

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 2C150 CA01 CA02 CA04 DA05 DA24 DA25 DA26 DA27 DA28 DF03 DF04 DF06 DF33 ED42 ED52 EF03 EF07 EF16 EF23 EF29 EF34 EF36 3F059 AA00 BA02 BB06 BC07 CA05 DB02 DB09 DC01 DD06 DD06 CA14 GD15 HA02 5H223 AA06 BB05 CC06 EE19 EE30

Claims (4)

[Claims] 1. An autonomous robot apparatus which recognizes an external and / or internal situation based on an output of an external sensor and / or an internal sensor and acts based on a result of the recognition. Means, storage means for storing a predetermined program, and action manifestation means for exhibiting a specific action uniquely associated in advance in accordance with the detection result of the detection means, based on the program read from the storage means. A robot device comprising: 2. The robot apparatus according to claim 1, wherein said storage means is provided in a main body of said robot apparatus. 3. A control method for an autonomous robot apparatus which recognizes an external and / or internal situation based on an output of an external sensor and / or an internal sensor and acts based on a result of the recognition. A first step of detecting, and a second step of expressing a specific action uniquely associated in advance according to the detection result based on a program stored in advance. A method for controlling a robot device. 4. The method according to claim 3, wherein in the second step, storage means for storing the program is provided in a main body of the robot device.