JP2003157489A - Operation control device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an autonomous control device and an agent device enabling a user to simply produce and add an autonomously corresponding content. SOLUTION: In the agent device, an agent image comprising a prescribed figure is displayed in a vehicle. A function for recognizing and deciding the surrounding state (including the motion or the voice of a person) from a detection result of a sensor or the like and outputting the motion or the voice corresponding to the result, which is a function of the agent device, is executed in connection with the motion or the voice of the figure of the agent. Hereby, the user feels such bodily sensation as if an agent having a pseudo character exists in the vehicle. Execution of a series of functions of the agent device is explained hereafter as an action or a motion of the agent. In the agent device in this execution form, the agent is allowed to perform various communications with a driver or deputization of an operation.

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an operation control device.
For example, by communicating with the vehicle occupant.
Autonomous control device that autonomously talks and operates equipment
Related to installation. 2. Description of the Related Art Sensors for detecting various states and state recognition.
Recognition devices are becoming more diversified and more accurate,
Furthermore, various types of equipment are used using the detection results and recognition results
With the development of control technology that controls the movement and display of
Information that controls various devices and interacts with users
Processing devices have been put to practical use. Such information processing device
For example, from the detection result of a sensor or the like, the surrounding state
(Including human movements and voices)
Autonomous, such as an autonomous robot that outputs a corresponding action or voice
There is a type control device. As an autonomous control device,
For example, pet-type robots such as dogs,
Provides guidance on operation of devices such as navigation devices and responds to situations.
Dialogue and response with the passengers by giving
Agent devices have been developed. In such an autonomous control device, the device
Assuming various possible states for
The actions to be taken when each of the states
Data and programs. And self
Program for autonomous judgment and response
System should be accepted by as many users as possible.
Developed assuming. For example, a pet robot
If so, not only do cute gestures but sometimes get angry
To make the user more friendly
Be programmed. In addition, agent equipment for vehicles
If it is a place, it detects midnight and proposes a dining room around the current location
Refueling by detecting the gasoline level
Prompted or detected as a pre-registered user hobby
Guide the start of the corresponding radio broadcast from the time
Is programmed as [0004] However, the conventional autonomous system
The controls are capable of handling many situations.
As much user satisfaction as possible,
Assuming the situation, plan as many types of responses as possible
Needs to be programmed. Against such a situation setting
Since the device developers are all responsible for the
And cost was high. Also develop
The program is accepted by as many users as possible
It is created assuming a situation like
May not be able to accept the content of autonomous response.
In such cases, depending on the user's usage and preferences,
Change hardware and software settings and functions
You will use the customize function, but customize
Possible contents are limited to those provided by the device developer
Have been. The best judgment and response for the user
In order to make an autonomous control device, each user must
Although it is desirable to program according to
Specialization in programming in assembly language, C language, etc.
Knowledge and environment were required. For this reason, ordinary users
Can autonomously respond to their preferences
Even if your programming knowledge matches,
Not configured to allow program rewriting
There were many devices. For example, in the agent device, the vehicle
And perform various actions according to the driver's condition.
The details of the correspondence and processing performed by the driver
Limited to actions according to the program set in the system
I have. And the user has added an agent action
Service that the user wishes to provide
Or the service provider adds an agent's action
Could not. The present invention has been made to solve the above problems.
Can respond autonomously to humans
It is a first object of the present invention to provide an operation control device that can be used. Ma
In addition, the control unit asked the human
Enter data that specifies the content of the question from the outside
The second is to provide an operation control device capable of performing
Aim. [0006] The invention according to claim 1
Now, perform a predetermined operation with an interactive interface
The interactive interface.
The control unit asks the human to ask for an answer.
A first step to be performed and a human being according to the first step
Of the answer to determine the content of the answer
And the contents of the question to the human being
An input device that enables input of prescribed data from outside,
At least according to the data input by the input device.
Also controlling the execution means for executing the first step.
The equipment is equipped. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an autonomous control device according to the present invention will be described.
Various types of navigation, including navigation device settings, etc.
An example of a vehicle agent device that responds
Embodiments will be described in detail with reference to FIGS. 1 to 27.
explain. (1) Overview of the Embodiment In the agent device of the present embodiment, the
Agent images (such as planar images, holography, etc.)
3D image) is displayed in the vehicle. And Ege
From the detection results of sensors, etc.
Recognize and judge the state (including human movements and voices)
The function of outputting actions and voices according to the results
-Executes in conjunction with the appearance movement and voice of the agent. An example
For example, "What genre do you like?"
(Japanese food, Western food, etc.)
Judge the contents of the user's response to the
Or from the selection of the answer selection button 54a), and
Executes processing according to the scene. Thus, from the device
Question that requires a response from
The user starts executing the operation of
Agent with personality is present in the vehicle
You will have a natural experience. In the following description,
The execution of a series of functions of the agent device
The actions and actions are described as follows. Agent of the present embodiment
The communication device provides this agent with various communications with the driver.
Communication and delegate operations. And age
Various actions (each action) that the agent performs autonomously
Consists of Nario. And a series of agents
Multiple scenarios that define the content of a series of actions and each scenario
Standardized by starting conditions for starting Nario deployment
Save the scenario data. Scenario is a scene
The minimum unit is composed of one or a plurality of continuous scenes.
It is. Autonomous processing contents and agent images and
A scene composed of at least one sound is one scene
is there. Scenario moves from one scene to the next
One transition condition (continuous condition) or multiple transition conditions
Condition (branch condition (for each state when multiple states occur)
Conditions for the transition of the scene)))
In response to the destination data that identifies the destination scene,
The development configuration of each scene is defined. On the other hand, the user of the agent device
Create your own scenario according to specified standards
Create using device. Scenario creation device
O Editing programs and data on personal computers
It can be configured by installing. make
Scenario is over a network such as the Internet.
Sent to the agent device or downloaded
And through a predetermined semiconductor memory
By storing it in a remote device,
Agent for desired actions (communication and processing)
Event. Of the scenario to create
Ask the first scene to ask for answers to humans
Scene scenario. And your own
From the external device to the input device
It is possible to input via. In this case, the input device
Stores the scenario data to be added to the semiconductor memory.
Read the contents of the semiconductor memory, if any
The media drive unit corresponds to a network such as the Internet.
Download from a specific server, etc.
Denotes a communication control device. This newly entered scenario
Execute the scenario according to the data. [0009] The agent device is a mail, browser software.
Software, and use a browser
And download the file.
The scenario data to make the agent function
Agent program for scenario data
So that it can be used by incorporating it. Also, add
Similarly, if the attached file is attached to the scenario
And if it is a scenario, the agent system
System for use. This
In addition to the traditional method, the existing communication infrastructure e-mail,
By using communication over networks, etc.,
In addition, through a semiconductor memory, etc.,
Add your own scenario created using a computer, etc.
Will be able to do it. In this way, the user
Scenarios to make agents work as desired
Also, since it can be easily created, agents
There is no resistance to autonomous operation of the device. (2) Details of Embodiment FIG. 1 is composed of an agent device and a scenario creating device.
This shows the overall system configuration. This system
Now, the agent device 1 of the present embodiment and the designated
Users who create scenarios with standards or third parties
Using Rio creator's scenario creation device 2 and server 3 etc.
And communication means such as the Internet.
Scenario creation device 2 uses the scenario editor
Create scenario data. And your own scenario
Creates a scenario in the semiconductor memory 7
I do. And the storage medium drive of the agent device 1
Reads the scenario from the semiconductor memory 7 and
You can create your own by incorporating it into a scenario that you remember.
The agent device 1 according to the scenario
It becomes possible. In the agent device 1,
Scenarios created by third parties can be transferred to Internet
Can be incorporated through the work. Also,
Wants to provide services to the user of the client device 1.
The third party that describes the scenario in a predetermined format
Create with the scenario creation device 2 using an editor,
On the homepage to make it downloadable,
Is the agent device 1 as an e-mail attachment
Send to Agent device 1 attached to e-mail
Received scenario 5 or the server 3
Download scenario file 4 via communication means such as
Is to be loaded. Also, the agent device 1
Is the user obtained according to the execution of the received scenario
In the body of the email 6 or in the attached file
It is transmitted to the scenario creator 2 of the narrator. First, a file created by a developer or user
Agents that function autonomously according to scenarios
The configuration and operation of the gent device will be described. Figure
2 shows the configuration of the agent device in the present embodiment.
FIG. Agent device 1 of the present embodiment
Now, the overall process that controls the entire communication function
A section 9 is provided. This overall processing unit 9 is used for the set purpose.
Search the route to the ground and provide guidance by voice or image display
Navigation processing unit 10, agent processing unit 11,
Navigation processing unit 10 and agent processing unit 11
External I / F unit 12, agent image and map image
Image processing unit 13 for processing image output and input images such as
Voice output and input such as
Voice control unit 14 for controlling the voice of the vehicle and the driver.
Information processing unit 1 that processes detection data of various situations
5, having an input control unit 16 and a storage device control unit 17
I have. Navigation processing unit 10 and agent
The processing unit 11 performs data processing and controls the operation of each unit.
A CPU (Central Processing Unit) and a data bus and
ROM, RAM connected by a bus line such as a control bus,
A timer and the like are provided. Both processing units 10 and 11 are network
Network connection (in-vehicle LAN (Local Area Net)
work)) to obtain the processing data of each other
You can do it. In the present embodiment,
The agent processing unit 11 performs navigation according to the scenario.
Information (such as destination data and travel route data)
Information from an external device such as a reporting center, or
Communication with the user based on the
Once you have obtained the specified destination,
The data is supplied to the gating processing unit 10. Na
In the navigation processing unit 10, the route is determined according to the acquired purpose.
Driving route obtained by searching, route search and reception
Provides route guidance based on the data. In this route guidance process
To provide guidance on course change directions, etc.
In such a case, the data necessary for the guidance
Supply to the agent processing unit 11 from the
It is also possible for agents to guide according to Nario
You. The ROM stores various data and programs for control by the CPU.
Is a read-only memory where the program is stored in advance.
RAM is used by the CPU as working memory
It is a random access memory. The navigation processing unit 10 of the present embodiment
The agent processing unit 11 has a CPU
Read various programs and execute various processes
It has become. Note that the CPU is a storage medium driving device 23.
Computer program from an external storage medium
Read the ram and read the agent data 30
Data 31 such as a hard disk not shown
Store (install) in another storage device, this storage device
Read necessary programs from RAM into
May be executed. In addition, necessary programs
Reads ram, etc. directly from storage medium drive 23 to RAM
May be executed. In this embodiment, the agent processing unit 1
1 is various communication of agents including conversation with driver
The communication operation is performed in various situations (state) of the vehicle or driver.
It will be performed according to the scenario created in advance
Swelling. That is, vehicle speed, time, running area,
Various conditions such as temperature, gasoline remaining amount, driver operation, etc.
As a starting condition to start Rio deployment, each situation will be autonomous
The content of the agent's response to judge and respond
A scenario is defined for each situation. In addition,
Starting deployment of a scenario is called launching a scenario.
The start condition for starting Rio deployment is called the scenario start condition.
U. Each scenario is made up of multiple consecutive scenes
I have. A scene is one scene in a scenario,
The agent asks one question for each scene,
And provide information. Each scene has a title, a squirrel
, Balloons, backgrounds, and other small units (parts)
Screens and voices of agents.
Each scene proceeds sequentially according to a scenario. Sina
Depending on the Rio, there are questions
Multiple choices depending on the driver's answer, vehicle conditions, etc.
Scene exists. In other words, depending on the answer in the middle,
There is a scenario where the branch branches. Scene branch exists
In existing scenarios, depending on user response and vehicle conditions, etc.
Will develop into different scenes. In a scenario where there is a branch, each scene is
They are arranged in a hierarchical structure. Specific scenarios
When o is executed, scene 1 of the first layer is developed
(E.g., with a print image or sound). Scene 1 ends
Then, the processing shifts to scene 2 of the second layer. This second layer
Scenes, for example, scenes 2a and 2b exist,
User response to scene 1 (including answers to questions)
) And the condition of the vehicle as branch conditions,
The second layer scene is determined. For example, if the branch condition is
User answers "yes" and "no" for scene 1
When the user's answer is "yes", the scene 2a is determined,
In the case of "No", the scene 2b is determined and the determined scene
Move to the Branch conditions and migration for each branch condition
The previous scene is defined for the scene before branching. Data of a scenario including a scene will be described later.
It is stored in the scenario data file 302. When
Information that defines where the scenario will be executed (scenario
E start condition) and what screen configuration when executed
Let the agent do what actions and conversations,
What about modules such as the gating processor 10
What to do next when you receive an event and receive an event
Data that defines what scene
The scenario data file 302
These are stored as nario data 1 to n. FIG. 3 shows the configuration of the agent processing unit 11.
It is a representation. As shown in FIG.
The event processing unit 11 executes the scenario determination unit 111 and executes the scenario.
Request determination unit 112, scenario driving unit 113, data reading
Embedding unit 114, drawing processing unit 115, speech recognition management unit 11
6. User input receiving unit 117, scenario embedding unit 12
1, data recording unit 118, external device management unit 119,
A gentle psychology unit 120 is provided. Each of these parts 111
To 120, CPUs provided in the agent processing unit 11;
Support engine (software) stored in ROM etc.
It is becoming more realizable. Situation judgment unit 111, etc.
The engines that make up each part of the
When transmitting, a corresponding event is issued. Each part 1
Some of the events issued by 11 to 120 are after branching
The event which specifies the scene of is included. The status determination unit 111 includes a time, a place,
Manage changes in the situation (including the addition of scenarios)
Issue events as needed. Judgment of situation change
Various situation detection devices 4 supplied from the situation information processing unit 15
0 based on the detection result, etc.
Is issued to the scenario execution request determination unit 112.
You. The scenario execution request determination unit 112
1 automatically receives the event
Determine whether to execute the scenario based on the motion judgment data
You. When executing a scenario, the event
Is issued to the scenario driving unit 113. The scenario driving section 113 has a
Of the scenario activation request issued by the request determination unit 112
It is determined whether or not to start up. Start up
When the scenario driver 113 determines that
The data reading unit 114 is requested to read the data.
In addition, drawing requests and capacities are drawn according to scenario data.
Request the drawing processing unit 115 to request the operation with the tractor
I do. It is also used for speech recognition according to scenario data.
To the speech recognition management unit 116. Scenario de
Request for control of external equipment (modem, etc.) and navigation according to data
do. Further, the scenario driving unit 113 outputs the result of voice recognition.
The scenario is driven by the result or user input (scene
Management). The scenario driver 113 receives an externally received
The scenario data file in the scenario embedding unit 121
Scenario stored in 302 (hereinafter referred to as external scenario)
The expiration date, and provide a scenario for expired scenarios.
Delete from the data file 302. Check expiration date
Is the reception date and time stored by the scenario embedding unit 121, or
Is determined from the expiration date and time. In addition, scenario driving unit 1
13 starts counting the number of times the external scenario is activated.
(Starting frequency counting means). And in external scenarios
When including the number of startups, each external scenario
The number of startups counted for the
If the external scenario is not
From the scenario data file 302.
ing. The number-of-starts counting means stores the data of the learning item data.
It can also be calculated using data. The data reading unit 114 is driven by a scenario.
Based on the request from the unit 113, the scenario data file
Load the corresponding scenario data from
U. The voice recognition management unit 116 transmits a finger to the voice recognition unit 142.
Control the indications. For example, notification from the scenario driving unit 113
Specify the dictionary to be used. Voice recognition management unit 116
Is determined by the speech recognition unit 142 according to the specified dictionary.
If the result of voice recognition is
Is issued to the scenario driving unit 113. With user
The force receiving unit 117 receives an input from the user (remote control / touch).
Event) corresponding to the input contents
Is issued to the scenario driving unit 113. From the user
Is determined based on information from the input control unit 16.
You. The scenario embedding unit 121 includes a communication control unit.
Whether the scenario received is included in the data received at 24
And if the scenario is included, the received
Scenario data file 3 with Nario as external scenario
02 is added to the scenario. Also new
If a new external scenario is added,
Issue a scenario addition event to the disconnection unit 112
You. The external scenario is a scenario data file 30.
2 may be stored in parallel. Scenario embedding unit 121
When storing a new external scenario,
Data to know the expiration date of Rio
Alternatively, the expiration date is stored in the RAM. The data recording unit 118 stores learning items and users.
The information to the learning item data 304 and the driver information 307.
Record (store) and manage. Also record on request
Tell me what items are memorized, such as the learning items you have. External device
The management unit 119 includes a navigation processing unit 10, a modem, and the like.
Is controlled by an instruction from the scenario driving unit 113.
You. The control result is returned to the scenario driving unit 113. D
The agent psychology unit 120 makes the judgment by the situation judgment unit 111.
Change the mental model parameters according to the
You. Also, mentally responded to the request from the scenario driver.
Change model parameters. Mental model
Judge the psychological state of the agent from the data. In FIG. 2, the external I / F unit 12 stores
The medium drive unit 23 and the communication control unit 24 are connected, and the
The display unit 27 and the imaging device 28 are connected to the
The audio control unit 14 includes an audio output device 25 and a microphone.
(Voice acquisition means) 26 is connected to the situation information processing unit 15
Is connected to various state detection devices 40, and the input control unit 16
Is connected to the input device 22. The various situation detecting devices 40 include a current position detecting device.
It has a device 41 and a situation detecting unit 42. Current position detection device
The location 41 detects the absolute position (depending on latitude and longitude) of the vehicle
The location of the vehicle using artificial satellites.
GPS (Global Positionion)
ng System) receiver 411 and azimuth sensor 4
12, a steering angle sensor 413, a distance sensor 414, a road
Bee receiving location information from beacon placed above
A receiver 415 or the like is used. GPS receiver 4
11 and beacon receiver 415 can measure position independently
However, the GPS receiver 411 and the beacon receiver 4
In a place where reception by the sensor 15 is impossible, the direction sensor 41
2. At least the steering angle sensor 413 and the distance sensor 414
To detect the current position by dead reckoning using one
It has become. The direction sensor 412 detects, for example, geomagnetism.
Geomagnetic sensor that detects the direction of the vehicle by detecting it, rotation of the vehicle
Detect angular velocity and integrate the angular velocity to determine the direction of the vehicle
Gyro such as gas rate gyro and optical fiber gyro
B. Arrange the left and right wheel sensors and output pulse difference (moving
Azimuth change by detecting the turning of the vehicle based on the
Wheel sensors, etc., which are used to calculate the
You. The steering angle sensor 413 is mounted on the rotating part of the steering.
Use an attached optical rotation sensor or rotation resistance volume
To detect the steering angle α. Distance sensor 41
4, for example, detects and counts the number of rotations of the wheels, or
Various methods such as detecting acceleration and integrating twice are used.
Used. Distance sensor 414 and steering angle sensor 413 drive
It also functions as an operation status detecting means. The situation detecting section 42 detects the situation of the driving operation.
Brake sensor that functions as a driving
421, vehicle speed sensor 422, direction indicator detector 42
3, shift lever sensor 424, side brake (park
(Braking brake) sensor 425. Also the situation
The detection unit 42 detects a device operation status, and detects a device operation status.
Air conditioner detector 427 functioning as detection means, wiper
A detector 428 and an audio detector 429 are provided. The brake sensor 421 is a foot brake
Detects whether the user is stepping on or not. Vehicle speed sensor 422
Detects the vehicle speed. The turn signal detector 423 operates
Whether the person is operating the turn signal, and the turn signal
Detects whether is blinking. Shift lever sensor 424
Indicates whether the driver is operating the shift lever, and
Detect the position of the bus. Side brake (parking brake
G) The sensor 425 indicates that the driver is operating the side brake.
Whether or not and the state of the handbrake (on or off)
To detect. The air conditioner detector 427 indicates that the driver
It is detected whether or not various switches are being operated. Wiper inspection
The output unit 428 detects whether the driver is operating the wiper.
You. The audio detector 429 allows the driver to use a radio, a CD,
Operate audio equipment such as players and cassette players.
Whether or not audio is being output and whether audio from the audio device is being output
Detect whether or not. The status detection unit 42 performs other operations on the device.
As a condition detection means, such as headlamps and room lamps
Light detection sensor that detects the operation status of lamps, driving
Belt detection that detects the user's seat belt attaching / detaching operation
It has sensors and other sensors. The input device 22 allows the driver to input driver information.
Or other agents in this embodiment
1 for the driver to respond to all inquiries, etc.
It is also a means. The input device 22 is used for navigation processing.
Location (starting point) and destination at the start of driving
(Arrival point), request information such as traffic congestion information from the information provider
In the specified driving environment (transmission conditions) of the vehicle you want to transmit,
Enter the type (model) of the mobile phone used
It is also for me. The input device 22 includes a touch panel
(Acts as a switch), keyboard, mouse, light
Remote control by pen, joystick, infrared ray, sound
Various devices such as a voice recognition device can be used. Also,
Remote control using infrared rays, etc.
May be provided for receiving various signals. Remote controller
To move the cursor displayed on the screen
In addition to the joystick, menu selection keys (buttons),
Various keys such as a numeric keypad are arranged. Input control unit 16
Detects data corresponding to the contents input by the input device 22.
Delivery agent processing unit 11 and navigation processing unit 1
Supply 0. The input control unit 16 controls the input operation by the driver.
By detecting whether or not it is working, it can be used as device operation status detection means.
Function. The recording medium driving device 23 is used for navigation.
The processing unit 10 and the agent processing unit 11 perform various processes
Computer program for reading from an external storage medium.
The drive used to insert. Record on recording medium
Computer programs include various programs.
RAM and data are included. Here, the recording medium is
A recording medium on which a computer program is recorded.
Physically, floppy disks and hard disks
Magnetic recording media such as disks and magnetic tapes, memory chips,
Semiconductor recording media such as IC cards, CD-ROMs and MOs,
Optically read information such as PD (phase change rewritable optical disk)
Recording media, paper cards, paper tape, etc.
The computer program is recorded on a recording medium or other various methods.
Recording media to be recorded are included. The recording medium driving device 23 stores these various
In addition to reading computer programs from recording media,
When the recording medium is written on a floppy disk or IC card, etc.
If the recording medium is writable,
RAM and storage device of the processing unit 10 and the agent processing unit 11
It is possible to write the data of the device 29 to the recording medium
It is. For example, an IC card has an agent function
Learning content (learning item data, response data) and driver information
Even when driving other vehicles by storing data such as information
Read the data from the IC card storing these
Use to learn according to your past response situations.
Communicate with trained agents
And become possible. This allows agents for each vehicle
Instead, the agent with the learning content unique to each driver
It becomes possible to appear in both. Also for IC card
A scenario data file may be stored.
As a result, each user has its own original
It is possible to be Rio. The communication control unit 24 receives signals from various wireless communication devices.
Mobile phones are to be connected. Communication control
The unit 24 includes a telephone line, a traffic congestion situation,
Information that provides data on traffic information such as traffic regulations
Used for communication with information providers and for karaoke on-board communications.
Communication with an information provider that provides karaoke data
Can be done. In addition, the communication control unit
24, learning data on the agent function,
It is also possible to send and receive scenario data files, etc.
You. In the agent processing unit 11 of the present embodiment, the communication system
The e-mail with the scenario attached is received via the control unit 24.
You can trust. Also,
The client processing unit 11 reads a homepage on the Internet.
It has browser software to display, and the communication control unit 24
Download data including scenarios from the homepage via
It can be downloaded. In addition,
The communication control unit 24 has a built-in wireless communication function such as a mobile phone.
You may make it so. The audio output device 25 is provided with a
Number of speakers and controlled by the voice control unit 14.
Voice, for example, guidance sound for voice route guidance
Voice and normal communication with the driver by the agent
For conversation and for obtaining driver information of this embodiment
The voice and sound of the question are output.
This audio output device 25 also serves as an audio speaker.
May be used. The microphone 26 is connected to the audio control unit 1
4, the voice to be recognized, for example, navigator
Input processing such as destination in
The driver's conversation with the driver (including the driver's response), etc.
It functions as voice input / output means for input / output. This microphone 2
6 has directivity to collect the driver's voice accurately.
Dedicated microphone is used. Note that the audio output device 25 and
By forming a hands-free unit with the microphone 26,
You may be able to talk. Microphone 26 and voice recognition
The unit 142 detects whether the driver is talking with the fellow passenger.
It functions as a conversation detecting means, and in this case, the microphone 26 and the sound
The voice recognition unit 142 detects a situation of the driver.
Functions as a step. The display device 27 includes a navigation processing unit.
Road map and various image information for route guidance by processing 10
Is displayed, or the agent processing unit 11
Various actions (video) of the agent and parts of the screen configuration are displayed
And so on. Also, in the imaging device 28,
The captured images inside and outside the vehicle are also processed by the image processing unit 13.
After it is displayed. The display device 27
Various display devices such as a liquid crystal display device and a CRT are used.
The display device 27 is, for example, a touch panel or the like.
It has the function of the input device 22 as well.
be able to. The image pickup device 28 has a C for picking up an image.
It consists of a camera with a CD (Charge Coupled Device)
In addition to the in-vehicle camera that images the driver,
Cameras on the left, right and left sides are located
ing. Image captured by each camera of the imaging device 28
Is supplied to the image processing unit 13 where processing such as image recognition is performed.
And each recognition result (passenger presence, driver recognition, etc.)
To be reflected in agent communication
Has become. The storage device 29 is connected to the storage device controller 17.
Data and data under the control of the storage controller 17.
Program reading and writing are now performed
ing. The storage device 29 has various types of age according to the present embodiment.
Necessary for realizing event and navigation functions
Agent (including programs)
Data 30 and navigation data 31 are stored
I have. The storage device 29 stores, for example, a floppy disk
Disk, hard disk, CD-ROM, optical disk, magnetic
Various recording media such as air tape, IC card, optical card,
The drive is used. In this case, for example, the learning term
Eye data 304, response data 305, driver information 307
With an easy-to-carry IC card or floppy disk
And configure other data on the hard disk
In other words, a plurality of different recording media and driving devices are used.
To use those storage media as drive devices
May be. The agent data 30 includes an agent
Program 301, scenario data file 302,
Voice data 303, learning item data 304, voice data
Response data 305 composed of
Image data 306 for displaying an image of the behavior, driver information
Information 307, each necessary for processing by other agents
The seed data is stored. The agent program 301 includes an agent
Agent processing program for realizing the agent function
Ram is stored. For example, the start condition of each scenario
Condition judgment processing for judging whether or not
If it is determined that the start condition is satisfied,
Starts the scenario corresponding to the
Scenario execution process for causing the agent to act
The data received by the communication control unit 24 is scenario data.
Data determination processing for determining whether or not
The received data contains scenario data
Scenario data file 3
02 and other various processing to be stored in
A processing program to be executed is stored. Learning item data 304 and response data 30
5 indicates that the agent responds to the driver's driving operation and response
This is data that stores the result of learning. Therefore, the learning term
The eye data 304 and the response data 305 are provided for each driver.
Data is stored and updated (learned)
You. The learning item data 304 includes, for example, an ignition
The number of times the fuel was turned on, the number of times the fuel was turned on each day, and the fuel
Agent's learning target such as data of fuel remaining amount at the time of oil
Is stored. The learning item data 304
Depending on the content of the study, for example, ignition
The greeting content when the agent appears changes according to the number of calls
Or the fuel level is below the average of the last five fuel levels
If so, the agent makes a refueling proposal. Response day
The user response to the agent's action is
An answer history is stored for each predetermined scenario. Response data
In 305, the response date and time and the response content are provided for each response item.
Stored for a fixed number of times. The response was ignored,
Rejected, accepted (accepted) cases, speech recognition
And from the input result of the input device 22 and stored. The scenario data file 302 contains
Agent that defines the behavior of the agent in
Rio data is stored. Figure 4 shows the scenario data
Conceptually the data structure and data contents of the data file 302
It is a representation. The scenario data file 302 is
Multiple scenario data that defines each scene are collected
One file is composed. Scenario data file
File 302 includes a file header and a scenario management table.
File, automatic start condition table, manual start possible scenario
View, each scenario data 1 to n, and additional resource data
It is composed of The file header is added to the scenario file.
Information about the file. Sina
Rio management table is included in the scenario file.
Data for managing scenario data, such as
Data such as priority when executing scenario data
Is stored. The automatic start condition table contains a scenario file.
To automatically start the scenario data recorded in the
Case data is defined and stored for each scenario data.
You. As an automatic start condition, for example, predetermined from a specific point A
Specific scenarios to run when you reach a distance point
Specific point and distance data, and agent at startup
As a condition for running the scenario in which he greets
Various self-services such as Nishon On
The dynamic start condition is stored. Also prohibit automatic startup
As an automatic start prohibition condition, for example, (A) manual start
Priority, (B) Do not automatically start while executing another scenario
(C) For example, China that has been executed within 30 minutes
Rio doesn't start, (D) starts when listening to music
No, (E) turning right or left on the curve (steering operation
(F) During rapid deceleration or rapid acceleration
Does not start, (G) starts when talking with a passenger
No, (H) Navigation device, audio, air
Control device, shift lever, side brake, wiper,
It does not start during operation of various devices such as turn indicators.
Does not start while audio is in use (listening to music or broadcast
Conditions for not disturbing: Audio must not be ON
And) (J) the traveling route by the navigation processing unit 10
Auto start prohibition conditions such as not starting during the guidance guidance of
Is also stored. However, these automatic start prohibition
Emergency conditions (fire departments, etc.)
Emergency start-up scenario, etc.)
The priorities are defined in the scenario management table. this
The automatic start prohibition condition is applied commonly to each scenario
There are conditions and conditions that apply to individual scenarios. Agent for collecting driver information
As a condition to launch a question scenario when a question is asked
Can determine that the driver's load state is a low load state
The following conditions and other conditions are specified as conditions for
Have been. (A) A scenario that asks a question within one hour has not been executed.
(Conditions to avoid frequent questions) (b) 15 minutes or more have passed since ignition was turned on
(Not immediately after starting to drive) (c) No conversation in the car (due to sound collection from microphone 26)
(D) Straight road (navigation data 31 and current position)
), There is no speed change for a predetermined time. (E) Waiting for traffic light (navigation data 3
1, Judgment based on current position and vehicle speed) These question activation conditions are essential conditions for activation
It is decided whether it is necessary to satisfy one or the other.
I have. For example, if the conditions (a) to (e) are limited,
(A) to (c) are all satisfied, and (d) or (e) is satisfied
When the question scenario start condition is met,
Rio is activated. Even in this case,
In case (A) ~, the question scenario starts
It is forbidden. In addition, automatic activation conditions including question activation conditions
The startup determination routine based on
Processing or an event specified in advance
It is supposed to be handled in case. Of the above conditions, condition (E),
Drive operation according to (F), (H), (J), (d) and (e)
The load on the crop is determined, and the conditions (C), (D),
(G), (I), (J), (a), (b), (c),
The degree of annoyance felt by the driver is determined more. The list of scenarios that can be manually started is described in the scenario file.
Manual start in the scenario data recorded in the file
List data of scenarios that can be activated is stored
I have. Additional resource data is used to expand storage
Audio data and image data to be added to the scenario data
And the like are stored. Each of the scenario data 1 to n is a scenario header.
Data, voice recognition dictionary DB, image data management table,
Table, each scene data 1 to m, and actual image data.
Has been established. Each scenario consists of multiple scenes
You. The scenario header is the header section of the scenario data.
Information about the scenario. voice
The recognition dictionary DB stores the D of the speech recognition dictionary used in the scenario.
At B, speech recognition dictionary data is stored. Image Day
Data management table specifies the images used in the scenario
Data is stored. Agent image
The image stored in the image data 306 and specific to the scenario is
It is stored in the actual image data at the end of the scenario data.
The scene data management table contains
Data for managing the scene data
I have. Each of the scene data 1 to m includes a scene header,
Screen configuration data, character motion instruction data, various processes
It consists of instruction data and a scene branch table.
The scene header is the header section of the sea data.
Information about scenes and each data section belonging to scene data.
Stores data for managing the application. Screen configuration
Is a section that defines the screen configuration displayed in the scene.
Each part of the screen configuration displayed on the display device 27
Data is stored. Character movement instruction data
Talks with the movement instructions of the character in the scene.
Instruction data on contents (voice data 303 to be used
Instruction data) is stored. Various processing instruction data
Data is processed by another module (for example, navigation) in the scene.
Data for requesting an instruction is stored. Scene branch
Table contains information on branching from the scene to the next scene.
In the section that is included, events
When the data that defines the next scene when the
Has been delivered. FIG. 5 shows a display device based on scene data.
27 shows an example of the scene screen displayed on the display 27.
You. The scene screen shown in FIG.
To get information about hobbies and preferences (meals)
Screen of a question scenario (scene number 0
x0001). The scene screen is shown in FIG.
Image of the agent (still image, video)
Agent display screen 51,
A balloon screen 52 in which the corresponding characters are displayed, a title screen
Plane 53 and image data unique to each scene (actual image data
Scene display that displays data images and answer selection buttons
It is composed of a screen 54. [0048] The agent processing unit 11 is provided with
When the question scenario is started, first the scene header
The screen configuration data of the scene specified by
Display from the file 302 and display the scene screen on the display device 2
7 is displayed, and a question voice equivalent to the question text is output.
The output from the force device 25 is provided. FIG. 5 (a)
In the question scenario scene screen, the balloon screen 52
"What genre of food do you like?" Is displayed.
Note that a sound corresponding to the display of the balloon screen 52 is output as a sound.
The force is output from the force device 25. Also figure
The scene display screen 54 in the scene screen of FIG.
Are "Japanese food" and "Yo" of the four answer selection buttons 54a.
"Food", "Chinese", and "none in particular" are displayed. The scene for asking a question to the driver includes
According to the driver's answer according to the
Scene is branched and continues. That is,
In the scene screen (scene number 0x0001) in FIG.
When the driver selects the answer selection button "Japanese food", the age
The event processing unit 11 displays a scene screen (b) corresponding to the answer.
Branched and displayed. In this scene screen (b), select
When the displayed “Japanese food” is displayed on the title screen 53,
And the balloon screen shows "I like Japanese food."
It is shown on the scene screen of the Japanese food after branching.
The actual image is read from the scenario data and the scene display
Displayed on surface 54. And to the agent processing unit 11
The driver's answer, for example, "Japanese food"
And stored in the hobby / preference data of the driver information 307.
It has become. In this way, the scenario
Scene image and sound are successively ordered until the last scene.
By displaying and outputting the next, the age in one scenario
The act of the agent will be completed. The audio data 303 in the storage device 29 (FIG. 2)
In accordance with the scene of the selected scenario,
Voice data for the driver to talk with the driver.
Have been. The voice data of the conversation by the agent is
Agent collects driver information according to embodiments
Voice data for asking a question is also stored.
Each data of the audio data 303 is a character of the scene data.
Is specified by the actuator operation instruction data. The image data 306 includes a finger according to a scenario.
Displays the state of the agent used in each specified scene.
Still images and moving images (animations)
Is stored. For example, an agent bows
Videos, nodding videos, videos raising the right hand, etc. are stored.
I have. Each of these still images and videos has an image code.
Have been. Agent stored in image data 306
The appearance of the person is human (male, female)
No need. For example, as a non-human agent,
Animals themselves such as octopus, chicks, dogs, cats, frogs and rats
Of animals and animals that are humanized in design (illustration)
It may be a figure, a robotic appearance, floors
In the appearance of a stand or tree, or the appearance of a specific character
There may be. Also, the age of the agent is constant
There is no need to be an agent learning function
Is the appearance of a child and grows over time
Changes (changes to the appearance of an adult, and the appearance of an elderly person)
May be changed). Image data 306
Contains images of the appearances of these various agents.
From the input device 22 etc. according to the driver's preference.
You can choose. The driver information 307 is information about the driver.
The driver's wishes and tastes are more
Used to match taste and taste. Figure 6
This is a conceptual representation of the configuration of the driver information 307.
As illustrated in FIG. 6, the driver information 307 includes:
Driver ID (identification information), name, age, gender, marriage
(Married or unmarried), luck consisting of children, number and age
Changer basic data and hobby / preference data are stored
It has become. Hobby and taste data include sports, drinking
Large items such as food and travel, and included in the concept of these large items
It consists of detailed items. For example, large item sports
There are favorite soccer teams, favorite baseball teams,
Detailed data such as interest in Ruff is stored
Has become. In this embodiment, each of these driver information
Are prioritized and agents are still ranked
In the driver information that is not stored,
They ask questions to the translators. Driver basic data
Has a higher priority than hobby preference data. The driver information 307 indicates that the vehicle is driven.
If there is more than one driver, it is created for each driver.
You. Then, the driver is identified and the corresponding driver information is used.
Used. The identification of the driver, for example,
At the same time, an agent common to all drivers appears and
Inquire about the transferee and identify from the answer. driver
Inquiries, select the name of the entered driver and other
Button on the display device and output a voice prompting the user to make a selection.
I do. New user registration screen when other is selected
Is displayed. Weight, fixed position of driver's seat (seat)
(Front and back position, backrest angle), rearview mirror angle,
Eye-gaze height, digitized face photo data, voice characteristics
Less driver-specific information such as ID parameter cards
One is stored in the driver information 307, and
The driver may be specified from the information. The navigation data 31 includes a route guidance
Various data files used for
Data file, drawing map data file, intersection data
File, node data file, road data file
Files, search data files, photo data files, etc.
Data is stored. The communication area data file includes a communication control unit.
24 or connected in the car with no connection
The area where the mobile phone can communicate from inside the car is displayed on the display device 27
Or use the area where communication is possible when searching for routes.
Communication area data for each mobile phone type
Have been. The drawing map data file includes the display device 2
7 is stored. this
Drawing map data is a hierarchical map, for example, the top layer
To Japan, Kanto region, Tokyo, Kanda
Diagram data is stored. Map data for each level is
Each has a map code. The intersection data file specifies each intersection.
Intersection number, intersection name, intersection coordinates (latitude and longitude
Degrees), the number of the road where the intersection is the start or end point
Signal, signal presence, etc. are stored as intersection data.
Have been. The node data file contains
Consists of information such as latitude and longitude that specify the coordinates of each point
Node data is stored. That is, this node
The data is data on one point on the road,
A node that connects nodes is called an arc.
By connecting arcs between each
Is expressed. The road data file specifies each road.
Road number, intersection number to be the start point and end point,
No. of road with end point, road thickness, prohibition of entry, etc.
Information, the photo number of the photo data described below, etc.
You. Intersection data file, node data file, road
Intersection data respectively stored in road data files,
Road network data consisting of node data and road data
Used for road search. In the search data file, a route search
Intersection sequence data and node sequence data that constitute the generated route
Data is stored. Intersection column data is the intersection
Photograph of name, intersection number, and characteristic scenery of the intersection
It consists of information such as numbers, turning angles and distances. Also the node
Column data contains the location of the node, such as east longitude and north latitude.
Consists of information. Photo data files include each intersection and
A photograph of a characteristic scenery, etc. that is visible while moving
Digital, analog, or negative corresponding to the true number
It is stored in film format. Next, the agent device configured as described above will be described.
The operation of the device will be described. FIG. 7 shows an agent device.
Flow chart showing the operation of scenario addition processing by
It is. The agent processing unit 11 controls the communication control unit 24
Monitoring whether data has been received via the
11). When the data is received (Step 11; Y),
Scenario incorporation unit 121 includes data in scenario
(Step 12: Data Judgment)
means). That is, the data received from the homepage
If the download data is
It is determined whether or not a scenario file is included. on the other hand,
If the received data is an e-mail,
Judge whether there is an attached file and add a scenario if there is an attached file.
It is determined whether it is attached. Note that the scenario in this embodiment is
The judgment as to whether or not it is based on the extension of the data
But in other ways, e.g. the file header indicates the scenario
For those who store data and make judgments based on this data
It may be a law. When scenario is included in received data
(Step 12; Y), the scenario embedding unit 121
The received scenario is stored in the agent
When stored in the scenario data file 302 of the data 30
In order to delete stored external scenarios,
Save. Then, the scenario embedding unit 121
In order to be able to launch external scenarios that have been added,
A scenario addition request is sent to the scenario execution request
An event is issued (step 14), and the process ends.
The display unit indicates that a new external scenario has been added.
Display on the display 27 and notify the user by audio output
You may do so. FIG. 8 shows the operation of the scenario activation judgment processing.
It is the flowchart which was performed. Agent processing unit 11
The scenario execution request determination unit 112 acquires status information
(Step 21). That is, the scenario execution request determination unit
Reference numeral 112 denotes various data supplied from the situation information processing unit 15.
The current status issued by the status determination unit 111 based on the data
Get events that show status or changes in status. Furthermore, China
In the automatic start condition table of the Rio data file 302.
Whether the scenario activation conditions are met.
In other words, a scenario (quality
(Including question scenarios)
22). Scenario start condition determined here
Is the external information incorporated by the scenario incorporating unit 121.
The starting conditions specified in the scenario are also included. Then, the scenario execution request determination unit 112
When the scenario start condition is satisfied (Step 23;
Y), the scenario execution request determination unit 112 satisfies
Event start request of scenario corresponding to scenario start condition
Is issued to the scenario driver 113 (step
24), and proceed to step 25. On the other hand, any startup
If the conditions are not satisfied (step 23; N),
Step without issuing an event
The process proceeds to step 25. In step 25, all scenarios
Make sure that the start conditions for
If there is no activation condition (; N), return to step 21
If all the start conditions are determined (Y), the process is terminated.
You. Next, in the above agent device,
The user makes a reservation for the inn, and the inn
E-mail with scenario attached to vehicle (agent device)
The following describes an example in which a rule is transmitted. In addition, inns and eating and drinking
There are no particular restrictions on how to book a store, etc.
You can make a reservation from the store's website, facsimile or
Reservations may be made by talking. Also, reservations by e-mail,
Reservations by other methods such as letters are also possible. However,
Attach scenario from the shop (inn) to e-mail
In order to receive and receive this,
The e-mail address of the agent device is also used for the reservation method
You need to contact. Please note that each inn and store
When downloading an external scenario from a web page, etc.
Or, when booking, not on the day of booking
Even when receiving Rio, the external scenario received
The scenario is executed in the same manner according to the start condition. sand
In other words, even if external scenarios are incorporated in advance,
If the startup date is set as a reservation date,
The same as when the scenario was received on the reservation date.
to start. FIG. 9 shows the information that the inn has sent to prospective guests.
The transition of the scene screen according to the scenario is shown for each scene.
Things. This guidance scenario includes multiple scene screens.
The scene screens (a) to (f) are displayed
Depending on the user's selection result for the scene screen (c).
Next scene screen branches to 0x0004 and 0x0006
are doing. Further, in the example of FIG.
On the screen (d), depending on the type of dish selected
The scene screen is displayed so that the
You may make it branch. In the following, according to FIG.
Explain each of Gent's actions. The following scene images
Agent behavior and screen display explained corresponding to the surface
Are stored in the scenario data of the external scenario.
Displayed according to the data, images and instructions given
It is. Also, as an operation of the agent,
Actually, the agent processing unit 11 performs the processing. When the reservation scenario is started, first, the number 0
The scene screen of x0001 is displayed on the display device 27.
In this scene, the agent display screen 51 displays the agent
, Bows and greets by voice. sound
The contents of the greeting by voice are displayed on the balloon screen 52.
Is the same as the sentence. Greeting by voice is on behalf of an inn
Agent will do it for you
Travel by displaying a photographic image of the general on the scene display screen 54
It is expressed that it is a greeting from the museum. Of this landlady
Images are images received and added as part of an external scenario
Is stored as actual image data of the scenario data.
(See FIG. 3). Instructions for agent behavior
According to the instruction stored in the character motion instruction data
U. When the greeting by the agent ends, the next scene
Transition to 0x002. In the next scene 0x0002, the scene display
An image of the open-air bath is displayed on the screen 54. And this
The agent points to the picture of the open-air bath, and the specialty of the inn
(This is a sale)
A description will be given of the display of the balloon screen 52. Agent story
Is completed, transition to the next scene 0x0003 is made
Display image 54 of the image of the meal (image of kaiseki cuisine)
And the agent explains the dish and
Ask whether or not. Note that the scene display screen 54 shows
According to the rules of the scenario, in addition to the image of kaiseki cuisine,
A button 54a for user response is displayed. To this question
If the user selects "Yes" as the answer to
Branch to scene 0x0004 and select “No”
In this case, the process branches to scene 0x0006. In scene 0x0004, other than kaiseki cuisine
A selectable list is displayed on the scene display screen 54. D
Agent points to the list on the scene display screen 54.
And ask which dish is better. And the user
If any one is selected, transition to scene 0x0005
You. In scene 0x0005, a change from kaiseki cuisine
A list of the number of people to be changed is displayed on the scene display screen 54,
Agents point to this list and ask
I do. When the user selects one, the scene
Transition to 0x0006. In scene 0x0006, a scene display screen
The inn's exterior photo image is displayed at 54, and the agent
Bow and say hello. And the agent
In the case of the guidance scenario shown in FIG.
Executing the answer result regarding the meal via the communication control unit 24
To the third party (inn) who sent the external scenario inside
You. Thus, if you want to get information about the user
In this case, the creator of the external scenario obtains the information
Questions in the scenario, and answer
Create a scenario to send by child mail. In addition,
If you need to submit a response, send the author's email address
In the scenario data. The last scene (Fig. 9
Now the story of the agent in scene 0x0006) is over
Then, the scenario ends. As described above, the agent processing unit 11
Is the last scene image and audio specified in the scenario.
Display and output sequentially up to the scene. The started scenario is over
Upon completion, the agent processing unit 11
Determine the start condition for execution. According to the embodiment described above,
The action of the event is specified in the scenario data,
Scenario consisting of multiple consecutive scenes with nario data
Has been standardized by the
Users and third parties can create and incorporate their own scenarios.
Can be. In this way, the differential stored in the
Users can add them to the default scenario.
Search for additional scenarios on the Internet etc.
And create your own scenarios.
The device can be used more comfortably. Also travel
Service providers such as museums, hotels, department stores, etc.
Promote the service and appeal to users
Create and provide scenarios that provide strategic information
Thus, customers can be secured in a friendly manner. For example, sir
A service provider can provide agents with their own services and products.
To set a store as a navigation destination
Create a scenario as follows. Select store as destination
By doing so, you can surely attract customers to the store
Wear. In addition, the service provider
Ask questions and collect the answers via communication
So you can get effective marketing information
You. Next, the user or a third party creates a unique scenario.
Configuration and operation of scenario creation device 2 to be created
explain. FIG. 10 shows the configuration of a scenario creation device.
Things. The scenario creation device 2 includes a control unit 210,
Force device 220, output device 230, communication control device 240,
Storage device 250, storage medium drive device 260, input / output IF
(Interface) 270. Each of these
The device is connected to a bus line 280 such as a data bus and a control bus.
Connected. The control unit 210 controls the entire scenario creation device 2
And to control. The control unit 210 includes the CPU 21
1. ROM (Read Only Memory) 21
2. RAM (Random Access Memory)
y) 213 and the like. The ROM 212 stores various calculations by the CPU 211.
Programs, data and parameters for controlling
This is a read-only memory in which data and the like are stored. R
The AM 213 is provided to the CPU 211 as a working memory.
This is the random access memory used. CPU 21
1 writes a program, data, etc. in the RAM 213
And can be erased. In this embodiment
In the RAM 213, the CPU 211 stores a scenario editor.
(Scenario editing program) to create a scenario,
An area for editing, storing, etc. can be secured.
You. The input device 220 is connected to the scenario creating device 2.
A device for inputting letters, numbers and other information
Yes, for example, composed of a keyboard and mouse
I have. The keyboard is used to enter kana and English characters
Key, numeric keypad for entering numbers, and various function keys.
Keys, cursor keys and other keys
I have. The keyboard is, for example, a user
Enter your login ID and password to log in to 2.
Or use speech synthesis or speech recognition when creating scenarios.
Used when inputting an elephant sentence. Mouse
It is a pointing device. GUI (Graphi
cal User Interface)
When operating the scenario creation device 2, it is displayed on the display device.
Click the clicked button or icon with the mouse.
Thus, predetermined information can be input. The output device 230 is, for example, a display device or a printing device.
Device. The display device is, for example, a CRT (Cath
mode Ray Tube) display, liquid crystal display
Play, plasma display, etc. are used. display
The device has a main screen for creating scenarios,
Various screens for selecting the screen configuration in the scene
The screen is displayed. The information selected on each screen
Information and entered information are displayed. mark
The printing device checks the entire flow of the generated scenario, for example.
And print the status of each scene
Device for printing. The printing device, for example,
Jet printer, laser printer, thermal transfer pudding
Printers such as printers and dot printers
You. The communication control device 240 communicates with the outside
A device for transmitting and receiving data and programs,
Modems, terminal adapters and other devices are used
You. The communication control device 240 is, for example, the Internet or L
AN (Local Area Network)
It is configured to be connectable. The communication control device 240
Other terminals or services connected to these networks
From the server device, etc.
Sending Rio Data and Scenarios Created by Third Parties
Receiving (downloading) data
Data required to create data
It has become. The communication control device 240 is a CPU 211
Is controlled by, for example, a predetermined protocol such as TCP / IP.
Communication with these terminal devices and server devices according to the protocol
Send and receive signals and data. The storage device 250 is a readable and writable storage medium.
Read programs and data from the body and its storage medium.
It is constituted by a driving device for writing. The
Hard disks are mainly used as storage media.
Besides, for example, magneto-optical disk, magnetic disk, semiconductor
Other readable and writable storage media, such as
It is also possible. The storage device 250 has
E editing program 251, scenario editing data 252,
And other program data 253 are stored.
You. As other programs, for example, a communication control device
240 and the network with the scenario creation device 2
Communication that maintains communication with connected terminal devices and server devices
Communication programs and scenarios such as memory management and I / O management
E is the basic software for operating the creation device 2.
OS (Operating System)
Stored in the storage device 250. FIG. 11 shows a scenario editing program 251.
And the contents and configuration of the scenario editing data 252
It is. As shown in FIG.
Gram 251 is composed of scenario editor 251a and DB
Collection tool (program) 251c and scenario compilation
251b. Scenario editing data 252
Are the common definition DB 252a and the local definition DB 252
b, scenario data 252c, and execution format data 25
2d. The scenario editor 251a
Application program to create Rio data
You. DB editing tool (program) 251c is common
Editing program for updating data stored in the
Program to directly change data, communication or
The data stored in the storage medium by the storage medium driving device 260
Data is read and updated, or the communication control unit 240
Via the Internet. Scenario con
The piler 251b is created by the scenario editor 251a.
Scenario that can be used on the agent device.
Software to convert to scenario data
Functions as a step. The common definition DB 252a stores each scenario.
(E) Data common to the creation devices is stored. Common definition D
B252a includes automatic start item data and scene
Data such as branch items and additional conditions are stored. local
Definition DB 252b indicates that the user is creating scenario data
The contents defined in are stored. For example, created by the user
Branch conditions, and the “good
Stores voice recognition data such as "Y", "Komaru", "No"
Is done. The scenario data 252c is a scenario editor
The data created in 251a is stored. Executable format
Data 252d is converted by the scenario compiler 251b.
The stored scenario data in the executable format is stored. FIG. 12 shows a scenario start condition (scenario).
Part of the contents of the start condition element that can be selected as the start condition)
Is an automatic start item table conceptually representing the above. Automatic start
The moving item table is stored in the common definition DB 252a.
I have. As shown in this auto-start item table,
Items that can be detected, recognized, and confirmed by
It is stored in the definition DB 252a. Scenario editor 2
Automatic start item starts when creating a scenario in 51a
Read from the common definition DB 252a as a condition element
(Start condition element acquisition means) is displayed in a list. This resource
One or more auto-start items (start
Scenario element) to start the scenario autonomously
Conditions (scenario start conditions) are created
(Start condition creation means). In the remarks column of the auto-start item, select
Items that can be referred to are listed in
Detailed conditions can be selected.
As shown in the list of 13 selectable items,
Are specified in the auto-start item number.
Have been. In this embodiment, the key of the automatic start item
Table is stored in the common definition DB 252a.
Via communication means such as Internet and communication control device 240
And obtain by downloading from external DB
You may make it. This acquisition method is common definition DB
252a is possible for all data stored
is there. FIG. 14 shows a branch from a scene to the next scene.
Branch that stores branch items (transition conditions) for
It is an item table. Scene branch item table is common
It is stored in the definition DB 252a. Scene branch item
Each item is read when creating the deployment configuration for each scene.
Is displayed in a list. Minutes from the displayed list
One or more branch conditions are selected as branch conditions, and
The branch condition not stored in the input device 220
, A scene development configuration is created. FIG. 15 sets the branch condition in more detail.
7 shows an additional condition table in the case. Eger
The client device uses the data recording unit 118 to copy scenarios.
Branches into several scenes, each with a different scene.
How to end when Nario ends as end ID
It is to be recorded. The data recording unit 118
Is the response of the user in the scenario, the total number of uses, etc.
Can be recorded as learning data.
ing. The additional condition table shown in FIG.
Recorded end ID, total number of uses, and other various
The item is defined as a more detailed branch item,
Can also be selected as branch conditions. The common definition DB 252a includes, in addition,
Use for speech recognition data and speech synthesis used for recognition
Both voice synthesis data are stored. Also see
Displayed on the agent display screen 51 of the application screen (see FIG. 5)
For selecting the display and behavior of the agent
Agent image data (the image of the agent
And stored on the scene display screen 54.
Screen configuration data for selecting a screen configuration is also stored.
Furthermore, items that can be selected as processing contents in the scene
Eyes, for example, actions that agents can handle,
Audio device on / off, channel selection, air conditioner
Device on / off and temperature setting, navigation processing unit 1
Set destination to supply to 0, turn on / off headlights
And other processing content item data are stored in the common definition DB 252a.
Is stored. In the present embodiment, these
Speech recognition data, speech synthesis data, agent image
Image data, screen configuration data, and process content item data
By reading from the communication definition DB 252a,
, But each of these data
At least one, communication means such as the Internet and communication
Download from external DB via control device 240
By doing so, it may be obtained. The storage medium drive 260 is a detachable storage
Drive for driving the storage medium 7 to read and write data
Device. As the removable storage medium 7, for example,
Magneto-optical disk, magnetic disk, magnetic tape, semiconductor memory
Moly, paper tape with punched data, CD-ROM, etc.
There is. In this embodiment, the scenario creation device
Scenario data collected is mainly written in semiconductor memory.
Is to be included. Scenario creation device 2
The storage medium drive device 260 drives the storage medium 7.
From the storage medium 7 storing the scenario data,
Acquire a scenario or create a scenario
Data from the storage medium drive to the storage medium 7
Can be. Input / output IF (interface) 270
Are, for example, serial interfaces and other standards
Interface. I / O IF
(Interface) 270
Scenario creation by connecting compatible external devices
The function of the device 2 can be extended. Such outside
For example, storage devices such as a hard disk,
There are speakers and microphones. Next, a scenario constructed as described above is created.
Screen transition for each operation of scenario creation by device
It will be described according to. FIG. 16 shows the scenario editor 251.
a main window displayed on the display device when
It shows the configuration of the dough. As shown in FIG.
As shown in the figure, the main window is
Image displayed on the display device 27 of the agent device 1
Screen (see FIG. 5)),
A setting screen 303 on which setting items for performing seed setting are displayed;
The scene development configuration (branch state) indicates the scene that represents each scene.
Scenes displayed by the hierarchical structure of the icon 307
An open screen 305 is configured. When the scenario editor is started, the main window
The start point is displayed on the scene development screen 305 of window.
308 is displayed. This start point 308
Select to edit the scenario properties. Choice
For example, the position of the point
Double-click the mouse at the
Click to select. Screen configuration change button 309
Click to select the screen configuration you want to display.
You. Double-click the agent display screen 311 with the mouse.
Click to edit the behavior of the agent (character)
The screen is displayed. Click the dialogue edit button 313
Then, the instruction of the character's dialogue can be edited. Button par
Double background / speech recognition dictionary setting 315
Click to edit the speech recognition dictionary to use.
Answer selection button 315a (54a) on scene screen 301
Word to recognize when you select the one displayed with the mark
Is displayed on the scene screen, and the dictionary list display section 31 is displayed.
When 5b is selected, the target for speech recognition is
Word names are not displayed. In the setting 317 of the maximum display time,
Set the maximum amount of time that the button can be displayed. External equipment, etc.
In the control instruction edit 319, external devices and the like (including navigation)
Set the control instruction of Next scene creation button 321
Click to edit the scenario flow (next sheet
Create a new application). With this scene creation button 321
By branching the flow of o, the development configuration of each scene
Is done. For example, when icon 1 of the scene is selected
Create the next scene in (active display state)
Click the button 321 to select the scene
The icon is displayed on the lower layer side and clicks multiple times
Scenes 2, 3, ... following scene 1 are created by branching
You. Click the end position creation button 323 for the scenario.
Click to create a scenario end position. make
At the end position of each scenario, the end number is the end ID.
It is made to be allocated. Data conversion
Click the button 325 to navigate the created scenario
To an executable format for use in
You can bart. FIG. 17 shows editing of scenario properties.
It shows the flow of screen operation. In (a) of FIG.
In the main window shown in FIG.
Double-click the start point 308 displayed in 5
Click to edit the scenario properties shown in (b)
The window is displayed on top of the main window
You. This scenario property edit window
, Scenario name input, Kana name input, icon
Selection, priority setting, expiration date
The upper limit of the time lag from the start to the actual start)
Settings, scenario start conditions (separate window),
Can input creator name and comment. On this screen
The scenario name input and kana name input that are input are executable
This becomes the scenario header in the scenario data of the expression. Shi
In the edit window of Nario properties,
Click on the button 402 to reflect the edited content to the data
To return to the main window. Meanwhile, Cancel button
If you click 403, it will not be reflected in the data,
Return to the window. Click the start condition setting button 401
Click to start the scenario start condition (start condition)
The edit screen window (Fig. 17 (c)) is displayed
It is. In the main edit window of the scenario start condition,
Open the folder displayed on the automatic start condition list screen.
By doing so, the automatic start items shown in FIG.
Listed by place. Auto-start items from this list
By selecting, the start condition (automatic start condition) is selected.
The selected item is displayed on the right. Of the scenario start condition
Click the Finish button in the main editing window to
Finish editing the Nario launch conditions and edit the scenario properties
Return to the collection window (b). FIG. 18 shows the main editing of the scenario start condition.
Screen operation to edit scenario start condition from window
It is a representation of the flow. Main scenario start condition
In the edit window (a), the user manually edits the scenario.
It can be set to start. Also,
The conditions under which the system automatically starts the scenario are displayed.
You. Click the new button 405 to start automatically
The judgment item selection window (b) is displayed and a new start
You can edit conditions. In addition, the automatic start condition list screen
Select the displayed automatic start condition and click the edit button.
Click the button to set additional conditions for the selected condition.
Automatic startup judgment item selection window for addition (b)
Is displayed. An automatic start judgment item selection window (FIG. 18)
In (b)), the judgment condition item to be set (category
Condition) to start automatically when you select (3) and click Enter.
Proceed to the range selection window (c) and click Cancel.
Click to edit the scenario start condition main edit window
Return to (a). In this automatic start judgment item selection window,
The items that can be displayed and selected are the automatic startup items shown in FIG.
Eyes. Selection window for condition range to be automatically activated
In FIG. 18 (c), the selection was made in the previous window (b).
Of the window according to the judgment condition items (categories)
Changes. Select the condition range that can be started automatically from the list.
Select (or enter in numbers) and click the Add button
Is set. Click the OK button to set the status
The main edit window of the scenario start condition
Return to (a) and click on Cancel to automatically launch
It returns to the cut item selection window (b). FIG. 19 is a diagram showing an automatic start in the case of a numerical value input.
This shows a configuration example of the condition range selection window.
You. In the automatic start judgment item selection window of FIG.
If the selected condition item is a numeric input item,
The numerical value input screen of FIG. 19 is displayed. This date specification field 4
11 Select the date and click the Add button to select
The selected date is displayed in the selection display column 412,
Click to select the automatic start judgment item in FIG.
Return to the window. FIG. 20 shows a table displayed on the agent display screen 51.
Shows the flow of screen operation to select the screen configuration you want to show
It is. The main window shown in FIG.
C The scene eye displayed on the scene development screen 305
Select and activate the computer 307
The scene screen 301 corresponding to the scene icon
It is. Then, a button for changing the screen configuration of the setting screen 303 is displayed.
Click 309 to open the screen configuration selection window
(B) is displayed. Selection window for this screen configuration
In (b), display is possible on the scene display screen 54 (see FIG. 5).
A list of available screen configurations is displayed. Group that displays nothing
This screen, two selection buttons with two selection buttons displayed
Screen, button selection screen with multiple selection buttons displayed, example
For example, a list of multiple items such as prefecture names
List selection screen, image display screen for displaying image data, etc.
Various selectable screens are displayed. Each image displayed in a list
Select one screen configuration from the screen configurations and click the OK button.
Click to confirm if you want to change the screen configuration
If you want to change after confirming with, change to the screen configuration and change
Return to the window (a). Return to main window
Changes the scene screen 301 to the newly selected screen configuration.
Displayed. FIG. 21 shows a character action (agent
Action) It shows the flow of screen operation to edit instructions
is there. Main window showing the editing status of the scene screen
In the dough (FIG. 21A), the agent display screen
Double-click 311 with the mouse to move the character
Direct editing window (b) or character movement
The status instruction edit window (c) is displayed. Motion finger
Is indicated by direct instructions of character movement
(B) is displayed and the finger is
If it indicates, (c) is displayed. Direct instruction of character movement in FIG. 21 (b)
In the edit window, motion (motion), facial expression (emotion)
Elements of expression), hairstyle (elements of growth expression), clothing (TPO table)
Current element), scale (character display area
If it is a frame, the camera angle element), mouth park
Range (the range to assign dialogue), operation instruction tie
Select the mining and background. In this window
Clicking the OK button will cause the edited content to
It is projected and returns to the main window (a). Cancelbo
Clicking the button will not be reflected in the data.
Return to window (a). Also, click the expression content designation button.
Tick, give instructions in the state you want the character to express
The state instruction edit window (c) is switched. The character movement state instruction shown in FIG.
In the edit window, the state represented by the character
Work elements, mental state elements, TPO expression elements,
Long representation element, scale element (character display area
If it is a frame of Mera, the element of the camera angle),
Select the operation instruction timing and background. This Wi
Window, click the OK button to save the edited data.
And the display returns to the main window (a). Can
Clicking the cell button will not reflect the data
Return to in-window (a). Click the direct designation button
If you click on the button,
The screen is switched to the instruction editing window (b). FIG. 22 shows the character (agent).
This shows the flow of screen operations for editing dialogue instructions.
You. Main window showing the editing status of the scene screen
C (FIG. 22 (a)), the dialogue of the setting screen 303
Click on the collection button 313 to edit the dialogue window
1 (b) is displayed. Edit window 1 for this line
In (b), a tab screen of “PCM” is displayed and PCM is displayed.
(Recorded and prepared voice data)
You. Click the "TSS" tab in the dialog edit window 1 (b)
When selected, the dialogue editing window 2 (c) is displayed
Thus, a TTS (synthesized voice) instruction can be given. Place of PCM
Enter the joint words in full-width hiragana (the above input box)
Search for applicable lines (click the search button)
: Two types of search methods are available: prefix match and exact match)
From the search results (there is a search result list display field).
Select a dialogue line to be spoken by the selected character (click the Add button).
Click to set). In case of TTS
Lyric name (speech balloon display data) and reading (synthesized speech)
To make the character speak
Set dialog (Click the Add button to set
). Dialogue editing window 1 (b), 2
In (c), click the OK button to edit
Is reflected as data on the balloon screen 430
Return to window (a) and click Cancel button
Then, it is not reflected in the data, and the main window (a)
Return. FIG. 23 is a screen operation for editing a speech recognition dictionary.
It shows the flow of the work. This operation is performed from the device side.
To recognize the answer to a question that requires an answer
Is the person who sets up the voice dictionary. Editing status of scene screen
In the main window (FIG. 23 (a))
The answer selection displayed according to the selected screen configuration.
Select button 315a (normal list box depending on the screen configuration)
Double-click on the sound
The voice recognition dictionary selection window (b) is displayed. Also,
Dictionary display unit 315b for recognition in the background
Double-click to select speech recognition dictionary selection window
(B) is displayed. A speech recognition dictionary selection window (FIG. 23)
In (b)), the dictionary name in the list display of dictionary candidates
Double-clicking will use the speech recognition dictionary
Then, it is displayed in the list selected as a general dictionary. Decision
Click the button to apply the edits to the data
Return to the main window (a) and click the Cancel button.
Clicking on the main window will not be reflected in the data
Return to (a). When you click the new button,
Speech recognition dictionary creation window to create a new recognition dictionary
(FIG. 23C) is displayed. A speech recognition dictionary creation window (FIG. 23)
In (c)), enter the name of the dictionary and click the add dictionary button.
Click to create a new voice recognition dictionary with that name
Window (d) for registering words in the voice recognition dictionary is displayed
Is done. OK button in the voice recognition dictionary creation window (c)
Click the button to finish creating the voice recognition dictionary and
The process returns to the recognition dictionary selection window (b). A window for registering words in the speech recognition dictionary
In FIG. 23 (d), the word to be registered is
Fill in the ligana field and click the OK button. Then first name
Select or enter a new name (the name you want to display)
Select the PCM audio for the callback (select None)
If a callback is made, the TTS is used).
After entering these three items, click the register button.
Then, the data is registered and added to the registered word list on the right.
When you have finished registering the words you want to register, click the back button.
Click to return to the voice recognition dictionary creation window (c)
You. FIG. 24 shows a screen for editing the flow of a scenario.
It shows the flow of operation. In the main window
Scene icon 307 (in the figure, the icon
Select 1) to make it active and create a new scene
When the button 321 is clicked, a branch event selection window is displayed.
The window (b) is displayed. Select this branch event
In the window (b), a new scene is created.
Select the condition to branch from the branch event list, and click the OK button.
Click the button to determine the condition (branch event)
To return to the main screen (a). Main screen after returning
In (c), a new scene is displayed on the scene development screen 305.
(In the figure, a scene icon (4) is created.
The scene icon created in
Is written as NEW in a red square to distinguish it from
You. It can be selected in the branch event selection window
The branch event is displayed in FIG. FIG. 25 shows a screen for editing the flow of a scenario.
This shows an extended version of the operation flow. Main win
In the window (a), click the scene creation button 321.
Click to display the branch event selection window (b).
It is. In the selection window (b) of this branch event
Even once selected branch event, more specific additional
If the condition is satisfied, the branch can be changed. For example
Select the word "I'm fine" in the voice recognition dictionary in scene 1
Branch condition is already defined when branching to scene 2
Have been. In addition, as an additional condition,
Branch to another scene if it ends in a command
You can also Dictionary of branch event list
Select the word "I'm fine" and select an additional condition type
Then, the items of the additional condition that can be selected in the selected type
Select the item whose list is displayed in the additional condition list, and click the OK button.
Click the button to select the condition (branch event + extension
), And returns to the main screen (a). At this time Sina
A new scene is created in the Rio diagram (Main screen
Fifth square marked NEW in a red square in (c)
Is a scene newly created by this operation). Can be selected
Additional condition type and additional condition items belonging to that type are as follows:
This is displayed in FIG. This allows, for example, voice recognition
When the word "I'm fine" is selected in the dictionary,
If the last scenario ended the scenario at the end position of ID 2
Branch to scene 5 otherwise branch to scene 2
Can be created. FIG. 26 shows the editing of the end position of the scenario.
It shows the flow of screen operation. Main window
In (a), a scenario end position creation button 323 is displayed.
Click on to display the end ID designation window (b).
Is displayed. In the window (b) for specifying this end ID
Here, the ID number of the end position mark is specified. Through
Normally, it is assigned automatically, but the
If you uncheck the check box, the editor operator
You can assign it yourself. Click OK button
Then the ID number is determined and the branch event selection window
(C) is displayed. In the branch event selection window (c),
Finish the scenario in the same way as when creating a new scene
Set the branch condition for The same applies to setting additional conditions.
Wear. Click OK button in this window
In that condition (branch event + additional condition if set)
Determined and returns to main window (d). At this time Sina
A new scenario end position 433 is created in the Rio diagram
I have. FIG. 27 shows the navigation of the created scenario.
To an executable format that can be used
It shows a flow of screen operation for performing the operation. Maine
In the window (a), the data conversion button 3
Click 25 to open the data converter window
(B) is displayed. The window of this data converter
C) In (b), a file for outputting the converted data
Specify the file name and select the scenario to be converted at the same time
(Simultaneously check the scenarios checked in the scenario list)
Convert) and click the compile button to save the data.
Start data conversion. Displays the data conversion status in the result display
Show. Click the Finish button to end data conversion
Then, the process returns to the main window (a). As described above, the embodiment of the agent device of the present invention is described.
However, the present invention is limited to the described embodiment.
It is not specified and is within the scope described in each claim.
It is possible to make various modifications. For example, explain
In the embodiment, the scenario is downloaded from the homepage.
When downloading, when the user performs download processing
Has been described, but the predetermined start conditions (date and time and location)
When the conditions are met, the scenario is automatically
It is also possible to make a scenario to acquire. New acquired
A system that guides services such as inns based on a scenario.
Nario is added and executed. As a modification of this embodiment, the operation
Control device, autonomous control device, agent device, scenario
Data creation device, scenario data creation program,
For example, meeting with the passenger of the vehicle
Autonomous control device, etc. that talks and operates equipment autonomously
May be configured as follows. In this case,
The user can easily create and add the contents of the legal correspondence
To provide autonomous control devices and agent devices that can operate
An autonomous control device such as
For creating and editing scenario data to be used on
Nario data creation device, scenario data creation program
The purpose is to provide B. (A) Where an interactive interface is provided
In the operation control device for executing a predetermined operation, the interactive type
The interface requests a human response from the controller
A first step of asking a question to ask
The contents of the human response by the
A second step of initiating a line.
Make it possible to externally input data that specifies
Input device and data input by the input device.
The execution means for executing at least the first step
An operation control device comprising: a step; (B) Judge the surrounding situation and condition and respond to the judgment result.
Autonomous control device that performs the same processing,
Is a scene, and is composed of one or a plurality of continuous scenes.
Scenarios that have been created, and
Scenario description that stores scenario data with start conditions
Storage means and whether the conditions for starting the scenario have been met.
A condition judging means for judging, and a start condition by the condition judging means.
If it is determined that the condition is satisfied, the system corresponding to the start condition
Scenario execution means for sequentially developing each scene of Nario,
Scenario acquisition means for acquiring scenario data from outside
And storing the acquired scenario data in the scenario storage device.
And a scenario adding means for storing in a column.
Autonomous control device. (C) Determine the condition of the vehicle interior and the vehicle, and
A function that autonomously performs the corresponding processing, the appearance of the agent
Agent device that executes in conjunction with the movement and voice of
Then, the image of the agent with a predetermined appearance is
Display means to be displayed in the device and processing performed by the device
Content and / or at least one of agent's image and sound
A scene composed of one scene, one or a plurality of continuous
Scenario composed of different scenes and deployment of the scenario
Scenario data having a start condition for starting
Scenario storage means that meets the scenario start condition.
Condition determining means for determining whether or not the condition has been determined;
If it is determined that the start condition is satisfied at the stage,
Scenario that sequentially develops each scene of the scenario corresponding to
Execution means and scenario to acquire scenario data from outside
E acquiring means, and the acquired scenario data
Scenario storing means stored in the Rio storage means.
An agent device characterized in that: (D) The scenario consists of a proposal by the agent, quality
Start with a scene that contains active actions such as questions, greetings, etc.
The agent device according to (c), wherein (E) Scenario data acquired by the scenario acquisition means
If the executable is not an executable,
Conversion means for converting to scenario data,
Adding means for converting the scenario data converted into an executable format into the executable data.
(D) is stored in the nario storage means.
Agent device described above. (F) One process autonomously performed by the device is defined as a scene.
Or, a scenario composed of multiple consecutive scenes
Autonomous type that starts when the start condition of the scenario is satisfied
Scenario data for creating scenario data for control devices
Creation device, selectable as scenario start condition
Start condition element acquisition means for acquiring multiple start condition elements
And an autonomous control device that can be selected as a scene component
Scene element acquisition means for acquiring the processing content of
To create a scenario start condition from the start condition elements
From the starting condition creation means and the acquired scene component,
A scene creating means for creating the processing content of each scene;
To move from one scene to the next
The transition conditions and the destination scene corresponding to each transition condition are specified.
Create deployment configuration for each scene from the migration destination data
And a scene development configuration creating unit that performs
Scenario data creation device. (G) The details of the processing that the device performs autonomously and the agent
A scene composed of at least one of an image and sound
A scene is composed of one or multiple consecutive scenes
Scenario when the start condition of the scenario is satisfied
Create scenario data for the agent device to be started
A scenario data creation device,
To get multiple start condition elements that can be selected
Selectable as condition element acquisition means and scene elements
Agent processing, agent image and audio data
Scene element acquisition means for acquiring at least one of data
And the start condition of the scenario from the obtained start condition element
Starting condition creating means for creating a scene structure,
From the components, the processing contents by the agent, the agent
A scene composed of at least one of the image of the
A scene creating means for creating a surface as one scene;
One or more transitions to transition from one scene to the next
Identify the destination scene according to the line condition and each transition condition
Create a deployment configuration for each scene from the migration destination data
And a scene development configuration creating means.
Scenario data creation device. (H) The scenario is a proposal by an agent, quality
Start with a scene that contains active actions such as questions, greetings, etc.
The scenario data described in (g).
apparatus. (I) The created start condition, scenario, and scene
From the expanded configuration, the executable format that can be used on the agent device
Conversion means to convert to scenario data
Creation of scenario data described in (g) or (h)
apparatus. (J) The start condition creating means includes the acquired start condition.
Window is displayed and the start condition element is selected.
And the scene creation unit creates the start condition.
Of the scene components
Image and audio data are displayed in separate windows, and
Create scenes from selected scene components.
Scenario data described in (g), (h) or (i)
Data creation device. (K) The scene development configuration creation means uses each of the created scenes.
Icons in the form of icons, and the deployment
(G) to (j), which are displayed in a dough display
The scenario data creation device according to any one of the above. (L) Processing contents autonomously performed by the device and agent
A scene composed of at least one of an image and sound
A scene is composed of one or multiple consecutive scenes
Scenario when the start condition of the scenario is satisfied
Create scenario data for the agent device to be started
Scenario data creation program
Acquires multiple start condition elements that can be selected as start conditions
Start condition element acquisition function
Selectable agent processing contents and agent images
Scene element acquisition for acquiring at least one of audio data
Start a scenario from the function and the obtained start condition element
A start condition creation function for creating conditions and the acquired
From the component elements, the content of processing by the agent,
And at least one of the output audio
Scene creation function to create a scene as one scene
1 for moving from the detected item to the next scene of the scene
Or, a plurality of transition conditions and the destination
Development of each scene from migration destination data specifying the scene
A scene development configuration creation function for creating a configuration
Data creation program for realizing
M According to this modification, the contents of the autonomous response
Autonomous system that allows users to easily create and add
A control device and an agent device are provided. Also this
Such as autonomous control devices and agent devices
Scenario data creation device to create and edit Rio data,
A scenario data creation program is provided. According to the operation control device of the present invention, the control device
The first step is to ask the human for an answer from the
Provide autonomous response to human beings with steps
be able to. In addition, the control device requests a response from humans.
The data that defines the content of the question
It can be executed by inputting from a section.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 shows a system configuration diagram of a scenario addition system using an agent device according to an embodiment of the present invention. FIG. 2 is a block diagram illustrating a configuration of an agent device according to an embodiment of the present invention. FIG. 3 is a configuration diagram of an agent processing unit in the agent device. FIG. 4 is an explanatory diagram conceptually showing a data structure and data contents of a scenario file in the agent device. FIG. 5 is an explanatory diagram showing a scene screen displayed on a display device based on scene data of a scenario. FIG. 6 is an explanatory diagram conceptually showing a configuration of driver information in the agent device. FIG. 7 is a flowchart illustrating an operation of a scenario adding process by the agent device. FIG. 8 is a flowchart illustrating an operation of a scenario activation determination process. FIG. 9 is an explanatory diagram showing, for each scene, transition of a scene screen according to a guidance scenario transmitted from a ryokan to a potential guest. FIG. 10 illustrates a configuration of a scenario creation device. FIG. 11 shows contents and configurations of a scenario editing program 251 and scenario editing data 252. FIG. 12 is an automatic activation item table conceptually showing a part of contents of a start condition element that can be selected as a scenario start condition (scenario start condition). FIG. 13 is a table showing detailed conditions of an item described as “selectable item list” in a remarks column of an automatic activation item in FIG. 12; FIG. 14 is a scene branch item table in which branch items (transition conditions) for branching from a scene to the next scene are stored. FIG. 15 illustrates an additional condition table for setting branch conditions in more detail. FIG. 16 illustrates a configuration of a main window displayed on a display device when a scenario editor 251 is activated. FIG. 17 illustrates a flow of a screen operation for editing a scenario property. FIG. 18 illustrates a flow of a screen operation for editing a scenario start condition from a scenario start condition main edit window. FIG. 19 illustrates a configuration example of a selection window of a condition range to be automatically activated in a case of inputting a numerical value. 20 shows a flow of a screen operation for selecting a screen configuration to be displayed on the agent display screen 51. FIG. FIG. 21 illustrates a flow of a screen operation for editing a character motion (agent operation) instruction. FIG. 22 illustrates a flow of a screen operation for editing a speech instruction of a character (agent). FIG. 23 illustrates a flow of a screen operation for editing a voice recognition dictionary. FIG. 24 illustrates a flow of a screen operation for editing a flow of a scenario. FIG. 25 illustrates an extended version of a screen operation flow for editing a scenario flow. FIG. 26 illustrates a flow of a screen operation for editing an end position of a scenario. FIG. 27 illustrates a flow of a screen operation for converting a created scenario into an executable format usable for navigation. [Description of Signs] 1 Agent device 2 Scenario creation device 3 Server 4 Scenario file 9 Overall processing unit 10 Navigation processing unit 11 Agent processing unit 12 External I / F unit 13 Image processing unit 14 Voice control unit 15 Situation information processing unit 16 Input Control unit 17 Storage device control unit 22 Input device 24 Communication control unit 25 Audio output device 26 Microphone 27 Display device 28 Imaging device 29 Storage device 30 Agent data 31 Navigation data 40 Various status detection devices 41 Current position detection device 42 Status detection unit

   ────────────────────────────────────────────────── ─── Continuation of front page    (72) Inventor Koji Hori             2-19-12 Sotokanda, Chiyoda-ku, Tokyo             Equos Research Co., Ltd. (72) Inventor Hiroaki Kondo             2-19-12 Sotokanda, Chiyoda-ku, Tokyo             Equos Research Co., Ltd. (72) Inventor Manabu Matsuda             2-19-12 Sotokanda, Chiyoda-ku, Tokyo             Equos Research Co., Ltd. (72) Inventor Tadashi Hirano             2-19-12 Sotokanda, Chiyoda-ku, Tokyo             Equos Research Co., Ltd. (72) Inventor Kazuaki Fujii             2-19-12 Sotokanda, Chiyoda-ku, Tokyo             Equos Research Co., Ltd. F term (reference) 2F029 AA02 AB01 AB07 AB13 AC02                       AC06 AC09 AC12 AC14 AC18                 5E501 AA22 AC16 BA13 BA17 CA04                       CA06 CB01 CB15 DA01 EA32                       EA34                 5H180 AA01 BB02 BB04 BB05 BB12                       BB13 CC04 FF04 FF05 FF12                       FF22 FF25 FF27 FF33 FF40

Claims (1)

Claims: 1. An operation control device for executing a predetermined operation having an interactive interface, wherein the interactive interface performs a first step of asking the control device to ask a human for an answer. And determining the contents of the human answer in the first step,
A second step of starting execution of the predetermined operation,
An input device for externally inputting data defining the contents of the inquiry to the human, and execution means for executing at least the first step in accordance with the data input by the input device. Operation control device.