JP7176244B2 - Robot, robot control method and program - Google Patents

Description

The present invention relates to a robot, a robot control method, and a program.

Robots for the purpose of voice interaction are used in general households and the like. Improving the precision of dialogue between a robot and a predetermined target including a user is important for the spread of robots intended for voice dialogue. When the distance between the robot and the predetermined target is longer than the dialogue distance required for dialogue between the robot and the predetermined target, or when the noise is large, it is difficult for the robot to interact with the predetermined target. be.

For example, Patent Literature 1 discloses an interactive robot that estimates the direction of a sound source from a predetermined target, moves the robot in the estimated direction of the sound source, and interacts with the predetermined target.

Japanese Patent Application Laid-Open No. 2006-181651

However, in the technique of Patent Literature 1, since the robot is moved from a predetermined target toward the sound source and only interacts with the predetermined target, when the predetermined target that is farther than the dialogue distance talks to the robot, Another problem is that it is difficult to improve the accuracy of speech recognition during dialogue between the robot and the predetermined target when the target speaks to the robot when there is a lot of noise.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a robot, a method for controlling the robot, and a program that achieve high speech recognition accuracy during dialogue.

In order to achieve the object of the present invention, one aspect of the robot according to the present invention is
an operation means for causing the device to operate;
a moving means for moving the own device;
When the device is spoken to by a predetermined target, the device detects the distance between the device and the predetermined target, and the detected distance can ensure the speech recognition accuracy of the dialogue with the predetermined target. a determination means for determining whether or not the interactive distance is exceeded ;
a control means for controlling the movement means so that the self-device moves toward the predetermined target when the determination means determines that the detected distance exceeds the interactive distance ; ,
with
When the predetermined target speaks to the device again while the device is moving toward the predetermined target, the control means determines the distance between the device and the predetermined target by the determination device. controlling the operating means so as to perform an operation indicating that the content spoken to during the movement is not recognized after it is determined that the distance is equal to or less than the interactive distance ;
It is characterized by

In order to achieve the object of the present invention, one aspect of the robot control method according to the present invention includes:
A control method for a robot comprising operating means for causing the own device to operate and moving means for moving the own device ,
When the device is spoken to by a predetermined target, the device detects the distance between the device and the predetermined target, and the detected distance can ensure the speech recognition accuracy of the dialogue with the predetermined target. a determination step of determining whether or not the interactive distance is exceeded ;
When the determination step determines that the detected distance exceeds the interactive distance, the operation means makes a first response for causing the predetermined target to stop the talking. a control step for controlling
including
if the determining step determines that the distance exceeds the interactive distance, controlling the moving means so that the own device moves toward the predetermined target;
When the predetermined target speaks to the device again while the device is moving toward the predetermined target, it is determined that the distance between the device and the predetermined target is equal to or less than the interactive distance. After that, controlling the operation means to perform an operation indicating that the content spoken to during the movement is not recognized.
It is characterized by

In order to achieve the object of the present invention, one aspect of the program according to the present invention is
A computer that controls a robot comprising operating means for causing the own device to move and moving means for moving the own device ,
When the device is spoken to by a predetermined target, the device detects the distance between the device and the predetermined target, and the detected distance can ensure the speech recognition accuracy of the dialogue with the predetermined target. Determination means for determining whether or not the interactive distance is exceeded ;
When the determining means determines that the detected distance exceeds the interactive distance, the operating means makes a first response for causing the predetermined target to stop talking. control means for controlling
function as
The control means is
controlling the movement means so that the device moves toward the predetermined target when the determination means determines that the distance exceeds the interactive distance;
When the predetermined target speaks to the device again while the device is moving toward the predetermined target, the determining means determines that the distance between the device and the predetermined target is equal to or less than the interactive distance. controlling the operating means to perform an operation indicating that the content spoken to during the movement is not recognized after it is determined that the
It is characterized by

Advantageous Effects of Invention According to the present invention, it is possible to provide a robot, a control method for the robot, and a program that achieve high speech recognition accuracy during dialogue.

It is a figure showing a robot concerning a 1st embodiment of the present invention. 1 is a block diagram showing the configuration of a robot according to a first embodiment of the present invention; FIG. 4 is a flow chart showing first interactive processing according to the first embodiment of the present invention; It is a figure explaining the 1st interactive processing based on the 1st Embodiment of this invention. It is a figure explaining the 1st interactive processing based on the 1st Embodiment of this invention. It is a figure explaining the 1st interactive processing based on the 1st Embodiment of this invention. It is a figure explaining the 1st interactive processing based on the 1st Embodiment of this invention. FIG. 4 is a block diagram showing the configuration of a robot according to a second embodiment of the present invention; FIG. 9 is a flow chart showing second interactive processing according to the second embodiment of the present invention; FIG. 10 is a diagram showing the relationship between the interactive distance and the volume of noise according to the modification of the present invention; FIG. 10 is a diagram showing the relationship between the reference value of the volume of noise and the distance between the user and the robot according to the modification of the present invention;

Hereinafter, robots according to embodiments for carrying out the present invention will be described with reference to the drawings.

(First embodiment)
As shown in FIG. 1, the robot 100 according to the first embodiment has a deformed shape of a human, and includes a head 101 on which members simulating eyes, a mouth, and a nose are arranged, and a leg simulating a leg. A body (housing) 102 in which the members are arranged, a hand 103 arranged in the body 102, a microphone (microphone) 104 arranged in the head 101, an imaging unit 105, a speaker 106, It has a display unit 107, a moving unit 108 arranged on the bottom, and an operation button 120 provided on the back of the body 102. The body 102 has a control unit 110 and a power supply unit 130 inside. . The hand portion 103, the microphone 104, the imaging portion 105, the speaker 106, and the display portion 107 function as operating means.

The hand portion 103 is operated by a driving portion (not shown) under the control of the control portion 110 . For example, by putting the hand part 103 behind the ear, a gesture is used to express that the voice cannot be heard.

The microphones 104 are arranged in the right ear, left ear, and back of the head 101 and pick up sounds. A microphone 104 placed in the right ear picks up sound emitted from the right front. A microphone 104 placed in the left ear collects sound emitted from the left front. A microphone 104 placed on the back of the head collects sounds emitted from behind. Microphone 104 outputs the collected sound to control unit 110 . Thus, the microphone 104 functions as an audio input means for inputting audio.

The imaging unit 105 is a camera provided at the nose position of the head 101 . The imaging unit 105 captures an image of a predetermined target such as the user U, and outputs data representing the captured image to the control unit 110 . Thus, the imaging unit 105 functions as imaging means for capturing an image.

The speaker 106 is provided at the position of the mouth of the head 101 and speaks a sound under the control of the control unit 110 . Thus, the speaker 106 functions as an audio output unit that outputs audio.

The display unit 107 is provided at the position of the eyes of the head 101 and displays an image of the eyes based on the control of the control unit 110 .

The moving unit 108 is composed of a motor and tires, and autonomously moves the robot 100 based on the control of the control unit 110 . The moving unit 108 moves the robot 100 forward, backward, turn right, turn left, turn right, and turn left. Thus, the moving unit 108 functions as moving means.

The control unit 110 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), and a RAM (Random Access Memory). The ROM is composed of non-volatile memory such as flash memory, and includes programs for realizing various functions of the control unit 110, data indicating the voice pattern of the calling voice, voice data indicating the notification of starting a dialogue, and countermeasures against talking while moving. store audio data for The RAM is composed of volatile memory, and is used as a work area for executing programs for the control unit 110 to perform various processes. In addition, the RAM stores whether the talking-while-moving Flag is ON or OFF. The control unit 110 has a voice analysis unit 111, a movement control unit 112, a determination unit 113, and an interactive control unit as shown in FIG. It functions as part 114 . The movement control unit 112 and the interaction control unit 114 function as control means.

The voice analysis unit 111 detects the voice uttered by the user U, and determines whether or not the voice uttered by the user U includes a pre-registered calling voice. The calling voice includes, for example, the name of the robot, "hey", "listen", and the like. Since the calling voice is matched with a pre-registered voice pattern, it can be detected even if the distance D exceeds the dialogue possible distance. Also, the voice analysis unit 111 detects the direction of the sound source from which the calling voice is emitted. Specifically, the voice analysis unit 111 determines the direction of the sound source from which the calling voice is emitted, the microphone 104 placed in the right ear of the head 101, the microphone 104 placed in the left ear, and the microphone 104 placed in the back of the head. It is detected by the volume difference or phase difference of the calling voice picked up by the microphone 104 . Also, the voice analysis unit 111 determines whether or not the user U is speaking to the user while moving. When determining that the user U is talking to the user U while moving, the voice analysis unit 111 turns on the moving talk flag.

The movement control unit 112 controls the movement unit 108 to turn the robot 100 so that it faces the direction of the sound source from which the calling voice is emitted. Specifically, the movement control unit 112 turns the robot 100 to the right or left in the direction detected by the voice analysis unit 111 or the direction of the face of the user U detected by the imaging unit 105, and the calling voice is emitted. facing the direction of the sound source. The movement control unit 112 also controls the movement unit 108 to move the robot 100 toward the user U. Further, after controlling the movement unit 108 so that the robot 100 moves toward the user U from a position beyond the conversational distance D, the movement control unit 112 moves the robot 100 to the conversational distance D or less. When it is determined that the robot 100 has approached, the moving unit 108 is controlled to stop the robot 100 .

When the user U speaks to the robot 100, the determination unit 113 determines whether or not the user U and the robot 100 can interact with each other. The face of the user U is detected from the image captured by the imaging unit 105, and the distance to the face of the user U is detected. The determination unit 113 may detect the distance from the size of the face, or may detect the distance to the face of the user U from the distance between the left and right eyes. After that, the determination unit 113 determines whether or not the user U and the robot 100 can interact with each other by determining whether or not the distance to the face of the user U is equal to or less than the interaction-possible distance D. The conversationable distance D is a distance at which speech can be recognized to such an extent that the speech collected by the microphone 104 can be converted into character information even if no speech pattern is registered in advance. The interactive distance D is, for example, 1 m. Thus, the determination unit 113 functions as determination means.

The dialog control unit 114 performs an operation to block the start of the dialog (first response for interrupting the dialog), a notification to start the dialog (a second response to continue the dialog different from the first response), Take countermeasures against talking while moving and execute dialogue. The dialogue control unit 114 performs an operation to prevent the dialogue from starting until the dialogue between the robot 100 and the user U becomes possible. Specifically, the dialog control unit 114 utters a voice such as "Hi, I'm home" or "Wait a minute" to prevent the start of the dialog, or puts the hand unit 103 behind the ear. perform an inaudible gesture against Further, when it is determined that the interaction between the robot 100 and the user U has become possible after executing the operation to prevent the interaction from starting, the interaction control unit 114 executes a notification to start the interaction. The dialog control unit 114 may utter the name of the user U, "What is it?", etc. from the speaker 106 as a notification to start the dialog. Gestures that initiate interactions may be performed. Further, the dialog control unit 114 determines whether or not the moving talk flag is ON, and if it is determined that the moving talk flag is ON, the speaker 106 utters a voice for countermeasures against the moving talk. do. Voices for countermeasures against talking while moving include "I didn't hear you" and "Tell me again".

The operation button 120 is provided on the back of the body section 102 and is a button for operating the robot 100, and includes a power button.

The power supply unit 130 is composed of a rechargeable battery built into the body unit 102 and supplies electric power to each unit of the robot 100 .

Next, the first interactive processing executed by the robot 100 having the above configuration will be described. The first dialogue process is a process of blocking the start of the dialogue when the distance between the user U and the robot 100 is greater than or equal to the dialogue possible distance D, and executing a notification to start the dialogue when the distance between the user U and the robot 100 approaches the dialogue possible distance D or less. is.

When the user U operates the operation button 120 to turn on the power, the robot 100 responds to the instruction to turn on the power and starts the first interactive process shown in FIG. The first interactive process executed by the robot 100 will be described below using a flowchart.

First, the voice analysis unit 111 detects voice uttered by the user U (step S101). Next, the voice analysis unit 111 determines whether or not the voice uttered by the user U includes a pre-registered calling voice (step S102). The calling voice includes, for example, the name of the robot, "hey", "listen", and the like. If it is determined that the calling voice is not included (step S102; No), steps S101 and S102 are repeated.

When it is determined that the voice uttered by the user U includes the calling voice (step S102; Yes), the voice analysis unit 111 turns off the moving talk flag (step S103). Next, the voice analysis unit 111 detects the direction of the sound source of the calling voice (step S104). Specifically, the voice analysis unit 111 determines the direction of the sound source from which the calling voice is emitted, the microphone 104 placed in the right ear of the head 101, the microphone 104 placed in the left ear, and the microphone 104 placed in the back of the head. It is detected by the volume difference or phase difference of the calling voice picked up by the microphone 104 . Next, the movement control unit 112 turns so that the robot 100 faces the direction of the sound source from which the calling voice is emitted (step S105). Specifically, the movement control unit 112 controls the movement unit 108 to turn the robot 100 to the right or left, and direct it toward the sound source from which the calling voice is issued. Next, the determination unit 113 detects the face of the user U (step S106). Next, the determination unit 113 detects the distance to the face of the user U (step S107). The determination unit 113 may detect the distance from the size of the face, or may detect the distance to the face of the user U from the distance between the left and right eyes.

Next, the determination unit 113 determines whether or not the distance to the face of the user U is equal to or less than the interactive distance D (step S108). The interactive distance D is, for example, 1 m. When it is determined that the distance to the user U's face is not equal to or less than the conversationable distance D (step S108; No), the dialogue control unit 114 performs an operation to block the start of dialogue (step S109). For the operation to prevent the start of the dialogue, a sound such as "Hi, I'm home", "Wait a minute", etc., may be uttered from the speaker 106 to prevent the start of the dialogue. Gestures may be performed. Next, the movement control unit 112 controls the moving unit 108 to move the robot 100 toward the user U (step S110). Next, the voice analysis unit 111 determines whether or not the user U is speaking to the user while moving (step S111). If it is determined that the user U is talking to the user U while moving (step S111; Yes), the voice analysis unit 111 turns on the moving talk flag (step S112), and returns to step S106. If it is determined that the user U has not spoken to the user during movement (step S111; No), the process returns to step S106.

When it is determined that the distance to the face of the user U is equal to or less than the conversationable distance D (step S108; Yes), the dialogue control unit 114 notifies the start of dialogue (step S113). Specifically, the dialogue control unit 114 may utter a notification to start the dialogue such as “What is it?” You may At this time, the movement control section 112 controls the movement section 108 to stop the robot 100 . Next, the dialogue control unit 114 determines whether or not the moving talking flag is ON (step S114). If it is determined that the talking-while-moving flag is ON (step S114; Yes), the speaker 106 speaks as a countermeasure against talking-while-moving (step S115). Voices for countermeasures against talking while moving include "I didn't hear you" and "Tell me again". Next, the dialogue control unit 114 executes dialogue (step S116). If it is determined that the moving talking flag is not ON (step S114; No), the dialogue control unit 114 executes dialogue (step S116). After that, the voice analysis unit 111 determines whether or not an end instruction has been received (step S117). If it is determined that the end instruction has not been received (step S117; No), the process returns to step S101, and steps S101 to S117 are repeated. If it is determined that an end instruction has been received during the dialogue (step S117; Yes), the first dialogue process is terminated.

Next, the first interactive processing executed by the robot 100 according to the present embodiment will be described based on specific examples with reference to FIGS. 4 to 7. FIG.

When the user U operates the operation button 120 to turn on the power, the robot 100 responds to the instruction to turn on the power and starts the first interactive process. It is assumed that the robot 100 runs by itself and moves to a position separated from the user U by exceeding the interactive distance D. FIG.

When the user U calls the robot 100 shown in FIG. 4 to "listen", the voice analysis unit 111 detects the voice uttered by the user U (step S101; FIG. 3). Since the voice uttered by the user U includes the voice "Listen to me", the voice analysis unit 111 determines that the voice includes a pre-registered calling voice (step S102; Yes; FIG. 3). .

Next, the voice analysis unit 111 turns off the talking-while-moving flag (step S103; FIG. 3). Next, the voice analysis unit 111 detects the direction of the sound source of the calling voice (step S104; FIG. 3). At this time, the user U is positioned on the right side of the robot 100, so the voice analysis unit 111 detects the call by using the volume difference or phase difference of the call voice picked up by the microphones 104 arranged in the right ear, the left ear, and the back of the head. The direction of the sound source from which the voice is emitted is detected as right. Next, as shown in FIG. 5, the movement control unit 112 turns right so that the robot 100 faces the direction of the sound source from which the calling voice was issued (step S105; FIG. 3). Next, the determination unit 113 detects the user's face (step S106; FIG. 3). Next, the determination unit 113 detects the distance D1 to the user's face (step S107; FIG. 3).

Next, the determination unit 113 determines whether or not the distance to the user's face is equal to or less than the interactive distance D (step S108; FIG. 3). At this time, since the distance D1 is greater than the interactive distance D, it is determined that the distance to the user's face is not equal to or smaller than the interactive distance D (step S108; No; FIG. An operation to prevent the start is performed (step S109; FIG. 3). As an action to prevent the start of the dialogue, as shown in FIG. Speak a voice that prevents it from starting. Next, the movement control unit 112 controls the moving unit 108 to move the robot 100 toward the user (step S110; FIG. 3). Next, when the user U speaks while moving, the voice analysis unit 111 determines that the user speaks while moving (step S111; Yes; FIG. 3). Next, the voice analysis unit 111 turns ON the speaking flag while moving (step S112; FIG. 3), and returns to step S106.

When the robot 100 moves to the position shown in FIG. 7, the determination unit 113 determines that the distance D2 to the user's face is equal to or less than the interactive distance D (step S108; Yes; FIG. 3). Next, the dialogue control unit 114 notifies the start of dialogue (step S113; FIG. 3). Specifically, the dialogue control unit 114 puts the hand part 103 forward to perform a gesture to start dialogue, and utters a notification to start dialogue such as “What is it?” from the speaker 106 . Next, the dialog control unit 114 determines that the moving talk flag is ON (step S114; Yes; FIG. 3), and utters the voice "I couldn't hear" from the speaker 106 as a countermeasure against the moving talk (step S114; Yes; FIG. 3). S115; FIG. 3). Next, the dialogue control unit 114 executes dialogue (step S116; FIG. 3). After that, when the speech analysis unit 111 determines that an end instruction has been received during the dialogue (step S117; Yes; FIG. 3), the first dialogue process ends.

As described above, according to the robot 100 of the present embodiment, when the robot 100 detects the calling voice and the distance between the user U and the robot 100 is greater than the dialogue-possible distance D at which the speech recognition accuracy of the dialogue can be ensured, the explicit It is possible to improve speech recognition accuracy by indicating that the user U cannot actually have a dialogue yet and preventing unnecessary dialogue start until the user U approaches a position within the dialogue possible distance D or less. Further, by notifying that the robot 100 will start dialogue, it is possible to smoothly conduct dialogue after movement.

(Second embodiment)
The robot 100 according to the first embodiment moves autonomously by the moving unit 108, and when the user U calls the robot 100, if the robot 100 is away from the user U, the robot 100 moves and approaches the user U. . On the other hand, in the case where the robot 200 according to the second embodiment does not move autonomously and the volume of the noise is equal to or greater than the reference value at which the user U can interact, the volume of the noise becomes equal to or less than the reference value. Until it becomes, the operation|movement which prevents the user U from starting a dialogue is performed.

As shown in FIG. 8, the robot 200 according to the second embodiment does not have the movement unit 108 and the movement control unit 112 that are included in the configuration of the robot 100. In addition to the configuration of the robot 100, a lip A motion detector 115 is provided.

The voice analysis unit 111 detects the voice uttered by the user U, and determines whether or not the voice uttered by the user U includes a pre-registered calling voice. Also, the voice analysis unit 111 detects sounds other than the voice uttered by the user U as noise. Also, the voice analysis unit 111 determines whether or not the noise is sound emitted from an audio device such as a television, an audio device, or a musical instrument. Whether or not the sound is emitted from an audio device is determined based on whether the sound includes a voice or the sound of a musical instrument, whether the sound has a rhythm, and the like.

The determination unit 113 determines that a dialogue between the robot 200 and the user U is possible when the volume of the noise detected by the voice analysis unit 111 is equal to or less than a reference value for dialogue with the user U. do. The determination unit 113 determines whether or not the user U has called out to the robot 200 based on the lip motion of the user U detected by the lip motion detection unit 115 . Specifically, the determination unit 113 determines whether or not the movement of the lips of the user U includes the movement when pronouncing the robot's name, "Hey", and "Listen".

When the determination unit 113 determines that the noise volume exceeds the reference value, the interaction control unit 114 performs an operation to prevent the user U from starting the interaction until the noise volume becomes equal to or lower than the reference value. do. Further, when the voice analysis unit 111 determines that the noise is emitted from an acoustic device such as a television, an audio device, or a musical instrument, the dialog control unit 114 controls the voice indicating an instruction to lower the volume from the acoustic device. utter the

The lip motion detection unit 115 detects the face of the user U from the image captured by the imaging unit 105 and detects the motion of the user U's lips. Thus, the lip motion detection unit 115 functions as lip motion detection means.

Next, a second interactive process executed by the robot 200 according to the second embodiment will be described.

When the user U operates the operation button 120 to turn on the power, the robot 200 responds to the instruction to turn on the power and starts the second interactive process shown in FIG. The second interactive processing executed by the robot 200 will be described below using a flowchart.

First, the voice analysis unit 111 detects voice uttered by the user U (step S201). Next, the voice analysis unit 111 determines whether or not the voice uttered by the user U includes a pre-registered calling voice (step S202). The calling voice includes, for example, the name of the robot 200, "hey", "listen", and the like. If it is determined that the calling voice is not included (step S202; No), the lip motion detection unit 115 detects the face of the user U from the image captured by the imaging unit 105 (step S203). Next, the lip motion detector 115 detects the lip motion of the user U (step S204). Next, the determination unit 113 determines whether or not the user U has called out to the robot 200 based on the motion of the lips of the user U (step S205). Specifically, the determination unit 113 determines whether or not the movement of the lips of the user U includes the movement when the name of the robot 200, “hey”, and “listen” are pronounced. If it is determined that there is no calling (step S205; No), steps S201 to S204 are repeated.

When the voice analysis unit 111 determines that the calling voice is included (step S202; Yes), or when the determination unit 113 determines that the user U has called the robot 200 (step S205; Yes), voice analysis is performed. The unit 111 detects sounds other than the voice uttered by the user U as noise (step S206). Next, the determination unit 113 determines whether or not the volume of the noise detected by the voice analysis unit 111 is equal to or less than a reference value at which the user U can have a dialogue (step S207).

When the determination unit 113 determines that the noise volume is not equal to or lower than the reference value (step S207; No), the dialog control unit 114 performs an operation to prevent the dialog from starting (step S208). For the operation to prevent the start of the dialogue, a sound such as "Hi, I'm home", "Wait a minute", etc., may be uttered from the speaker 106 to prevent the start of the dialogue. Gestures may be performed. The voice analysis unit 111 determines whether or not the noise is sound emitted from an audio device such as a television, an audio device, or a musical instrument (step S209). Whether or not the sound is emitted from an audio device is determined based on whether the sound includes a voice or the sound of a musical instrument, whether the sound has a rhythm, and the like. When it is determined that the noise is the sound emitted from the audio equipment (step S209; Yes), the dialogue control unit 114 utters an instruction to lower the volume from the audio equipment, for example, a voice indicating "Turn down the volume". (step S210) and returns to step S206. If it is determined that the noise is not the sound emitted from the audio equipment (step S209; No), the process returns to step S206.

When the determination unit 113 determines that the volume of noise is equal to or less than the reference value (step S207; Yes), the dialogue control unit 114 notifies the start of dialogue (step S211). Specifically, the dialogue control unit 114 may utter a notification to start the dialogue such as “What is it?” You may Next, the dialogue control unit 114 executes dialogue (step S212). After that, the voice analysis unit 111 determines whether or not an end instruction has been received (step S213). If it is determined that the end instruction has not been received (step S213; No), the process returns to step S201, and steps S201 to S213 are repeated. If it is determined that an end instruction has been received during the dialogue (step S213; Yes), the second dialogue process is terminated.

As described above, according to the robot 200 of the second embodiment, when the volume of noise exceeds the reference value, the action of preventing the start of dialogue is performed. As a result, the speech recognition accuracy can be improved because the dialogue is stopped when the dialogue is difficult due to noise. In addition, by notifying that the robot 200 will start a dialogue, it is possible to have a smooth dialogue after movement. Further, the lip motion detection unit 115 detects the motion of the user U's lips from the image captured by the imaging unit 105, and the determination unit 113 determines whether the user U has called the robot 200 based on the motion of the user U's lips. determine whether This allows the robot 200 to determine whether or not the user U has called the robot 200 even when the noise is loud.

(Modification)
In the above-described embodiment, the determination unit 113 of the robot 100 according to the first embodiment determines whether or not the dialogue is difficult based on whether or not the distance to the user U exceeds the dialogue-possible distance D. did. The determination unit 113 of the robot 200 according to the second embodiment determines whether or not the dialogue is difficult based on whether or not the noise volume exceeds the reference value. The determination unit 113 of the robots 100 and 200 only needs to be able to determine whether or not the dialogue is difficult. For example, the determination unit 113 of the robots 100 and 200 may determine whether or not it is difficult to interact with the user U based on the distance to the user U and noise. For example, when the distance from the user U exceeds the conversationable distance D and the volume of noise exceeds a reference value, it may be determined whether or not the conversation is difficult. Further, the interactive distance D may be changed according to the volume level of noise. For example, as shown in FIG. 10, when the noise volume is high, the interactive distance D is decreased, and when the noise volume is low, the interactive distance D is increased. In this case, the robots 100 and 200 may calculate the interactive distance D using a function, or obtain the interactive distance D from a table stored in advance in the ROM. In addition, the reference value of the volume of noise with which the user U can have a dialogue may vary depending on the distance between the user U and the robots 100 and 200 . For example, as shown in FIG. 11, when the distance between the user U and the robots 100 and 200 is large, the noise volume reference value is decreased. In this case, the robots 100 and 200 may calculate the noise volume reference value using a function, or obtain the noise volume reference value from a table stored in advance in the ROM. Further, the determination unit 113 of the robots 100 and 200 may determine whether or not the dialogue is difficult based on the volume of the voice of the user U, the rate at which speech recognition can be converted into character data, the language used for the dialogue, and the like. .

The interaction control unit 114 of the robots 100 and 200 of the above embodiments may display an eye image on the display unit 107 so that the display unit 107 representing the eyes and the eyes of the user U make eye contact. . By doing so, the robots 100 and 200 can express more human-like motions. The action of blocking the start of the dialogue performed by the dialogue control unit 114 is to display an image with the eyes closed on the display unit 107 arranged in the ears and eyes where the microphone 104 is arranged, and the robots 100 and 200 You may pretend that you are not listening to the voice and that the user U is not looking at it.

In the first embodiment described above, the movement control unit 112 controls the movement unit 108 to move the robot 100 toward the user U, and stops the robot 100 when the robot 100 approaches the interactive distance D or less. An example of controlling the moving unit 108 has been described. The movement control unit 112 may control the robot 100 so as to approach the conversationable distance D or less. For example, when the determination unit 113 determines that the distance between the robot 100 and the user U exceeds the interactive distance D, the movement control unit 112 determines that the robot 100 is moving in an arbitrary direction. The moving unit 108 is controlled so that the robot 100 changes its direction of movement and moves toward the user U, and when the robot 100 is stopped, the robot 100 starts moving toward the user U. You may control the moving part 108 to. When the determination unit 113 determines that the distance between the robot 100 and the user U is equal to or less than the conversationable distance, the movement control unit 112 stops the robot 100 when the robot 100 is moving. The moving unit 108 may be controlled to keep the robot 100 stopped when the robot 100 is stopped. In this case, the dialogue control unit 114 may control the operation means to give a response to the user U to continue the dialogue.

In the above-described embodiment, an example in which imaging unit 105 is a camera provided at the position of the nose of head 101 has been described. The imaging unit 105 may be a camera provided at the position of one eye of the head 101, or may be a stereo camera provided for both eyes.

In the above embodiments, the distance between the robots 100 and 200 and the face of the user U is detected from the size of the face of the user U captured in the image captured by the imaging unit 105 or the distance between the left and right eyes. explained. The method of measuring the distance between the robots 100 and 200 and the face of the user U is not limited, and may be measured by a non-contact distance sensor. The non-contact distance sensor is, for example, an ultrasonic sensor or a laser sensor installed on the body 102 of the robot 100,200. When a stereo camera is used for the imaging unit 105, the distance between the robots 100 and 200 and the face of the user U may be detected from images captured by the stereo camera.

In the above embodiment, an example in which the predetermined target is the user U has been described. The predetermined object may be any object that can interact with the robot 100 , and may be a person, an animal such as a dog or a cat, or a robot other than the robot 100 .

In the above-described embodiment, an example in which the moving unit 108 is composed of a motor and tires has been described. The moving unit 108 only needs to be able to move the robot 100, and may be composed of, for example, a plurality of legs and motors for moving the legs. By doing so, the shape of the robot 100 can be made closer to a person or an animal.

In the above-described embodiments, robots 100 and 200 have a human-like shape, but the shape of robots 100 and 200 is not particularly limited. It may have a shape, and may be a shape imitating an animation character or an imaginary creature. Also, the robots 100 and 200 may be any robots that have an interactive function, and include cleaning robots that autonomously run to clean floors, security robots that patrol and monitor, and the like.

In the above-described embodiment, when the robot is spoken to by the predetermined target, the determining means determines whether or not the robot can interact with the predetermined target. The determining means may determine whether or not it is possible to improve the speech recognition accuracy between the robot and the predetermined target when the robot is spoken to by the predetermined target.

Further, in the above-described embodiment, the control means makes the first response for interrupting the dialogue when the determination means determines that the dialogue between the robot and the predetermined object is not possible. However, it is not limited to dialogue, and the control means controls the predetermined target when the determination means determines that it is not possible to improve the speech recognition accuracy between the robot and the predetermined target. A first response may be given to cause a break in speaking.

In addition, the first response may include output of sound by the audio output means to prompt the predetermined target to stop the dialogue, or output of posture by the action means to prompt the predetermined target to stop the dialogue. In addition, the first response is to intentionally pretend that the voice input means and imaging means are not functioning, and the robots 100 and 200 do not hear voices and do not see images. may include doing

In addition, the second response may include output of voice by the voice output means prompting the predetermined target to continue the dialogue, or output of posture by the action means prompting the predetermined target to continue the dialogue. In addition, the second response is to pretend that the voice input means and the imaging unit 105 are functioning intentionally, and the robots 100 and 200 are listening to the voice and also looking at the image. may include doing

Further, in the above-described embodiments, a robot was described as an example, but the present invention is not limited to a robot, and an electronic device such as an AI (Artificial Intelligence) speaker may be used.

In addition, the main part of the processing executed by the control unit 110, which is composed of a CPU, RAM, ROM, etc., is not based on a dedicated system, but a normal information portable terminal (smartphone, tablet PC (Personal Computer)), It can be executed using a personal computer or the like. For example, a computer program for executing the above-described operations can be stored on a computer-readable recording medium (flexible disk, CD-ROM (Compact Disc Read Only Memory), DVD-ROM (Digital Versatile Disc Read Only Memory), etc.). By storing, distributing, and installing this computer program in a portable information terminal or the like, an information terminal that executes the above-described processing may be constructed. Alternatively, the computer program may be stored in a storage device of a server device on a communication network such as the Internet, and may be downloaded by an ordinary information processing terminal to constitute an information processing device.

In addition, when the functions of the control unit 110 are realized by sharing the functions of an OS (Operating System) and an application program, or by cooperation between the OS and an application program, only the application program portion is stored in a recording medium or storage device. You may

It is also possible to superimpose a computer program on a carrier wave and distribute it via a communication network. For example, the computer program may be posted on a bulletin board system (BBS: Bulletin Board System) on a communication network and distributed over the network. Then, this computer program may be activated and executed in the same manner as other application programs under the control of the OS so that the above processing can be executed.

The present invention is capable of various embodiments and modifications without departing from the broader spirit and scope of the invention. Moreover, the above-described embodiments are for explaining the present invention, and do not limit the scope of the present invention. In other words, the scope of the present invention is indicated by the claims rather than the embodiments. Various modifications made within the scope of the claims and within the meaning of the invention equivalent thereto are considered to be within the scope of the present invention. The invention described in the original claims of the present application is appended below.

(Appendix 1)
an operation means for causing the device to operate;
determination means for determining whether or not speech recognition accuracy between the device and the predetermined target can be improved when the device is spoken to by a predetermined target;
When the determination means determines that the accuracy of speech recognition between the device itself and the predetermined target cannot be improved, a first response for causing the predetermined target to stop talking to the target is provided. a control means for controlling the operating means to
comprising
A robot characterized by:

(Appendix 2)
The determination means determines whether or not a dialogue between the device and the predetermined object is possible when the device is spoken to by the predetermined object,
The control means makes the first response for interrupting the dialogue when the determination means determines that the dialogue between the device and the predetermined target is not possible. controlling the operating means to
The robot according to appendix 1, characterized by:

(Appendix 3)
The control means controls the operation means to make the first response for interrupting the dialogue until the dialogue between the device and the predetermined target becomes possible. do,
The robot according to appendix 2, characterized by:

(Appendix 4)
After the control means controls the operation means to make the first response, if the determination means determines that the interaction between the device and the predetermined target is possible, the control means for controlling the operating means to make a second response different from the first response for causing the predetermined target to continue the dialogue;
The robot according to appendix 2 or 3, characterized by:

(Appendix 5)
The determination means detects a distance between the device and the predetermined target when the device is spoken to by the predetermined target, and the determination means detects the distance required for the dialogue with the predetermined target. Determining that the interaction between the device and the predetermined target is not possible when the interaction possible distance is exceeded;
5. The robot according to any one of Appendices 2 to 4, characterized by:

(Appendix 6)
further comprising moving means for moving the device,
When the determining means determines that the distance exceeds the interactive distance, the control means controls the moving means so that the own device moves toward the predetermined target.
The robot according to appendix 5, characterized by:

(Appendix 7)
controlling the operating means to make the first response until the distance becomes equal to or less than the interactive distance;
The robot according to appendix 5 or 6, characterized by:

(Appendix 8)
voice input means for inputting voice;
a voice analysis means for analyzing the voice input by the voice input means;
further comprising
When the volume of the noise analyzed by the speech analysis means exceeds a reference value at which the dialogue with the predetermined object is possible, the determination means determines whether the dialogue between the device itself and the predetermined object is possible. is not possible,
When the determination means determines that the volume of the noise exceeds the reference value, the control means makes the first response until the volume of the noise becomes equal to or less than the reference value. controlling the operating means;
8. The robot according to any one of Appendices 2 to 7, characterized by:

(Appendix 9)
an imaging means for imaging an image;
lip motion detection means for detecting the face of the predetermined target from the image captured by the imaging means and detecting lip motion included in the detected face;
further comprising
The determination means determines whether or not the predetermined target has spoken to the lip motion detected by the lip motion detection means.
9. The robot according to any one of Appendices 2 to 8, characterized by:

(Appendix 10)
when the determination means determines that the distance between the device and the predetermined target exceeds the interactive distance,
The control means controls the movement means so that, when the device is moving in an arbitrary direction, the device changes the direction of movement and moves toward the predetermined target. , controlling the moving means so that the own device starts moving toward the predetermined target when the own device is stopped;
The robot according to any one of Appendices 2 to 9, characterized in that:

(Appendix 11)
when the determination means determines that the distance between the device and the predetermined object is equal to or less than the interactive distance,
The control means controls the moving means to stop the own device when the own device is moving, and stops the own device when the own device is stopped. controlling the moving means to maintain
The control means controls the action means to give a second response for causing the predetermined target to continue the dialogue.
11. The robot according to any one of Appendices 2 to 10, characterized by:

(Appendix 12)
further comprising audio output means for outputting audio,
The first response prompts the predetermined target to discontinue the dialogue, outputs voice by the voice output means, or prompts the predetermined target to discontinue the dialogue, and changes posture by the action means. including output,
The robot according to appendix 2, characterized by:

(Appendix 13)
further comprising audio output means for outputting audio,
The second response prompts the predetermined target to continue the dialogue, outputs voice by the voice output means, or prompts the predetermined target to continue the dialogue, and changes posture by the action means. including output,
The robot according to appendix 4 or 11, characterized by:

(Appendix 14)
voice input means for inputting voice;
an imaging means for imaging an image;
further comprising
The first response is to intentionally pretend that the voice input means and the imaging means are not functioning, and to express an emotion such that the self-device neither hears voices nor sees images. including doing
The robot according to appendix 2, characterized by:

(Appendix 15)
voice input means for inputting voice;
an imaging means for imaging an image;
further comprising
The second response is to pretend that the voice input means and the imaging means are functioning intentionally, and to express emotions such that the self-device is listening to voices and also seeing images. including to
The robot according to appendix 4 or 11, characterized by:

(Appendix 16)
the predetermined target includes a human or animal or other robot;
16. The robot according to any one of Appendices 1 to 15, characterized by:

(Appendix 17)
A control method for a robot comprising operating means for causing the robot to operate,
a determination step of determining whether or not it is possible to improve speech recognition accuracy between the device and the predetermined target when the device itself is spoken to by a predetermined target;
When it is determined by the determination step that the improvement of the speech recognition accuracy between the device itself and the predetermined target is not possible, a first response for causing the predetermined target to stop the speech is provided. a control step of controlling the operating means to
including,
A robot control method characterized by:

(Appendix 18)
A computer that controls a robot equipped with operating means that causes the device to operate,
Determination means for determining whether or not speech recognition accuracy between the device and the predetermined target can be improved when the device is spoken to by a predetermined target;
When the determination means determines that the accuracy of speech recognition between the device itself and the predetermined target cannot be improved, a first response for causing the predetermined target to stop talking to the target is provided. control means for controlling the operating means to
to function as
A program characterized by

DESCRIPTION OF SYMBOLS 100, 200... Robot 101... Head 102... Body part 103... Hand part 104... Microphone 105... Imaging part 106... Speaker 107... Display part 108... Moving part 110... Control part 111 112 movement control unit 113 determination unit 114 dialogue control unit 115 lip motion detection unit 120 operation button 130 power supply unit

Claims (16)

an operation means for causing the device to operate;
a moving means for moving the own device;
When the device is spoken to by a predetermined target, the device detects the distance between the device and the predetermined target, and the detected distance can ensure the speech recognition accuracy of the dialogue with the predetermined target. a determination means for determining whether or not the interactive distance is exceeded ;
a control means for controlling the movement means so that the self-device moves toward the predetermined target when the determination means determines that the detected distance exceeds the interactive distance ; ,
with
When the predetermined target speaks to the device again while the device is moving toward the predetermined target, the control means determines the distance between the device and the predetermined target by the determination device. controlling the operating means so as to perform an operation indicating that the content spoken to during the movement is not recognized after it is determined that the distance is equal to or less than the interactive distance ;
A robot characterized by: When the determination means determines that the detected distance exceeds the interactive distance, the control means makes a first response for causing the predetermined target to stop the talking. controlling said operating means to
The robot according to claim 1, characterized by: The control means causes the operation means to make the first response for interrupting the talking until the distance between the device and the predetermined object becomes equal to or less than the interactive distance . to control the
3. The robot according to claim 2, characterized by: After the control means controls the operation means to make the first response, the determination means determines that the distance between the own device and the predetermined object has become equal to or less than the interactive distance. Then, the control means controls the operation means to make a second response different from the first response to cause the predetermined target to continue the dialogue.
4. The robot according to claim 2 or 3, characterized in that: The second response is performed before an action indicating that the content spoken to during the movement is not recognized .
5. The robot according to claim 4 , characterized in that: voice input means for inputting voice;
a memory in which a calling voice is stored;
further comprising
The determining means compares the voice input from the voice input means with the calling voice stored in the memory to determine whether the predetermined target is speaking to the user .
6. The robot according to any one of claims 1 to 5 , characterized in that: when the determining means determines that the distance between the device and the predetermined target exceeds the interactive distance,
The control means controls the movement means so that, when the device is moving in an arbitrary direction, the device changes the direction of movement and moves toward the predetermined target. and controlling the moving means so that the own device starts moving toward the predetermined target when the own device is stopped;
7. The robot according to any one of claims 1 to 6 , characterized in that: when the determining means determines that the distance between the device and the predetermined target is equal to or less than the interactive distance,
The control means controls the moving means to stop the own device when the own device is moving, and stops the own device when the own device is stopped. controlling the moving means to maintain
The control means controls the action means to give the second response for causing the predetermined target to continue the dialogue.
6. The robot according to claim 4 or 5, characterized in that: further comprising audio output means for outputting audio,
The action indicating that the content spoken to while moving is not recognized is a voice requesting to repeat the content of the speech from the predetermined target, or recognizing the content of the speech from the predetermined target. An operation of outputting a voice indicating that it is not done from the voice output means ,
9. The robot according to any one of claims 1 to 8 , characterized in that: further comprising audio output means for outputting audio,
The first response prompts the predetermined target to discontinue the dialogue, outputs voice by the voice output means, or prompts the predetermined target to discontinue the dialogue, and changes posture by the action means. including output,
3. The robot according to claim 2 , characterized by: voice input means for inputting voice;
an imaging means for imaging an image;
further comprising
The first response is to intentionally pretend that the voice input means and the imaging means are not functioning, and to express an emotion such that the self-device neither hears voices nor sees images. including doing
3. The robot according to claim 2, characterized by: further comprising audio output means for outputting audio,
The second response prompts the predetermined target to continue the dialogue, outputs voice by the voice output means, or prompts the predetermined target to continue the dialogue, and changes posture by the action means. including output,
9. The robot according to claim 4 , 5 or 8 , characterized in that: voice input means for inputting voice;
an imaging means for imaging an image;
further comprising
The second response is to pretend that the voice input means and the imaging means are functioning intentionally, and to express emotions such that the self-device is listening to voices and also seeing images. including to
9. The robot according to claim 4 , 5 or 8 , characterized in that: the predetermined target includes a human or animal or other robot;
14. The robot according to any one of claims 1 to 13 , characterized in that: A control method for a robot comprising operating means for causing the own device to operate and moving means for moving the own device ,
When the device is spoken to by a predetermined target, the device detects the distance between the device and the predetermined target, and the detected distance can ensure the speech recognition accuracy of the dialogue with the predetermined target. a determination step of determining whether or not the interactive distance is exceeded ;
When the determination step determines that the detected distance exceeds the interactive distance, the operation means makes a first response for causing the predetermined target to stop the talking. a control step for controlling
including
if the determining step determines that the distance exceeds the interactive distance, controlling the moving means so that the own device moves toward the predetermined target;
When the predetermined target speaks to the device again while the device is moving toward the predetermined target, it is determined that the distance between the device and the predetermined target is equal to or less than the interactive distance. After that, controlling the operation means to perform an operation indicating that the content spoken to during the movement is not recognized.
A robot control method characterized by: A computer that controls a robot comprising operating means for causing the own device to move and moving means for moving the own device ,
When the device is spoken to by a predetermined target, the device detects the distance between the device and the predetermined target, and the detected distance can ensure the speech recognition accuracy of the dialogue with the predetermined target. Determination means for determining whether or not the interactive distance is exceeded ;
When the determining means determines that the detected distance exceeds the interactive distance, the operating means makes a first response for causing the predetermined target to stop talking. control means for controlling
function as
The control means is
controlling the movement means so that the device moves toward the predetermined target when the determination means determines that the distance exceeds the interactive distance;
When the predetermined target speaks to the device again while the device is moving toward the predetermined target, the determining means determines that the distance between the device and the predetermined target is equal to or less than the interactive distance. controlling the operating means to perform an operation indicating that the content spoken to during the movement is not recognized after it is determined that the
A program characterized by

Images