JP6744529B2 - In-vehicle device and vehicle - Google Patents

Description

Reference to related applications

This international application claims priority based on Japanese Patent Application No. 2014-262907 filed with the Japan Patent Office on December 25, 2014, and the Japanese Patent Application No. 2014-262907 The entire contents of this International Application are incorporated by reference.

The present disclosure relates to robots.

In recent years, robots that can talk to people have attracted attention. This robot acquires a voice uttered by a person with a microphone and estimates the meaning of the voice by voice recognition. Then, the preliminarily associated answer is given to the estimated meaning content (see Patent Document 1).

Japanese Patent No. 4015424

It is possible to improve the conversation ability of the robot by learning using artificial intelligence. However, even if the conversation ability of a specific robot is improved by learning, for example, when using another robot in a different place, learning must be performed from the beginning. One aspect of the present disclosure is to provide a robot that can reduce the burden on the user regarding learning of the robot.

An in- vehicle device according to an embodiment of the present disclosure includes a voice acquisition unit configured to acquire a voice of a vehicle occupant, and an instruction recognition configured to recognize an instruction to the vehicle from the voice acquired by the voice acquisition unit. A unit, a control content determination unit configured to determine a control content of the vehicle using a program according to the instruction recognized by the instruction recognition unit, and the control content determined by the control content determination unit To a vehicle control unit that controls the vehicle, a vehicle state acquisition unit configured to obtain a state of the vehicle using a sensor, and the control A learning unit configured to perform learning based on the control content output by the content output unit and the vehicle state acquired by the vehicle state acquisition unit, and the learning unit configured to reflect the learning result of the learning unit. A program update unit configured to update the program.

It is a block diagram showing an electric configuration of a robot of one embodiment. It is a front view showing the structure of a robot. 8 is a flowchart showing a program setting process executed by the robot. It is a flow chart showing conversation processing which a robot performs. It is a flow chart showing learning stop judgment processing which a robot performs. FIG. 6A is an explanatory diagram showing an example of an operation for instructing to stop learning, and FIG. 6B is an explanatory diagram showing an example of an operation for instructing to restart learning. It is a block diagram showing the electric constitution of a computer. It is a perspective view showing the external appearance of a computer. It is a block diagram showing the electric constitution of an in-vehicle device. It is a flow chart showing vehicle control processing which an in-vehicle device performs. It is a block diagram showing the electric constitution of the robot of another embodiment. It is a flow chart showing the program installation processing which a robot performs. It is a flow chart showing conversation processing which a robot performs. It is a flow chart showing the 2nd artificial intelligence unit learning stop judgment processing which a robot performs. 7 is a flowchart showing a program upload process executed by the robot. It is a block diagram showing the electric constitution of the robot of further another embodiment. It is a flow chart showing conversation processing which a robot performs. It is explanatory drawing showing the structure for performing a telephone call process. 7 is a flowchart showing a call pair setting process executed by a host computer. It is explanatory drawing showing the display which is displaying the icon.

1, 301... Robot, 3... Control unit, 5... Microphone, 7... Camera, 9... Touch panel, 11... Sensor group, 13... GPS, 15... Speaker, 17... Motor group, 19... Display, 21... Input unit, 23... Standard program storage unit, 25... Artificial intelligence unit, 26... Communication unit, 27... Arithmetic unit, 29... Output unit, 31... Cloud network, 33... Terminal, 35... Head, 37... Trunk, 39... Right arm Part, 41... Left arm part, 43... Right leg part, 45... Left leg part, 47, 49... Wheel, 51... Movement motor, 53... Neck joint, 55... Shoulder joint, 57... Elbow joint, 59... Wrist joint , 61... Joint motor, 63... Keyboard, 65... Mouse, 67... Casing, 69... Terminal, 71... Vehicle control unit, 101... Computer, 201... In-vehicle unit

An embodiment of the present disclosure will be described based on the drawings.
<First Embodiment>
1. Configuration of Robot 1 The configuration of the robot 1 will be described with reference to FIGS. 1 and 2. As shown in FIG. 1, the robot 1 includes a control unit 3, a microphone 5, a camera 7, a touch panel 9, a sensor group 11, a GPS 13, a speaker 15, a motor group 17, and a display 19.

The control unit 3 includes a microcomputer. The control unit specifically includes an input unit 21, a standard program storage unit 23, an artificial intelligence unit 25, a communication unit 26, an arithmetic unit 27, and an output unit 29.

The input unit 21 acquires information from the microphone 5, the camera 7, the touch panel 9, the sensor group 11, and the GPS 13, and outputs the information to the arithmetic unit 27 and the artificial intelligence unit 25.

The standard program storage unit 23 constantly stores a standard program for executing various operations of the robot 1. The standard program is different from the AI program described later,
It is a program whose contents do not change.

The artificial intelligence unit 25 can store an AI (artificial intelligence) program for executing various operations of the robot 1. When storing the AI program, the artificial intelligence unit 25 changes and develops the AI program by learning. Further, the artificial intelligence unit 25 can newly install the AI program and delete the AI program.

The communication unit 26 communicates with the external cloud network 31, the terminal 33, and the like. The communication may be wireless communication or wire communication.
The arithmetic unit 27 uses a standard program or an AI program to execute the arithmetic necessary for realizing various operations of the robot 1.

The output unit 29 outputs the calculation result of the calculation unit 27 to the speaker 15, the motor group 17, and the display 19. Detailed functions of each unit belonging to the control unit 3 will be described later.

As shown in FIG. 2, the robot 1 has a humanoid outer shape and includes a head portion 35, a body portion 37, a right arm portion 39, a left arm portion 41, a right leg portion 43, and a left leg portion 45.
The microphone 5, the camera 7, and the speaker 15 are provided on the head 35. The display 19 and the touch panel 9 are provided on the front side of the body portion 37.

The right leg portion 43 and the left leg portion 45 are provided with wheels 47 and 49 for movement, respectively. Two wheels 47 and 49 are provided in the front-rear direction, respectively. Therefore, the robot 1 contacts the ground with a total of four wheels. The wheels 47, 49 are driven by a moving motor 51 belonging to the motor group 17. By rotating the wheels 47, 49 by the driving force of the moving motor 51, the robot 1 can move back and forth. Further, by making the rotation speed of the wheel 47 different from the rotation speed of the wheel 49, the robot 1 can turn right or left.

The robot 1 includes a neck joint 53, a shoulder joint 55, an elbow joint 57, and a wrist joint 59. The degree of freedom of each joint can be appropriately set from 1 to 3. Each joint is operated by the driving force of the joint motor 61 belonging to the motor group 17. By appropriately selecting one of the above-mentioned joints with the joint motor 61, the robot 1 performs a predetermined motion and expresses a predetermined posture.

The plurality of sensors belonging to the sensor group 11 detect the position, speed, acceleration, inclination, joint angle, etc. of each part of the robot 1. The detection result is fed back to the arithmetic unit 27 via the input unit 21 and used for motion control of the robot 1.

The standard program storage unit 23, the artificial intelligence unit 25, and the arithmetic unit 27 are examples of a voice recognition unit, a processing unit, and a learning stop unit. The artificial intelligence unit 25 is an example of a learning unit. The communication unit 26 is an example of a storage unit and a learning result acquisition unit. The input unit 21 and the communication unit 26 are an example of an identification information acquisition unit. The cloud network 31 is an example of an external storage device.

2. Process Performed by Robot 1 (2-1) Program Setting Process The control unit 3 executes the program setting process shown in FIG. 3 in order to determine the program to be used from the standard program and the AI program. This program setting process is repeatedly executed at predetermined time intervals when the robot 1 is powered on.

In step 1, it is determined whether or not the AI program is being used at that time. When the AI program is not used (that is, when the standard program is used), the process proceeds to step 2, and when the AI program is used, the process proceeds to step 5.

In step 2, it is determined whether or not the identification information of the user has been input. The identification information can be input, for example, by the following method.
-Sending identification information from the terminal 33 to the communication unit 26.

Transmitting an electromagnetic wave (for example, radio wave, infrared ray, etc.) representing the identification information to the communication unit 26. The electromagnetic wave representing the identification information may be transmitted from the terminal 33, or may be transmitted from a fixed device (for example, a beacon, a wireless LAN access point, etc.). The electromagnetic wave representing the identification information may be transmitted periodically, or may be transmitted according to a user's instruction, or may be transmitted when the terminal 33, a fixed device, or the like detects the robot 1. May be.

-The user speaks out the content of the identification information. The microphone 5 acquires the voice and identifies the identification information by voice recognition.
The camera 7 captures a one-dimensional barcode or two-dimensional barcode representing identification information.

-Enter identification information using the touch panel 9.
The identification information may be composed of numbers or characters, may be a one-dimensional or two-dimensional image (for example, a one-dimensional bar code or a two-dimensional bar code), or the biometric information of the user ( For example, it may be a fingerprint, a pattern of veins on any part of the body, an iris, a face, etc.) or may be composed of voice.

If the identification information has been input, the process proceeds to step 3, and if the identification information has not been input, this processing ends.
In step 3, the AI program and data set corresponding to the identification information determined to be input in step 2 are installed from the cloud network 31. The installed AI program and data set are stored in the artificial intelligence unit 25.

The cloud network 31 stores the identification information, the AI program, and the data set in association with each other. The data set is a data set used in the use and learning of the AI program, and includes dictionary data used in voice recognition and voice synthesis.

In step 4, the program used by the robot 1 is changed from the standard program to the AI program installed in step 3 above. After this point, the robot 1 uses the AI program.

On the other hand, if an affirmative judgment is made in step 1, it is judged in step 5 whether or not the use termination condition of the robot 1 is satisfied. The use termination condition can be, for example, the following.

-The user does not operate the robot 1 for a predetermined time or longer.
-The state in which the user cannot be recognized in the image of the camera 7 or the voice acquired by the microphone 5 continues for a predetermined time or longer.

-The user performs an input corresponding to the end of use to the robot 1 (for example, the user says "end of use" or inputs contents corresponding to the end of use on the touch panel 9). ..

-The power of the robot 1 is turned off.
・It becomes the preset time.
If the use end condition is satisfied, the process proceeds to step 6, and if the use end condition is not satisfied, this process ends.

In step 6, the AI program and data set are uploaded to the cloud network 31. At this time, it is uploaded in association with the user identification information. When the learning described below is performed, the AI program and data set to be uploaded are those after learning.

In step 7, the AI program and the data set are deleted from the artificial intelligence unit 25.
In step 8, the program used is changed from the AI program to the standard program. After this point, the robot 1 uses the standard program.

(2-2) Conversation processing The control unit 3 executes the conversation processing shown in FIG. This process is executed when the microphone 5 detects a sound volume equal to or higher than a predetermined threshold.

In step 11, the voice is acquired using the microphone 5.
In step 12, the contents of the voice acquired in step 11 are recognized by a known voice recognition technique. At this time, if the standard program is used, the contents of the voice are recognized using the standard dictionary data that the robot 1 has in advance. When the AI program is used, the contents of the voice are recognized using the dictionary data installed from the cloud network 31 together with the AI program and reinforced by past learning.

In step 13, data of a voice to be answered (hereinafter referred to as answer voice data) is created for the content of the voice recognized in step 12. At this time, if the standard program is used, the reply voice data is created using the standard dictionary data that the robot 1 has in advance. When the AI program is used, the answer voice data is created using the dictionary data installed from the cloud network 31 together with the AI program and reinforced by past learning.

The content of the answer voice data can be created, for example, by detecting a keyword from the content of the voice recognized in the above step 12 and searching the dictionary data for an item targeted in advance for the keyword. Further, the content of the response voice data may be the content of the voice recognized in step 12 that is inferred using artificial intelligence.

In step 14, based on the answer voice data created in step 13, the speaker 15 is used to make a sound. That is, the answer to the voice acquired in step 11 is pronounced.

In step 15, it is determined whether or not the AI program is being used at that time. If the AI program is being used, the process proceeds to step 16. If the standard program is being used, this process ends.

In step 16, it is determined whether learning is stopped at that time. The learning stop is set by a learning stop determination process described later. If learning is not stopped, the process proceeds to step 17, and if learning is stopped, this process is ended.

In step 17, the learning restriction content set at that time is acquired. As the learning restriction contents, for example, an image including a family (family name) of a user (user corresponding to the identification information determined to be input in step 2) and acquaintance (name, address, telephone number, mail address, history, face) ) Etc.

In step 18, the contents of the voice recognized in step 12 are learned. Examples of learning include machine learning. Machine learning may be supervised learning or unsupervised learning. Further, the learning may be learning by artificial incompetence. In this case, a keyword can be extracted from the voice recognized in step 12 and added to the data set (for example, dictionary data used for voice recognition). This keyword can be used, for example, in the process of creating answer voice data (step 13).

However, even if the content of the voice recognized in step 12 corresponds to the learning restriction content acquired in step 17, the learning is not performed.
In step 19, the AI program and the data set are updated so as to reflect the learning result obtained in step 18. The AI program and data set updated to reflect the learning result is an example of the learning result.

(2-3) Learning Stop Judgment Process The control unit 3 repeatedly executes the learning stop judgment process shown in FIG. 5 at predetermined time intervals. In step 21 of FIG. 5, it is determined whether the AI program is being used. If the AI program is being used, the process proceeds to step 22. If the standard program is being used, this process ends.

In step 22, the position information of the robot 1 is acquired using the GPS 13.
In step 23, the camera 7 is used to acquire an image of the surroundings of the robot 1.
In step 24, the voice is acquired using the microphone 5.

In step 25, it is determined whether or not learning is stopped at that time. Note that the learning stopped state is started in step 28 described later, and ends when learning is restarted in step 30 described later. If learning is not stopped, the process proceeds to step 26, and if learning is stopped, the process proceeds to step 29.

In step 26, it is determined whether or not the image acquired in step 23 or the sound acquired in step 24 causes learning to stop. The trigger for stopping learning is, for example, as follows.

-In the image acquired in step 23, a preset motion as a motion to instruct learning stop is recognized. As the operation, for example, as shown in FIG. 6A, there is an operation in which the index finger is raised and placed in front of the mouth. Further, as another operation, wink can be cited.

-In the voice acquired in step 24, a voice preset as a keyword for instructing to stop learning is recognized. The keywords include, for example, “secret”, “off record”, and “private”. Moreover, as another keyword, for example, names of family members and acquaintances of the user can be cited. The user can register the names of family members and acquaintances as keywords in the robot 1 in advance. Further, the robot 1 infers which word is the name of the user's family or acquaintance based on the voice data recognized in the past, and registers the word inferred as the name of the user's family or acquaintance as a keyword. May be.

The face of a person registered in advance is recognized in the image acquired in step 23. Examples of this person include a user's family and acquaintances. It should be noted that the user can register in advance the face image of the person who causes the learning to stop.

If there is a trigger for stopping learning, the process proceeds to step 28, and if there is no trigger for stopping the learning, the process proceeds to step 27.
In step 27, it is determined whether or not the position information acquired in step 22 corresponds to a position registered in advance as a position where learning should be stopped. Examples of the position where learning should be stopped include the user's home, a conference room, and the like.

The user can previously register the position where learning should be stopped. Further, the robot 1 can infer a place where learning should be stopped based on the past data and register the inferred place. For example, the robot 1 can infer, as a place where learning should be stopped, a place where a trigger for stopping learning is recognized in the past.

In step 28, the learning stop state is started. From this point onward, an affirmative decision is made in step 16 of the conversation process, and learning in step 18 is not performed.
On the other hand, if an affirmative decision is made in step 25, it is decided in step 29 whether or not the image obtained in step 23 or the voice obtained in step 24 provides an opportunity to restart learning. To do. The following are examples of triggers for restarting learning.

-In the image acquired in step 23, a preset motion as a motion for instructing restart of learning is recognized. As the operation, for example, as shown in FIG. 6B, there is an operation of forming a ring with a thumb and an index finger (indicating OK).

-In the voice acquired in step 24, a voice preset as a keyword for instructing restart of learning is recognized. The keywords include, for example, “OK”, “restart”, “learning” and the like.

If there is a trigger for restarting learning, the process proceeds to step 30, and if there is no trigger for restarting learning, this processing ends.
In step 30, learning is restarted. From this point onward, a negative determination is made in step 16 in the conversation process, and the learning in step 18 is performed.

(2-4) Schedule Management Process The robot 1 can execute the following schedule management process. The user inputs his or her schedule information into the robot 1 in advance. The schedule information may be input using, for example, the touch panel 9 or may be input by voice. Further, the schedule information may be transmitted from the terminal 33 to the communication unit 26.

The schedule information includes at least a deadline and items to be executed by the deadline. The robot 1 is scheduled by the user based on the sound acquired by the microphone 5, the image acquired by the camera 7, the information acquired by the terminal 33, and the like at a predetermined time (for example, one day, three hours) before the due date. It is determined whether the items included in the information have been executed, and if they have not been executed, the user is warned by the sound of the speaker 15 or the image displayed on the display 19.

3. Effects of Robot 1 (1A) The robot 1 can change and develop the AI program by learning. Further, the robot 1 can enhance the contents of the data set (for example, dictionary data) by learning. Then, the robot 1 can upload the AI program and data set after learning to the cloud network 31. The robot 1 can also install the AI program and data set uploaded to the cloud network 31.

The robot 1 that installs the AI program and data set after learning may be the same as or different from the robot 1 that has previously learned. Further, the robot 1 which installs the AI program and the data set after learning may be at a place where learning has been performed in the past or may be at a different place.

Therefore, the user uses the cloud network 31 to store the AI program and the data set learned through his/her own use in the robot 1 at the place where the user is at that time (for example, various places such as workplace, store, and home). The robot 1 can be installed and used.

(1B) The robot 1 permits the installation of the AI program and data set corresponding to the identification information, provided that the identification information of the user is input. Therefore, it is possible to prevent other people from arbitrarily using the AI program and data set of a certain user.

(1C) The robot 1 can recognize a voice of a person and pronounce a voice of an answer to the voice. That is, the robot 1 can talk with a person. In addition, when the robot 1 uses the AI program, the robot 1 learns based on the recognition result of the voice and creates an answer voice using the learning result. Therefore, the more advanced the learning, the more advanced the conversation. It can be carried out.

(1D) The robot 1 stops learning in accordance with the movements of people around it and the results of identification of people. Therefore, it is possible to prevent the robot 1 from learning a matter that is not desirable for the user and later talking to another person.

(1E) The robot 1 stops learning according to the place where it exists. Therefore, it is possible to prevent the robot 1 from later talking to another person about an item learned in a place where he/she does not want to learn.

(1F) The robot 1 restarts learning in accordance with the movements of people around it. Therefore, the learning of the robot 1 can be promoted.
(1G) The robot 1 does not learn the matters corresponding to the learning restriction content. Therefore, it is possible to prevent the robot 1 from learning a matter that is not desirable for the user and later talking to another person.

(1H) The robot 1 can manage the schedule of the user.
(1I) The robot 1 erases the AI program and the data set when the use termination condition of the robot 1 is satisfied. Therefore, the user can suppress the use of his/her AI program and data set by another person later.
<Second Embodiment>
1. Configuration of Computer 101 The configuration of the computer 101 will be described with reference to FIGS. 7 and 8. The electrical configuration of the computer 101 is basically the same as that of the robot 1 in the first embodiment. However, the input unit 21 in the computer 101 is connected to the external microphone 5, the camera 7, the keyboard 63, the mouse 65, and the touch panel 9. The output unit 29 is also connected to the external speaker 15 and the display 19.

As shown in FIG. 8, the computer 101 includes a box-shaped casing 67, and the components are housed inside the casing 67. In addition, the computer 101 includes a plurality of terminals 69 to which external devices (for example, the microphone 5, the camera 7, the keyboard 63, the mouse 65, the touch panel 9, the speaker 15, the display 19 and the like) can be connected. The computer 101 is a robot in a broad sense.

2. Processing Executed by Computer 101 The computer 101 executes program setting processing, conversation processing, learning stop determination processing, and schedule management processing, as with the robot 1 of the first embodiment. Further, the computer 101 has the same function as a known computer.

3. Effects of Computer 101 The computer 101 has the effects (1A) to (1I).
<Third Embodiment>
1. Configuration of vehicle-mounted device 201 The configuration of the vehicle-mounted device 201 will be described with reference to FIG. 9. The vehicle-mounted device 201 is mounted on the vehicle. The electrical configuration of the vehicle-mounted device 201 is basically the same as that of the computer 101 in the second embodiment. However, the input unit 21 in the vehicle-mounted device 201 is connected to the microphone 5, the camera 7, the touch panel 9, the sensor group 11, and the GPS 13 provided in the vehicle.

The microphone 5 is provided in the passenger compartment of the vehicle and detects the voice of the vehicle occupant (driver or other occupant). The camera 7 photographs the occupant. The touch panel 9 is provided in the passenger compartment of the vehicle and is operated by an occupant. The sensor group 11 includes a driver's driving operation (steering angle, accelerator depression amount, brake depression amount, shift position, etc.), vehicle state (speed, acceleration, yaw rate, power unit (internal combustion engine, motor, etc.) state, Fuel level, battery level, etc.) and

Moreover, the output unit 29 is connected to the speaker 15, the display 19, and the vehicle control unit 71. The speaker 15 and the display 19 are provided inside the vehicle compartment. The vehicle control unit 71 performs various controls regarding the vehicle (for example, steering, acceleration, deceleration, shift change, etc.). The vehicle-mounted device 201 is a robot in a broad sense.

2. Processing executed by the on-vehicle device 201 (2-1) Processing similar to that of the robot 1 The on-vehicle device 201, like the robot 1 of the first embodiment, has a program setting process, a conversation process, a learning stop determination process, and a schedule. Perform management processing.

(2-2) Vehicle Control Process The control unit 3 of the vehicle-mounted device 201 repeatedly executes the vehicle control process shown in FIG. 10 at predetermined time intervals. This process is executed when the microphone 5 detects a sound volume equal to or higher than a predetermined threshold.

In step 31, voice is acquired using the microphone 5. In step 32, an instruction to the vehicle (for example, start, stop, deceleration, acceleration, right/left turn, lane change, shift change, etc.) is recognized from the voice acquired in step 31 by a known voice recognition technique. At this time, when the standard program is used, the above-mentioned instruction is recognized using the standard dictionary data that the vehicle-mounted device 201 has in advance. When the AI program is used, the above-mentioned instruction is recognized using the dictionary data that is installed from the cloud network 31 together with the AI program and is reinforced by past learning.

In step 33, the control content of the vehicle is determined according to the instruction to the vehicle recognized in step 32. For example, when the instruction to the vehicle is to start, the timing to release the brake, the increase amount of the engine speed, the increase speed, etc. are determined. At this time, if the standard program is used, the control content of the vehicle is determined using the standard program. When the AI program is used, the control content is determined using the AI program that has evolved through past learning.

In step 34, the control content determined in step 33 is output to the vehicle control unit 71. The vehicle control unit 71 controls the vehicle according to the control content.
In step 35, it is determined whether or not the AI program is being used at that time. If the AI program is being used, the process proceeds to step 36. If the standard program is being used, this process ends.

In step 36, the detection result regarding the state of the vehicle is acquired from the sensor group 11.
In step 37, learning is performed based on the control content output in step 34 and the sensor output acquired in step 36. The learning is to correct the control content so that the sensor output acquired in step 36 is within the preset optimum range.

For example, if the control content output in step 34 is a start, it is confirmed whether or not the sensor output (for example, sensor output such as speed or acceleration) in the process of starting is out of the optimum range. If so, the control content of the start is corrected so that the sensor output at the time of the next and subsequent starts approaches the optimum range.

In step 38, the AI program is updated so as to reflect the learning result in step 37.
3. Effects of the in-vehicle device 201 The in-vehicle device 201 has the effects (1A) to (1I). Further, the vehicle-mounted device 201 also has the following effects.

(3A) The vehicle-mounted device 201 can recognize a human voice and perform vehicle control corresponding to the voice. In addition, since the vehicle-mounted device 201 performs learning based on the content of the vehicle control output to the vehicle control unit 71 and the sensor output indicating the state of the vehicle when the AI program is used, as the learning progresses, More advanced vehicle control can be performed.

(3B) The vehicle-mounted device 201 can change and develop the AI program regarding the vehicle control process by learning. Further, the vehicle-mounted device 201 can install the AI program uploaded to the cloud network 31.

The vehicle-mounted device 201 on which the AI program after learning is installed may be the same as or different from the vehicle-mounted device 201 that has learned in the past.
Therefore, the user can install and use the AI program learned through his/her own use in the in-vehicle device 201 of any vehicle from the cloud network 31.
<Fourth Embodiment>
1. Configuration of Robot 301 The configuration of the robot 301 of this embodiment is basically the same as that of the robot 1 of the first embodiment. Below, it demonstrates centering around difference with 1st Embodiment. As shown in FIG. 11, the robot 301 includes a first artificial intelligence unit 73 and a second artificial intelligence unit 75.

The first artificial intelligence unit 73 stores an AI program and a data set for executing various operations of the robot 301. Hereinafter, this program will be referred to as a first AI program, and this data set will be referred to as a first data set.

The first artificial intelligence unit 73 changes and develops the first AI program and the first data set by learning. The learning performed by the first artificial intelligence unit 73 is not affected by learning stop and learning restriction, which will be described later. Further, the first AI program and the first data set are not uploaded to the cloud network 31.

The second artificial intelligence unit 75 is capable of storing AI programs and data sets. Hereinafter, this program will be referred to as a second AI program, and this data set will be referred to as a second data set.

The second AI program is basically the same as the first AI program, but is not used to execute various operations of the robot 301. The second artificial intelligence unit 75 changes and develops the second AI program and the second data set by learning. The second AI program and the second data set accumulate learning results and are uploaded to the cloud network 31. Thereby, the learning result can be stored in the cloud network 31.

However, the learning performed by the second artificial intelligence unit 75 is limited by learning stop and learning limitation described later. Therefore, the learning result stored in the cloud network 31 by uploading the second AI program and the second data set is limited.

The second AI program and the second data set can be downloaded from the cloud network 31 to the second artificial intelligence unit 75. Then, the learning of the first AI program and the first data set can be performed based on the downloaded second AI program and the second data set. Details will be described later.

2. Processing Performed by Robot 301 (2-1) Program Installation Processing The control unit 3 executes the program installation processing shown in FIG. 12 when the power of the robot 301 is turned on.

In step 41, it is determined whether the identification information of the user has been input. The identification information can be the same as that in the first embodiment. If the identification information has been input, the process proceeds to step 42, and if the identification information has not been input, this processing ends.

In step 42, the second AI program and the second data set corresponding to the identification information determined to be input in step 41 are installed from the cloud network 31. The installed second AI program and the second data set are stored in the second artificial intelligence unit 75. If the second AI program and the second data set are already stored in the second artificial intelligence unit 75, they are overwritten.

In step 43, contents included in the second AI program and the second data set installed in step 42 but not stored in the first AI program and the first data set are stored. learn.

In step 44, the contents learned in step 43 are added to the first AI program and the first data set and updated. That is, the learning of the first AI program and the first data set is performed based on the downloaded second AI program and the second data set.

(2-2) Conversation processing The control unit 3 executes the conversation processing shown in FIG. This process is executed when the microphone 5 detects a sound volume equal to or higher than a predetermined threshold. The conversation process is performed using the first AI program and the first data set stored in the first artificial intelligence unit 73.

The processing of steps 51 to 54 is the same as the processing of steps 11 to 14 in the first embodiment.
In step 55, the first artificial intelligence unit 73 learns the content of the voice recognized in step 52. The contents of learning are the same as those in the first embodiment.

In step 56, the first AI program and the first data set are updated to reflect the learning result in step 55.
In step 57, it is determined whether the second artificial intelligence unit 75 is currently in the learning stopped state. The learning stop of the second artificial intelligence unit 75 is set by a second artificial intelligence unit learning stop determination process, which will be described later. If the second artificial intelligence unit 75 is not in the learning stopped state, the process proceeds to step 58, and if the learning is stopped, the process ends.

In step 58, the learning restriction content set at that time is acquired. As the learning restriction content, for example, an image including family (family name), acquaintance (name, address, telephone number, mail address, history, face) of the user (the user corresponding to the identification information determined to be input in step 41) ) Etc.

In step 59, the second artificial intelligence unit 75 learns the content of the voice recognized in step 52. The contents of learning are the same as those in the first embodiment. However, even if the content of the voice recognized in step 52 is applicable, the matter corresponding to the learning restriction content acquired in step 58 is not learned.

In step 60, the second AI program and the second data set are updated so as to reflect the learning result in step 59.
(2-3) Second Artificial Intelligence Unit Learning Stop Judgment Process The control unit 3 repeatedly executes the second artificial intelligence unit learning stop judgment process shown in FIG. 14 every predetermined time.

In step 71, the position information of the robot 301 is acquired using the GPS 13.
In step 72, the camera 7 is used to acquire an image of the surroundings of the robot 301.

In step 73, the voice is acquired using the microphone 5.
In step 74, it is determined whether the second artificial intelligence unit 75 is currently in the learning stopped state. Note that the learning stopped state is started in step 77 described later, and ends when learning is restarted in step 79 described later. If learning is not stopped, the process proceeds to step 75, and if learning is stopped, the process proceeds to step 78.

In step 75, it is determined whether or not the image acquired in step 72 or the sound acquired in step 73 causes learning to stop. The trigger for stopping learning is the same as in the first embodiment. If there is a trigger to stop learning, the process proceeds to step 77, and if there is no trigger to stop the learning, the process proceeds to step 76.

In step 76, it is determined whether or not the position information acquired in step 71 corresponds to a position registered in advance as a position where learning should be stopped. The position at which learning should be stopped is the same as in the first embodiment.

In step 77, the learning stop state is started for the second artificial intelligence unit 75. After this point, an affirmative decision is made in step 57 of the conversation process, and the learning in step 59 is not performed.

On the other hand, if an affirmative decision is made in step 74, then it is decided in step 78 whether or not the image acquired in step 72 or the sound acquired in step 73 is a trigger for restarting learning. To do. The trigger for restarting learning is the same as that in the first embodiment. If there is a trigger for restarting learning, the process proceeds to step 79, and if there is no trigger for restarting learning, this processing ends.

In step 79, the learning in the second artificial intelligence unit 75 is restarted. From this point onward, a negative determination is made in step 57 in the conversation process, and the learning in step 59 is performed.

(2-4) Program Upload Process The control unit 3 repeatedly executes the program upload process shown in FIG. 15 at predetermined time intervals.

In step 81, it is determined whether or not the use termination condition of the robot 301 is satisfied. The use end condition is the same as that determined in step 5 of the first embodiment. If the use end condition is satisfied, the process proceeds to step 82, and if the use end condition is not satisfied, this process ends.

In step 82, the second AI program and the second data set are uploaded to the cloud network 31. At this time, it is uploaded in association with the user identification information. When the above-mentioned learning is performed, the second AI program and the second data set to be uploaded are those after learning.

3. Effect of Robot 301 The robot 301 has the effects of (1B), (1C), (1F), and (1H). Furthermore, the robot 301 also has the following effects.

(4A) The robot 301 can change and develop the first AI program and the first data set by learning. Further, the learning in the first AI program and the first data set is not affected by the learning stop processing and the learning restriction processing.

Further, since the first AI program and the first data set are not uploaded to the cloud network 31, it is possible to suppress the contents from being known to others.
(4B) The robot 301 can upload the learned second AI program and the second data set to the cloud network 31. Further, the robot 301 can install the second AI program and the second data set from the cloud network 31.

The robot 301 that installs the second AI program and the second data set after learning may be the same as or different from the robot 301 that has learned in the past. In addition, the robot 301 that installs the second AI program and the second data set after learning may be at the place where the learning was performed in the past or may be at a different place. Good.

Therefore, the user uses the robot 301 at the place where the user is at the time (for example, various places such as a workplace, a store, a home) of the second AI program and the second data set learned through his/her own use in the past. In addition, it can be installed from the cloud network 31. Then, learning can be performed using the second AI program and the second data set to change and develop the first AI program and the first data set.

(4C) The robot 301 restricts learning about the second AI program and the second data set uploaded to the cloud network 31. That is, the robot 301 stops learning about the second AI program and the second data set according to the motion of the people around it, the identification result of the people, and the like. Further, the robot 301 stops learning in the second AI program and the second data set, depending on the place where the robot 301 exists. Further, the robot 301 does not include the items corresponding to the learning restriction contents in the second AI program and the second data set.

Therefore, it is possible to suppress uploading of the second AI program and the second data set, which include items that the user does not want others to know, to the cloud network 31. <Fifth Embodiment>
1. Configuration of Robot 401 The configuration of the robot 401 of this embodiment is basically the same as that of the robot 1 of the first embodiment. Below, it demonstrates centering around difference with 1st Embodiment. The robot 401 can be connected to the Internet 77 using the communication unit 26, as shown in FIG. The Internet 77 is an example of a network. Further, the robot 401 can be placed inside the vehicle. In this case, the robot 401 can have a conversation with an occupant of the vehicle by a process such as a conversation described later.

2. Processing executed by the robot 401 (2-1) Processing similar to that of the robot 1 The robot 401 executes program setting processing, learning stop determination processing, and schedule management processing, as in the robot 1 of the first embodiment. ..

(2-2) Conversation etc. Processing The control unit 3 of the robot 401 repeatedly executes the conversation etc. processing shown in FIG. This process is executed when the microphone 5 detects a sound volume equal to or higher than a predetermined threshold.

The processing of steps 91 and 92 is the same as the processing of steps 11 and 12 in the first embodiment, respectively.
In step 93, the information related to the content of the voice recognized in step 92 is searched and obtained on the Internet 77 using the communication unit 26. For example, a keyword is extracted from the content of the voice recognized in step 92, and information including a matter previously associated with the keyword is searched. Examples of the search target include SNS (social networking service), blog, electronic bulletin board, and the like. If the voice content recognized in step 92 is a question, the answer to the question is searched. As a search method, for example, a known search engine can be used.

In step 94, basically, similar to step 13 in the first embodiment, answer voice data is created for the content of the voice recognized in step 92. However, in this step, the answer voice data is created by using the information acquired in step 93. Using the information obtained in step 93 means, for example, including the voiced version of the information in the response voice data.

In step 95, the type of voice to be pronounced in step 96, which will be described later, is determined according to the content of the answer created in step 94. Examples of the type of voice include male voice, female voice, adult voice, and child voice.

Specifically, the type of voice is determined as follows. First, the control unit 3 extracts a characteristic (for example, a characteristic peculiar to a man, a characteristic peculiar to a woman, a characteristic peculiar to an adult, a characteristic peculiar to a child, etc.) from the contents of the answer created in step 94. ..

The control unit 3 previously includes a map in which the characteristics of the content of the answer and the type of voice are associated with each other. The control unit 3 inputs the feature extracted as described above into the map to determine the type of voice corresponding to the extracted feature.

For example, if the characteristic peculiar to male and the characteristic peculiar to adult are extracted from the content of the answer created in step 94, the control unit 3 determines the voice of the adult male.
In step 96, the speaker 15 is used to produce sound based on the answer voice data created in step 94. At this time, the voice of the type set in the step 95 is used to pronounce.

In step 97, first, the content of the voice recognized in step 92 or the emotion (for example, joy, anger, sadness, calmness, etc.) corresponding to the content of the answer created in step 94 is acquired. This emotion is acquired as follows.

The control unit 3 is provided with a map that associates emotions with the features that appear in the content of the voice and the content of the answer in advance. The features that appear in the content of the voice include, for example, volume, voice intonation, pitch of the pitch, and the like. In addition, examples of features that appear in the content of the answer include words and phrases that reflect the emotions, phrases that reflect the emotions, and the like that are included in the content of the response.

The control unit 3 acquires the corresponding emotion by inputting the content of the voice recognized in step 92 or the feature extracted from the answer created in step 94 into the map.

Next, the control unit 3 creates a face image of a person or a character expressing the acquired emotion and displays it on the display 19. For example, when the acquired emotion is joy, a face image of a smiling person or character is displayed on the display 19. When the acquired emotion is sadness, a face image of a person or a character with a crying face is displayed on the display 19.

The processing of steps 98 to 102 is the same as the processing of steps 15 to 19 in the first embodiment, respectively.
3. Effects of Robot 401 The robot 401 has the effects (1A) to (1I). Furthermore, the robot 401 also has the following effects.

(5A) The robot 401 can search and acquire information related to the content of voice on the Internet 77. Then, the robot 401 creates reply voice data by using the information thus obtained. Thereby, more appropriate answer voice data can be created.

(5B) The robot 401 can determine the type of voice according to the content of the answer. Therefore, the user is less likely to feel a dissonance between the content of the answer and the type of voice that sounds it.

(5C) The robot 401 acquires an emotion corresponding to the content of the recognized voice or the content of the answer, and displays a face image of a person or a character expressing the acquired emotion on the display 19. As a result, the user can feel the robot 401 as if it were a human.
<Sixth Embodiment>
1. Configuration of Robot 501 The configuration of the robot 501 of this embodiment is basically the same as that of the robot 1 of the first embodiment.

2. Processing Executed by Robot 501 (2-1) Processing Similar to Robot 1 The robot 501 has the same program setting processing, conversation processing, learning stop determination processing, and schedule management processing as the robot 1 of the first embodiment. To execute.

(2-2) Call Processing As shown in FIG. 18, the plurality of robots 501 enable a call between the user 79 and the user 81. For example, the robot 501 on the side of the user 79 (hereinafter referred to as the robot 501A) acquires the voice produced by the user 79 with the microphone 5. In addition, the robot 501A receives, via the touch panel 9, an input by the user 79 for designating the robot 501 (hereinafter, robot 501B) of the other party of the call.

The robot 501A communicates a signal obtained by converting the voice produced by the user 79 (hereinafter referred to as a voice conversion signal), the identification signal of the robot 501A, and the identification signal of the robot 501B of the communication partner designated by the user 79. It is transmitted to the host computer 83 using the unit 26.

The host computer 83 determines whether or not the received identification signal of the robot 501A and the received identification signal of the robot 501B of the call partner are set as a call pair. If it is set as a call pair, the voice conversion signal is transferred to the robot 501B of the call partner. The setting of the call pair will be described later.

The robot 501B receives the transferred voice conversion signal using the communication unit 26, and produces a sound using the speaker 15 based on the voice conversion signal. Through the above processing, the voice pronounced by the user 79 is transmitted to the user 81 on the robot 501B side. In addition to pronouncing a voice, the robot 501B also converts the voice into a character and displays the character on the display 19.

Further, the robot 501B performs the above-described processing of the robot 501A and the robot 501A performs the above-described processing of the robot 501B, whereby the voice pronounced by the user 81 can be transmitted to the user 79.

On the other hand, the host computer 83 does not transfer the voice conversion signal unless the received identification signal of the robot 501A and the received identification signal of the robot 501B of the call partner are set as a call pair.

(2-3) Call Pair Setting Process Next, a process in which the host computer 83 sets the above call pair will be described with reference to FIG.

In step 111, it is determined whether or not a call pair setting request has been newly received. The call pair setting request is a request transmitted from the robot 501A to the host computer 83. In the call pair setting request, the identification signal of the robot 501A that is the transmission source of the request, the characteristics of the user 79 on the side of the robot 501A (for example, hobbies, food/living/housing preferences, matters of interest, etc.), and the user 79's The image data of the face photograph is included.

The call pair setting request may be transmitted by the robot 501A or automatically by the robot 501A according to a user's instruction. When the host computer 83 newly receives the call pair setting request, the process proceeds to step 112, and when it is not received, this process ends.

In step 112, the characteristics of the user 79 included in the call pair setting request (hereinafter referred to as a new setting request) determined to be received in step 111 are searched in the setting waiting list. Here, the setting waiting list is a list of call pair setting requests (hereinafter, referred to as past setting requests) received from any of the robots 501 in the past.

In step 113, as a result of the search in step 112, it is determined whether or not a past setting request having a characteristic that matches the characteristic of the user 79 included in the new setting request is found. If such a past setting request is found, the process proceeds to step 114, and if not, the process proceeds to step 116.

In step 114, the robot 501B, which is the transmission source of the past setting request, and the robot 501A, which is the transmission source of the new setting request, found in step 113 are set as a call pair.

In step 115, each of the robots 501A and 501B set as the call pair is notified of information regarding the partner robot 501. That is, the robot 501A is notified of information about the robot 501B, and the robot 501B is notified of information about the robot 501A. The information to be notified includes the identification signal of the robot 501 of the other party, the image data of the corresponding face photograph of the user, and the like.

On the other hand, if a negative determination is made in step 113, a new setting request is added to the setting waiting list in step 116. Thereafter, the new setting request becomes a part of the list of past setting requests.

The robot 501 notified of the information about the partner robot 501 uses the information to create an icon 85 representing the other party of the call pair, and displays it on the display 19 as shown in FIG. The icon 85 includes a face picture of the user on the other side of the call pair. The robot 501 can display a plurality of icons 85 on the display 19.

When the specific icon 85 is touched by the user, the robot 501 recognizes the user corresponding to the icon 85 as a call partner. Then, when transmitting the voice conversion signal to the host computer 83 as described above, the robot 501 transmits the identification signal of the robot 501 corresponding to the touched icon 85 to the host computer 83.

(2-4) Other processing The host computer 83 analyzes the content of the voice conversion signal transmitted by the robot 501A during the call processing of (2-2). When the analysis result corresponds to a preset prohibited item (for example, a crime-related item), the call process is not started or the call process is terminated halfway. Therefore, the host computer 83 can suppress the call processing from being used for crimes and the like.

Further, in the call processing of (2-2), the robot 501B also performs learning by using the voice transmitted from the other computer 501A. Therefore, learning can be performed more efficiently.

Further, in the call processing of (2-2), when the pronunciation of the user 79 is interrupted for a predetermined time or more, or when the user 79 gives an instruction, the robot 501A sends the voice conversion signal created by itself to the host computer 83. Send. In this case, the robot 501A and the user 81 will talk. The computer 501A can create a voice conversion signal using a learning result obtained by past call processing. Further, the robot 501A may store a voice (for example, a hammer etc.) sounded by the robot 501A or the robot 501B in a past call and create a voice conversion signal corresponding to the voice.

3. Effects of Robot 501 The robot 501 has the effects (1A) to (1I). Furthermore, the robot 501 also has the following effects.

(6A) The user 79 on the robot 501A side and the user 81 on the robot 501B side can talk.
(6B) The call between the user 79 and the user 81 is premised on that the robot 501A and the robot 501B are set as a call pair. The call pair is set when the characteristics of the user 79 and the user 81 match. Therefore, the robot 501 selectively enables a call between users having the same feature.

(6C) The robot 501 creates an icon 85 including a facial photograph of the user who is the other party of the call and displays it on the display 19. The user can easily select the other party by touching the icon 85. In addition, since the icon 85 includes a facial photograph of the call partner, the user can easily understand which icon 85 corresponds to which call partner.
<Other embodiments>
(1) In the first to sixth embodiments, the robots 1, 301, 401, 501, the computer 101, and the vehicle-mounted device 201 may increase or decrease the learning restriction content according to the situation. For example, when it is recognized that an item originally belonging to the learning restriction content is spoken by a plurality of persons, the item may be excluded from the learning restriction content.

(2) In the first to sixth embodiments, the robot 1, 301, 401, 501, the computer 101, and the vehicle-mounted device 201 determine the voice color of the voice when the voice is acquired by the microphone 5, and the determination is made. Depending on the result, it may be decided whether or not to learn the content of the voice.

(3) In the first to sixth embodiments, the processing performed according to the voice recognition result is something other than the creation of answer voice data (step 13) and the determination of the control content of the vehicle (step 33). It may be. For example, the robots 1, 301, 401, 501 may be moved or an external device connected to the computer 101 may be operated according to the result of voice recognition. In addition, for example, a predetermined operation (for example, opening/closing a box, opening/closing a window, locking/unlocking a key, operating a home electric appliance, etc.) may be performed according to the result of voice recognition.

(4) In the first to sixth embodiments, the one that stores the AI program and the data set may be something other than the cloud network 31. For example, it may be stored in a known server, storage medium, or the like.

(5) In the third embodiment, the in-vehicle device 201 may be mounted on a moving body other than a vehicle (for example, a railway vehicle, an aircraft, a ship, etc.) and control them.
(6) In the first and fourth to sixth embodiments, the forms of the robots 1, 301, 401, 501 do not have to be humanoid. For example, it may be in the form of an animal, a fish, or an imaginary character.

(7) In the first to third, fifth, and sixth embodiments, the robots 1, 401, 501, computer 101, and vehicle-mounted device 201 may use the standard program and the AI program at the same time. .. In this case, the standard program can execute the basic processing, and the AI program can execute the additional processing obtained as a result of the learning.

(8) In the first to sixth embodiments, the forms of the robots 1, 301, 401, 501, the computer 101, and the vehicle-mounted device 201 are home electric appliances (for example, TV, refrigerator, air conditioner, vacuum cleaner, washing machine, etc.). ), a mobile terminal (for example, a mobile phone (including a smartphone), a glasses-type terminal, a wristwatch-type terminal) and the like.

(9) In the first to sixth embodiments, the robots 1, 301, 401, 501, the computer 101, and the vehicle-mounted device 201 automatically acquire images using the camera 7, and acquire the acquired images on the network. It may have a function of sending to. The transmission destination may be a server, a terminal, a vehicle, another robot, or the like. The image to be transmitted may be the image itself obtained by using the camera 7, or a part extracted from the image obtained by using the camera 7 (for example, a person's face, a person's whole body, a vehicle, a vehicle number). It may be an image of a plate or the like).

The robots 1, 301, 401, 501, the computer 101, and the vehicle-mounted device 201 having the above-described functions can be used for crime prevention, monitoring of other robots, and the like. The robots 1, 301, 401, 501, the computer 101, and the vehicle-mounted device 201 having the above-mentioned functions are, for example, indoors, on the road, at an entrance of a specific facility (for example, a residence, a condominium, an office, a parking lot, etc.) or another robot. It can be installed in the vicinity of.

Further, the robots 1, 301, 401, 501, the computer 101, and the vehicle-mounted device 201 having the above-described functions have the function of flying in the air or can be mounted on a flying object. In that case, an image of the ground taken from the viewpoint of the sky can be transmitted on the network. In this case, the robots 1, 301, 401, 501, the computer 101, and the vehicle-mounted device 201 can recognize the white line of the road in the captured image and move along the white line. In addition, the robots 1, 301, 401, 501, the computer 101, and the vehicle-mounted device 201 recognize the traffic light in an image on the ground taken from a viewpoint in the sky, and display the contents (red signal, blue signal, etc.) on the vehicle on the ground. Can be sent.

(10) In the first to sixth embodiments, the robots 1, 301, 401, 501, the computer 101, and the vehicle-mounted device 201 recognize the action of a person using the camera 7 and are associated with the recognition result in advance. You may output the sound.

For example, when a person recognizes the action of leaving the kitchen with the gas ignited, a warning sound for the person can be output. In addition, when a person recognizes an action of searching for a predetermined object, the person can search for the place of the object and inform by voice of the existence of the object. In addition, when a person recognizes the action of entering the room from the entrance of the house, the voice "Welcome back" can be output. A person's behavior and voice associated with it can be increased by learning.

When a voice is emitted as described above, the type of the voice may be associated with the situation at that time and the content of the voice. For example, when the father is recognized in the image, the type of voice "Welcome back" can be the voice of the child.

(11) In the first to sixth embodiments, the control unit includes a microcomputer, but may be a combination of individual electronic circuits or an AISI (Application Specific Integrated Circuit). It may be a programmable logic device such as FPGA (Field Programmable Gate Array) or a combination thereof.
(12) Part or all of the configurations of the first to sixth embodiments may be combined appropriately. For example, the configurations of the fourth to sixth embodiments may be applied to the second and third embodiments.

Claims (5)

A voice capture unit configured to capture voice of a vehicle occupant;
An instruction recognition unit configured to recognize an instruction to the vehicle from the voice acquired by the voice acquisition unit,
In response to the instruction recognized by the instruction recognition unit, a control content determination unit configured to determine the control content of the vehicle using a program,
A control content output unit configured to output the control content determined by the control content determination unit to a vehicle control unit that controls the vehicle;
A vehicle condition acquisition unit configured to acquire a condition of the vehicle using a sensor;
A learning unit configured to perform learning based on the control content output by the control content output unit and the vehicle state acquired by the vehicle state acquisition unit,
A program update unit configured to update the program to reflect a learning result of the learning unit;
Equipped with
The learning performed by the learning unit is to correct the control content of the vehicle determined by the control content determination unit so that the vehicle state acquired by the vehicle state acquisition unit falls within a preset optimum range. Is an in- vehicle device. The vehicle-mounted device according to claim 1,
The in-vehicle device, wherein the instruction is one or more selected from the group consisting of start, stop, deceleration, acceleration, right/left turn, lane change, and shift change. The vehicle-mounted device according to claim 1 or 2, wherein
An in-vehicle device further comprising an installation unit configured to install the program from the cloud. The vehicle-mounted device according to any one of claims 1 to 3,
Further comprising a standard program that is not updated by the program update unit,
The in-vehicle device configured such that the control content determination unit can determine the control content of the vehicle by using the standard program instead of the program. A vehicle comprising the vehicle-mounted device according to any one of claims 1 to 4.

Images