JP2004062063A - Interactive apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an interactive apparatus which can switch an interface for interaction according to the contents of the interaction. <P>SOLUTION: The interactive apparatus 1 is equipped with a first interface 80 for interacting with another interactive apparatus 1 in a first style, a second interface 70 for interacting with the latter interactive apparatus 1 in a second style different from the first style, a determining means 100 of determining the contents of the interaction with the latter interaction apparatus, and an interface selecting means 60 of selecting one of the first interface 80 and second interface 70 according to the contents of the interaction determined by the determining means 100, and those interactive apparatus 1 interact with each other through the interface selected by the interface selecting means 60. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an interactive device capable of switching an interface for performing an interaction according to the content of the interaction.
[0002]
[Prior art]
2. Description of the Related Art A speech recognition reaction device that recognizes a word and returns a response to the word and an operation corresponding to the word is known. For example, see JP-A-6-175689, JP-A-6-327842, and JP-A-9-131468. In particular, Japanese Unexamined Patent Publication No. Hei 9-131468 discloses a voice interactive toy in which toys interact with each other.
[0003]
As a pet robot or an entertainment robot imitating a four-legged animal such as a dog or a cat, for example, AIBO is already on the market.
[0004]
[Problems to be solved by the invention]
However, in the related art, although conversation can be performed between the voice interactive toys, no consideration is given to the owner (owner) who owns the voice interactive toy.
[0005]
In addition, as a result of a questionnaire survey conducted by the Japan Management Association Research Institute, the most popular dream robot drawn by consumers was “a pet robot that is closer to the real thing”. There is a long-awaited need for a living-type symbiotic or entertainment-type robot that shares a living space with humans.
[0006]
SUMMARY OF THE INVENTION It is an object of the present invention to provide an interactive device capable of switching an interface for performing the dialog according to the content of the dialog.
[0007]
[Means for Solving the Problems]
An interactive device according to the present invention is an interactive device for interacting with another interactive device, wherein a first interface for performing a dialog with the other interactive device in a first mode; A second interface for performing a dialog with the interactive device in a second mode different from the first mode, determining means for determining the content of the dialog with the other interactive device, and determining Interface selecting means for selecting one of the first interface and the second interface in accordance with the content of the dialogue determined by the means, wherein the interface between the interactive device and the other interactive device is provided. Interaction between them is performed via the interface selected by the interface selecting means, thereby achieving the above object.
[0008]
The first interface is an interface for interacting with the other interactive device in a manner understandable to the owner of the interactive device, and the second interface is an interface for owning the interactive device. It may be an interface for interacting with the other interactive device in a manner that cannot be understood by the user.
[0009]
The first interface is a voice interface for performing a dialog with the other interactive device by voice, and the second interface is a voice interface for performing a dialog with the other interactive device non-voice. It may be a non-voice interface.
[0010]
The non-voice interface may be a wireless communication interface.
[0011]
When the content of the dialogue determined by the determining unit is a content that may be heard by the owner of the interactive device, the interface selecting unit may select the first interface.
[0012]
If the content of the dialogue determined by the determining means is a content that the owner of the interactive device does not want to hear, the interface selecting means may select the second interface.
[0013]
The interactive device may further include a moving unit that allows the interactive device to move.
[0014]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0015]
FIG. 1 shows a configuration of a robot A and a robot B as an example of an interactive device 1 of the present invention. The robots A and B are configured to be able to interact with each other.
[0016]
Robots A and B shown in FIG. 1 include a camera 10 corresponding to “eyes”, a speaker 84 and an antenna 74 corresponding to “mouth”, and a microphone 30 and an antenna 74 corresponding to “ears”. .
[0017]
Note that, as will be described later, the speaker 84 forms at least a part of the audio interface 80, and the antenna 74 forms at least a part of the non-audio interface 70.
[0018]
In the example shown in FIG. 1, it is assumed that the owner (master) A of the robot A is a single woman and the owner (master) B of the robot B is a single man. It is also assumed that robots A and B have their names registered in advance.
[0019]
The robots A and B may be autonomous mobile robots (mobile robots) having a moving unit 160 that allows the robots to move, or may be of a type that cannot move by themselves. Good.
[0020]
Any mechanism that enables the robots A and B to move can be used. For example, the robots A and B may be configured to move forward and backward by controlling the rotation of rollers installed on the limbs, or may be a mobile robot such as a tire type or a leg type. The robots A and B may be humanoid robots that imitate animals that walk bipedally, such as humans, or pet robots that imitate quadrupedal animals.
[0021]
Although the interactive robot has been described as an example of the interactive apparatus 1, the interactive apparatus 1 is not limited to this. The interactive device 1 may be any device configured to be able to interact with another interactive device 1. The interactive device 1 may be, for example, an interactive toy or an interactive agent.
[0022]
FIG. 2 shows an internal configuration of the interactive device 1.
[0023]
The image recognition unit 20 captures an image from the camera (image input unit) 10, recognizes a person or another interactive device 1 from the captured image, and outputs the recognition result to the subject identification unit 50.
[0024]
The voice recognition unit 40 captures voice from the microphone (voice input unit) 30, recognizes the captured voice, and outputs the recognition result to the target person specifying unit 50.
[0025]
The target person specifying unit 50 specifies a target person such as a human (for example, a master) or another interactive device 1 based on the image recognition result and the voice recognition result, and indicates the specified target person. The target person specifying information is output to the interface (IF) selecting unit 60. The target person identification information includes owner recognition information indicating whether or not the owner of the interactive device 1 has been recognized.
[0026]
The target person specifying information makes it possible to detect the presence of the master or another interactive device 1 near the interactive device 1.
[0027]
Note that the example in which the interactive device 1 includes one system of the image input unit 10 and the image recognition unit 20 and one system of the voice input unit 30 and the speech recognition unit 40 has been described. It is not limited to this. The interactive device 1 may include only one input unit and recognition unit, or may include three or more input units and recognition units.
[0028]
The language processing unit 100 determines the content of the dialog with the other interactive device 1. Here, the language processing unit 100 can determine the content of the dialogue with another interactive device 1 in various modes.
[0029]
For example, the language processing unit 100 may determine the content of the dialogue by referring to the dialogue pattern stored in the dialogue database 120, or may refer to the personal information stored in the personal database 130. The content of the dialog may be determined, the content of the dialog may be determined by referring to information stored in the information database 140, or the content of the dialog may be determined according to a combination of information obtained from these databases. The content may be determined.
[0030]
The conversation database 120 stores conversation patterns and rules for generating response sentences. The personal database 130 stores information (eg, information on gender, age, name, occupation, character, hobby, date of birth, etc.) regarding the owner of the interactive device 1 and the acquaintance of the owner, and other interactive devices 1. And information about its owner. The information database 140 stores information such as today's weather and news, and knowledge such as various common senses. Information such as today's weather and news is acquired from the outside of the interactive device 1 via the transmission / reception unit 72 and the language processing unit 100, for example, and stored in the information database 140.
[0031]
Further, the personal database 130 may store a history of past conversations. By referring to the history of past conversations, it is possible to avoid repeating the same conversations as in the past, and it is possible to conduct conversations that are more natural.
[0032]
Alternatively, the language processing unit 100 may determine the content of the dialogue according to the recognition result of the image input from the image input unit 10, or may determine the content of the dialogue according to the recognition result of the voice input from the voice input unit 30. May be determined, or the content of the dialog may be determined according to a combination of these recognition results.
[0033]
Alternatively, the language processing unit 100 may determine the content of the dialog according to the target person specifying information output from the target person specifying unit 50.
[0034]
Alternatively, the language processing unit 100 may determine the content of the dialog according to a randomly generated random number.
[0035]
As described above, the language processing unit 100 functions as a determination unit that determines the content of the dialogue with another interactive device 1. The language processing unit 100 includes, for example, a memory 102 that stores a program, and a CPU 104 that executes the program stored in the memory 102. In this case, the functions and operations of the language processing unit 100 are realized by the CPU 104 executing a program stored in the memory 102.
[0036]
Furthermore, the language processing unit 100 may provide information indicating the content of the dialogue determined by the language processing unit 100 (for example, whether the content of the dialogue is not desired to be heard by the owner of the interactive device 1 or may be heard. Is output to the IF selection unit 60.
[0037]
It should be noted that whether the content of the dialogue is the content that the owner of the interactive device 1 does not want to hear or the content that may be heard can be determined, for example, together with the dialogue pattern stored in the dialogue database 120 together with the dialogue pattern. It can be defined in the database 120. Such an interaction pattern may be fixed to the default set, or may be added or changed from the default set by the education of the owner of the interactive device 1.
[0038]
The IF selection unit 60 receives the information indicating the content of the dialogue output from the language processing unit 100, and according to the information, determines whether the voice interface (first interface) 80 or the non-voice interface (second interface) 70 Choose one of them. Alternatively, the IF selection unit 60 considers the subject identification information output from the subject identification unit 50 in addition to the information indicating the content of the dialogue output from the language processing unit 100, and One of the audio interfaces 70 may be selected. Which of the voice interface 80 and the non-voice interface 70 is selected is determined, for example, by referring to a predetermined rule stored in a memory (not shown) provided inside the IF selection unit 60. It is determined.
[0039]
As described above, the IF selection unit 60 performs the first interaction with the other interactive device 1 in the first mode according to the content of the interaction determined by the language processing unit 100 (determination unit). The interface functions as an interface selection unit that selects one of the second interfaces for performing the interaction with the interface and the other interactive device 1 in a second mode different from the first mode.
[0040]
In the example shown in FIG. 2, the voice interface 80 is realized by a voice synthesis unit 82 and a voice output unit 84. The audio output unit 84 is, for example, a speaker. In the example illustrated in FIG. 2, the non-voice interface 70 is a communication interface using wireless communication, and is realized by the transmission / reception unit 72 and the antenna 74. An arbitrary communication interface (for example, a communication interface for optical communication or infrared communication (IrDA)) can be used as the communication interface using wireless communication.
[0041]
When the IF selection unit 60 selects the voice interface 80, the dialog between the interactive device 1 and another interactive device 1 is performed via the voice interface 80. In this case, the contents of the dialogue output from the language processing unit 100 are synthesized into a voice by the voice synthesis unit 82 and output from the voice output unit 84. When the IF selecting unit 60 selects the non-voice interface 70, the dialog between the interactive device 1 and another interactive device 1 is performed via the non-voice interface 70. In this case, the content of the dialogue output from the language processing unit 100 is converted into a communicable format by the transmission / reception unit 72 and output wirelessly from the antenna 74.
[0042]
Furthermore, the language processing unit 100 generates a control signal according to the recognition result of voice (word) or non-voice (wireless), and outputs the control signal to the operation control unit 110.
[0043]
The operation control unit 110 drives various actuators according to a control signal output from the language processing unit 100. This allows the interactive device 1 to operate in a desired manner.
[0044]
FIG. 3 shows a procedure of a process in which the interactive device 1 determines how to interact with another interactive device 1. This processing can be stored in the memory 102 in the language processing unit 100 in the form of a program, for example. This processing can be executed by, for example, the CPU 104 in the language processing unit 100.
[0045]
Here, the procedure of the process will be described by taking, as an example, a case where a robot B, which is an example of the interactive device 1, interacts with a robot A, which is an example of another interactive device 1. It is assumed that the robot A meets the robot B and a single man, Mr. B, who is the owner of the robot B in a city (for example, a park). Robot A may or may not know robot B. Whether or not the robot A knows the robot B can be determined, for example, by referring to the personal database 130 of the robot A.
[0046]
Step ST1: The robot B detects the robot A. For example, when the image recognition unit 20 of the robot B recognizes the image of the robot B, the robot B can detect that the robot A exists near the robot B.
[0047]
Step ST2: The robot B determines the content of the dialogue with the robot A. Here, the robot B can determine the content of the dialog in various modes. For example, the language processing unit 100 of the robot B may determine the content of the dialogue by referring to the dialogue pattern stored in the dialogue database 120 of the robot B. Alternatively, the language processing unit 100 of the robot B sends the personal information stored in the personal database 130 of the robot B (or the personal database 130 of the robot A) (for example, the owner A of the robot A is a single woman. , Etc.) may be used to determine the content of the dialogue.
[0048]
Step ST3: The robot B selects one of a voice interface (first interface) and a non-voice interface (second interface) according to the content of the dialogue with the robot A. If the robot B has selected the non-voice interface, the process proceeds to step ST4. If the robot B has selected the voice interface, the process proceeds to step ST5.
[0049]
Step ST4: The robot B interacts with the robot A via the non-voice interface.
[0050]
FIG. 4A shows a situation where the robot B and the robot A are interacting via a non-voice interface.
[0051]
Step ST5: The robot B interacts with the robot A via the voice interface.
[0052]
FIG. 4B shows a state where the robot B and the robot A are interacting via the voice interface.
[0053]
Step ST6: The robot B determines whether or not the conversation has ended. If “No”, the process returns to step ST2, and if “Yes”, the process ends.
[0054]
Whether the robot B interacts with the robot A via the voice interface or the robot A via the non-voice interface is determined according to a predetermined rule. The predetermined rule includes, for example, the following <Rule 1> and <Rule 2>.
[0055]
<Rule 1>: If the content of the dialogue with the robot A does not want to be heard by the owner (master) of the robot B, the robot B performs a dialogue with the robot A via the non-voice interface. .
[0056]
Example 1: Robot B collects information from Robot A for the purpose of creating a lover for his master, or sells his master to Robot A, and uses Robot A via a non-voice interface. Interact with. For example, the robot B can hear information such as the hobby and personality of the master of the robot A from the robot A, and can appeal the goodness of the master to the robot A so that the master of the robot A speaks implicitly. Requesting this corresponds to this case.
[0057]
Example 2: When the robot B tells the robot A the bad mood (habit, personality, etc.) of its master, or when the robot A asks the robot A to hear the complaint of the master, the robot B communicates with the robot A via the non-voice interface. Conduct a dialogue. For example, when the robot B has an agent function and the robot B complains to the robot A that "the master's command is difficult to change" or " This case corresponds to a case in which the robot A spills a complaint that the master is angry.
[0058]
Example 3: When the robot B asks the robot A what he does not know, the robot B interacts with the robot A via a non-voice interface. For example, this case corresponds to a case where the robot B does not find an answer to the homework when its master has given homework.
[0059]
<Rule 2>: If the content of the dialogue with the robot A is the content that the owner (master) of the robot B may be able to hear (or want to hear), the robot B communicates via the voice interface. Interact with the robot A.
[0060]
Example 1: Robot B interacts with Robot A via a voice interface when playing with his master. For example, this case corresponds to a case where the robot B wants to allow the robot B to answer the homework from the master by fumbling with the master.
[0061]
It is not essential that the robot B detect the robot A. This is because the robot B can spontaneously speak from itself regardless of the presence or absence of the other party. Therefore, step ST1 in FIG. 3 can be omitted.
[0062]
Note that in step ST3 of FIG. 3, the robot B determines whether or not its own master is close to itself, and only when it is determined that its own master is close to itself, the robot B One of a voice interface and a non-voice interface may be selected.
[0063]
In the above-described embodiment, the interactive device selects one of the voice interface and the non-voice interface according to the content of the dialog. However, the interface to be selected is the voice interface or the non-voice interface. It is not limited to. Whatever interface is used, a plurality of interfaces for performing a dialog with another robot in different modes, and an interface selecting unit for selecting one of the plurality of interfaces according to the content of the dialog. Interactive devices that include are within the scope of the invention.
[0064]
For example, a first non-voice interface (first interface) may be used instead of the voice interface, and a second non-voice interface (second interface) may be used instead of the non-voice interface. Even when such an interface is used, an adapter (interface converter) that converts a dialogue via the first non-voice interface into a dialogue that can be understood by the owner of the interactive device is provided by the owner of the interactive device. The same effect as in the case where the voice interface and the non-voice interface are used can be obtained by mounting. Interacting in a manner that is understandable to the owner of the interactive device has the same effect as conducting an audio conversation.
[0065]
Alternatively, an interface (first interface) for performing interaction with another interactive device in a manner understandable to the owner of the interactive device is used instead of the voice interface, and the dialog is used instead of the non-voice interface. An interface (second interface) for interacting with another interactive device in a manner that cannot be understood by the owner of the model device may be used. Even when such an interface is used, the same effect as when the voice interface and the non-voice interface are used can be obtained. Interacting in a manner incomprehensible to the owner of the interactive device has the same result as interacting non-speech.
[0066]
An example of an interface for interacting with another interactive device in a manner that the owner of the interactive device cannot understand is, for example, that the owner can only understand the first language (eg, Japanese) In some cases, a language interface is provided for the interactive device to interact with another interactive device in a second language (eg, English).
[0067]
【The invention's effect】
ADVANTAGE OF THE INVENTION According to the interactive apparatus of this invention, it becomes possible to switch the interface for performing the dialog according to the content of the dialog. As a result, the interactive device can perform a behavior close to human behavior, and the satisfaction and desire of the owner of the interactive device can be increased. As a result, such an interactive device can be a life-based symbiotic interactive toy and a pet robot closer to the real thing.
[Brief description of the drawings]
FIG. 1 is a diagram showing a configuration of a robot A and a robot B as an example of an interactive device 1 of the present invention. FIG. 2 is a diagram showing an internal configuration of the interactive device 1. FIG. FIG. 4A is a flowchart showing a procedure of a process for determining how to interact with the interactive apparatus 1. FIG. 4A is a diagram showing a situation where the robot B and the robot A are interacting via a non-voice interface. Showing the dialogue between the robot and the robot A via the voice interface.
60 IF selection unit 70 Non-voice interface 80 Voice interface 100 Language processing unit

Claims (7)

An interactive device for interacting with another interactive device,
A first interface for performing interaction with the other interactive device in a first manner;
A second interface for performing a dialog with the other interactive device in a second mode different from the first mode;
Determining means for determining the content of the dialogue with the other interactive device,
Interface selecting means for selecting one of the first interface and the second interface in accordance with the content of the dialogue determined by the determining means,
An interactive device, wherein the interaction between the interactive device and the other interactive device is performed via an interface selected by the interface selecting means. The first interface is an interface for interacting with the other interactive device in a manner that can be understood by the owner of the interactive device, and the second interface is an owner of the interactive device. The interactive device according to claim 1, wherein the interface is an interface for performing interaction with the other interactive device in a manner that cannot be understood by a user. The first interface is a voice interface for performing a dialog with the other interactive device by voice, and the second interface is a voice interface for performing a dialog with the other interactive device non-voice. The interactive device of claim 1, wherein the interactive device is a non-voice interface. The interactive device according to claim 3, wherein the non-voice interface is a wireless communication interface. 3. The interface selecting unit selects the first interface when the content of the dialogue determined by the determining unit is a content that may be heard by the owner of the interactive device. 4. An interactive device according to claim 4. 3. The interface according to claim 2, wherein when the content of the dialogue determined by the determining unit is a content that the owner of the interactive device does not want to be heard, the interface selecting unit selects the second interface. 4. Item 6. The interactive device according to any one of Items 4. The interactive device according to claim 1, wherein the interactive device further comprises a moving unit that allows the interactive device to move.

Images