KR100572683B1 - Authoring tool for content creation of intelligent robot and content creation method of intelligent robot using the same - Google Patents

Abstract

The present invention relates to an authoring tool for creating an intelligent robot's content and a method for producing an intelligent robot's content using the same, and more particularly, to intelligently program an intelligent robot based on a multimedia and a method of effectively controlling hardware and a display device of the intelligent robot. The present invention relates to an authoring tool that can be used to provide a method of producing content for a robot that provides a user with a more human-friendly interface.

By using the robot authoring tool of the present invention, even a user without hardware control programming experience can control the robot's head and wheels, the eye and mouth LEDs forming a facial expression, and the chest LED expressing emotions with a little learning. By modifying various different parameter values for each type, you can conveniently create the desired motion of the robot, which reduces the time required for programming the robot and reduces the effort. Therefore, the general public can easily create a multimedia-based robot program using his robot. In addition, it was developed to check the synchronization of the content even without a robot through a three-dimensional simulator.

Intelligent Robot, Robot Control, Robot Programming, Robot Content, Robot Multimedia, Robot Editor, Robot Simulator

Description

Contents authoring tool for intelligent robots and contents developing method using the tool}             

1 is a screenshot showing the authoring environment of the present invention,

2 shows a part directly driven by the present invention as a robot under development,

FIG. 3 is a robot action window included in the static object information window of FIG. 1 which can define the control of an actual robot in detail.

4A shows head action types and parameters,

4b shows wheel action types and parameters,

4C is a chest LED (Light Emitting Diode) type and parameters,

4d shows eye LED types and parameters,

4e, 4f and 4g show the LED type and parameters,
4h is a diagram illustrating an example of a frame in which an eye LED is operated;
4i illustrates an example of a frame in which an input LED is operated;

5A is an LED action parameter upper and lower limit restriction formula,

5B is a head action parameter upper and lower limit restriction formula,

5C is a wheel action parameter upper and lower limit restriction formula,

Figure 6a is a voice recognition window having voice recognition information of the robot,

6B is a window for registering file and page global recognition words;

6C is a window for registering a frame section recognition word;

7 is a window defining a voice synthesis (TTS) of the robot,

8 is a player screen for playing content produced by an authoring tool;

9 shows the structure of a script used for the purpose of controlling content for a robot in an authoring tool,

10A to 10E are script embedded keyword lists, FIG. 10A is a sentence type of a script, FIG. 10B is a script global function type, FIG. 10C is a type of script event, FIG. 10D is a type of script system function, and FIG. 10E is a script usage symbol. Each represents a type of,

11 shows an embodiment of robot content creation,

12 is a timeline structure diagram;

13 is a robot action structure diagram,

14 is a speech recognition structure diagram.

<Explanation of symbols for the main parts of the drawings>

101: content screen area

102: Timeline Area 103: Head Action Layer

104: Wheel Action Layer 105: Eye LED Layer

106: mouth LED layer 107: chest LED layer

108: speech synthesis layer 109: speech recognition layer

110: drawing toolbar 111: general object information window

112: static object information window

201: head 202: wheels

203: eyes 204: mouth

205: Bust 206: TFT LCD (touch screen)

301: Command type 302: Command code

303: Command name 304: Parameter type

305: Upper and lower limit 306: Action type

801: content player 802: player control bar

803: 3D Simulator

901 lexical analyzer 902 parser

903: Launcher

1101: content screen area 1102: timeline window

1103: Voice recognition registration window

The present invention relates to an authoring tool for contents creation of an intelligent robot and a method for producing contents of an intelligent robot using the same, and more particularly, an authoring tool for programming a multimedia-based intelligent robot for the purpose of producing contents for a robot, and The present invention relates to a method for producing contents of an intelligent robot that can effectively control hardware and display devices of an intelligent robot.

In particular, the present invention relates to a multimedia-based control program and content production method developed for programming a robot.

Until now, in order to control robots, programmers had to code in the program on a case-by-case basis the overall details such as robot actions, speech recognition, speech synthesis, and graphical user interface (GUI) at the code level. As well as professional knowledge, there was a problem that takes a lot of time and money.

Due to these technical limitations, the commercialization of intelligent robots could not meet the necessity of developing a large amount of contents for robots that effectively deliver systematic information to end users by utilizing various robot resources.

There was a multimedia authoring tool for real-time lectures developed in the past, but this authoring program downloads multimedia contents existing in the server to a personal computer (PC) in real time and plays data such as general objects, sounds, and videos through a separate player. It was developed for the purpose of remote lectures, so it could not produce content for robots that contained various robot actions.

The present invention is to provide a multimedia-based control program and a method for manufacturing the robot developed in order to program the robot in view of the above points, the general user, not a programmer, the action of the robot, speech recognition, speech synthesis, An object of the present invention is to provide an authoring tool for content creation of an intelligent robot that can effectively control a robot by combining a GUI and the like, and a method for producing an intelligent robot content using the same.

In accordance with the achievement of the above object, the present invention may facilitate modification of changes and future updates, thereby saving a lot of cost and time when producing contents for a robot.

In addition, in the conventional case, it was possible to process only the defined action of the robot, but by providing various types of parameter attributes for each kind of action, the content creator can obtain the effect of implementing the more accurate action of the robot.

In addition, unlike preconceived notions about robots, robot content creators can arrange robot actions in the form of timeline to effectively use LEDs attached to eyes, mouth, and chest to provide more human-friendly services. Can be applied.

Particularly, according to the present invention, a content producer who does not own a real robot can create a content using an authoring tool, and then completely simulate the synchronization between the content and the robot through a player, a speech synthesis engine, a recognition engine, a simulator, and the like.

The present invention for achieving the above object is the entire robot control and multimedia-based authoring function, such as robot action, speech recognition, speech synthesis, and the content player to enable the same type of simulation as the robot even without the robot 3 Including dimensional simulators; The fixed layer includes the head action layer 103, the wheel action layer 104, the eye LED layer 105, the mouth LED layer 106, the chest LED layer 107, the speech synthesis layer 108, and the speech recognition layer. There is a layer list that displays the parts related to the direct control with the robot, including (109), the drawing toolbar 110, general object information window 111, robot action information and voice synthesis, and voice recognition registration required for content creation. Provides an authoring tool for producing content of an intelligent robot, characterized in that it comprises an authoring environment with a static object information window 112 for.

The fixed layer includes a layer that displays multimedia objects such as flash and sound and script control, and the timeline includes a robot action that can be driven in real time so that various objects and robot actions can be controlled in a complex manner. Inserting is characterized in that the structure is executed in synchronization with the content.

In the authoring tool, a script used for the purpose of controlling robot contents is used, wherein the script interpreter is a lexical analyzer 901 for separating each token into input of a script source input by a user, and a generated grammar as an input. It is characterized in that it consists of a parser (902) for retrieving, and generating a table to be used at run time, and an executor (903) for executing an action on an object with the run table as input.

The method of the present invention for achieving the above object is defined by a unique code for the command type 301 for distinguishing the robot control parts such as the head, wheel, eyes, mouth, chest, and the like and possible action patterns for each controlled part. Command code 302, command name 303 given for each action code, parameter type 304 for detailed control of the robot action, upper and lower limit values indicating the limit range that can be processed by the hardware of the robot 305 Using the authoring tool of claim 1 comprising a robot action window composed of

In the timeline of the authoring tool, the operation of the head 201 and the wheel 202 is controlled through the structure of the frame progress method, the effective face expression is controlled through the control of the eye 203 and the mouth LED 204, and the chest LED is controlled. Through 205, the emotional state of the robot can be delivered to the end user.

In addition, after the robot content is produced using the authoring tool, when the PC is played on the PC, the voice recognition, the synthesis engine, and the 3D robot simulator are driven together with the content player; In the content player 801, the robot content produced by the authoring tool can be simulated in advance, and in a real robot, a command generated in the player is transmitted to the robot, and in a PC, a command is transmitted to the simulator; The driving of the three-dimensional simulator is characterized in that the movements of the head, wheels, eyes, mouth, chest, etc., which are the same as those of the actual robot, can be realistically expressed.

At this time, the content player and the simulator is characterized in that it operates in the same way as the robot receives the input by the TCP / IP communication.

In particular, a lexical analyzer 901 for separating each token with input of a script source input by a user, a parser 902 for retrieving a grammar using input generated tokens, and generating a table to be used at execution time, an execution table After inputting a script with the authoring tool in the step of controlling the contents for the robot in the authoring tool using a script composed of an executor 903 that executes an action on an object by inputting the script, the script source is The lexical analysis and parsing point is before the content is executed, and if there is no grammatical error as a result of the lexical analysis and parsing, a syntax tree is created. It is characterized in that the execution.

Robot authoring tool according to the present invention (hereinafter, authoring tool) is a program developed for the purpose of producing the content that can control the robot using the authoring tool multimedia content that can be linked to the robot in a short period of time even if the creator without knowledge of robot programming It is proposed to produce.

Authoring tool of the present invention is composed of the entire robot control, such as robot action, speech recognition, speech synthesis, multimedia-based authoring function, and the content player and 3D simulator to enable the same type of simulation even without the robot It is.

The simulation is designed to provide the PC with the same environment as the robot such as player playback of contents, speech recognition engine, speech synthesis engine, and simulator.

By utilizing the various functions provided by the authoring tool, it is possible to provide the information needed by the user more effectively by developing high quality contents required by the robot.

Hereinafter, the present invention will be described in detail with reference to the drawings.

1 is a screenshot showing the authoring environment of the present invention.

The content screen area 101 in the center of the screen is a screen area for editing content to be displayed on the robot. The timeline area 102 at the bottom of the screen is composed of an area showing a layer list of the timeline window and an area displaying a frame and an action block.

The layer is divided into a fixed layer related to robot control and a general layer displaying general objects. The fixed layer includes a head action layer 103, a wheel action layer 104, an eye LED layer 105, and an mouth LED layer 106. There is a chest LED layer 107, and there is a layer list that displays parts related to direct control with the robot, including speech synthesis layer 108 and speech recognition layer 109.

In addition, a layer for displaying multimedia objects and script controls such as flash and sound is included in the fixed layer. The timeline is designed to effectively control different objects and robot actions in a complex manner. When the robot action that can be driven is inserted into the frame of the timeline, the timeline is synchronized with the contents and executed.

In addition, there is a drawing toolbar 110 and a general object information window 111 necessary for content creation on the left and right sides of the screen, and a static object information window 112 on the right center for robot action information, voice synthesis, and voice recognition registration. The pattern registration, voice synthesis, and voice recognition words of the new robot action are all registered in the static object information window.

2 is a view showing a part directly driving in the present invention as a robot currently in development. In the timeline of the authoring tool, it is possible to control the operation of the head 201 and the wheel 202 through the structure of the frame progression method, and the effective face expression can be controlled through the control of the eye 203 and the mouth LED 204. The chest LED 205 may transmit the emotional state of the robot to the end user.

At this time, the LED is defined by three colors for each emotion to express the mood state of the robot through the color.

For example, when the chest LED is orange, it expresses a happy feeling, when it is red, it expresses feelings of anxiety, and when it is blue, it expresses feelings of sadness.

In addition, the layer structure of the timeline provided by the present invention provides a production environment that allows content creators to directly define matters related to Korean-English speech synthesis (TTS) and Korean speech recognition. Reference numeral 206 denotes a TFT LCD (touch screen).

FIG. 3 is a robot action window included in the static object information window of FIG. 1 that can define the control of an actual robot in detail. The authoring tool basically provides various action types for robot control. In addition, the user can register and use a new robot action by defining parameters for the robot action required for content creation.

The window of the robot action is composed of an instruction type 301, an instruction code 302, an instruction name 303, a parameter type 304, and an upper and lower limit value 305. The command type 301 distinguishes robot control parts such as head, wheels, eyes, mouth, and chest.

The command code 302 is defined as a unique code for the action pattern possible for each part to be controlled. For example, the action of shaking the head up and down is defined as HUD0 combining H (Head) U (Up) D (Down), and the action code to rotate the wheel left and right is W (Wheel) L (Left) It is defined as WLR0 combining R (Right). The command code does not change even if the robot content creator registers a new action by changing the parameter value property.

The command name 303 is a name given to each action code, and when the action type is added by user definition, the action code is unique, but the command name can be defined and identified differently.

The parameter 304 is designed to allow detailed control of the robot action. The parameter 304 has different parameter types according to the characteristics of the controlled part, and has a holding time and a repetition frequency parameter value in common.

Every robot action has a holding time of one second and a value of one iteration. The retention time and the number of repetitions can be arbitrarily set by the robot content creator, and the types of parameters such as the moving angle, the moving distance, the moving length, the moving end point, and the moving direction vary depending on the characteristics of the action.

If the creator wants to increase the holding time, change the number of repetitions of the action, or change the parameter value such as the moving angle or the moving distance for synchronizing the TTS or narration sound, etc., the desired parameter value in the robot action window You can add a new action pattern by typing.

At this time, the repetition frequency of the robot action is affected by the holding time, the moving distance or the angle, and all parameters have an upper and lower limit value range.

The upper and lower limit values 305 represent the limit ranges that can be processed by the hardware of the robot. The upper and lower limit values 305 check the upper and lower limits internally with respect to the parameter values input by the user and notify the user when a message is exceeded.

The default values for all robot actions are once per second, and content creators can define values within the upper and lower limits.

For example, the maximum head angle of the robot head is 10 degrees. If you enter more than 11 degrees of the moving angle parameter, the parameter upper limit exceeded message will occur. Alternatively, excessive hardware control, such as performing the head of the robot up and down 100 times or more in a holding time of 1 second, is prevented.

As defined above, the action pattern defined in the robot action registration window has a structure in which the robot action is automatically inserted into the corresponding layer by action type when dragging and dropping to the desired frame of the timeline.

The action type 306 shows a list of action codes that can be used for each part controlled by the robot. If the list is selected and placed in a frame of the timeline, the action is added to the timeline.

4A to 4G define the types and parameter attributes of robot actions. In the table, CODE is the action code, VALUE is the default value, ACTIVE NAME is the action name, PARAMETER is the parameter, DEFAULT is the default value, MAXIMUM is the maximum value, MINIMUM is the minimum value, and COLOR is the color.

While developing contents for robots, it was necessary to define various motion patterns by effectively controlling various control parts on hardware, and it was required to express various motions with more precise control in content planning. By assigning parameter types and values appropriate to the characteristics, the robot can be freely controlled at the level of content creation in the authoring tool.

The types of parameters are basically the maintenance time and the number of repetitions of the robot action. In addition, there are various types of head and wheel movement angles, movement lengths, movement distances, movement end points, and movement directions. Referring to this table, an example of defining the operations of various robots in one action pattern will be described.

-Define the top left view of the robot head

HSE0 (View specific direction) Holding time 10 seconds, 1 repetition time, Moving end point (left 45 degrees, upper 10 degrees)

-Nod five times up and down the robot head

HDN0 (nod up and down) holding time 10 seconds, 5 repetitions, moving angle (10 degrees)

-Return to home after 10 minutes

HST0 (Stare in a specific direction) Holding time 1 second, 1 repetition time, Moving end point (0 degree left, 10 degree upward)

After 10 minutes,

HST0 (Stare in a specific direction) Holding time 1 second, 1 repetition number, Moving end point (0 degrees left, 10 degrees upward)

-Rotate left and right 10 times

WLR0 (rotate left and right) holding time 20 seconds, repeat number 10 times, moving angle (45 degree)

-Blue chest LED keeps on for 1 minute

LHG0 (Sad Feelings) Duration 60 seconds, 1 repetition time, Color Definition Blue

-Blue chest LED flashes continuously for 1 minute

LHG0 (Sad Feelings) Duration 60 seconds, 60 repetitions, Color definition Blue

Red Eye LED flashes for 1 minute

LES0 (surprise) retention time 60 seconds, 30 repetitions, color definition red

-Specific green LED blinks for 5 seconds

LSM0 (Smile) Duration 5 seconds, 5 repetitions, Color definition Green

-Orange specific mouth LED blinks for 10 seconds

LSP0 (Surprise) Retention time 10 seconds, Repetition frequency 10 times, Color definition Green

4A shows seven types of action patterns related to head control.

Action name defined by action code and default value, minimum value and maximum value of each parameter are defined. The holding time refers to the time to keep the action. The unit is seconds. You can define up to 3,600 seconds, or 1 hour. The number of repetitions is an attribute that repeats the action with respect to the holding time. The maximum value of the number of repetitions is affected by the holding time, the moving angle, the rotation angle, and the number of shaking times. Refer to FIG. 5 for an upper and lower limit restriction formula of the repetition number.

The types of head action parameters vary and vary depending on the characteristics controlled during operation. In case of action code HDN0, which moves head up and down, the parameter type is only moving angle except for holding time and repetition frequency.However, in case of action code HSK0, parameter to be defined as shaking count, shaking angle, leaning angle, etc. There are more varieties than HDN0.

4B is a total of four types of action patterns related to wheel control.

Moving forward and backward through the action code WFB0 is defined as returning to the starting position after reciprocating a 10cm travel distance at a rate of once per second. At this time, the maximum movement distance of the wheel is 10cm, and if a parameter value of 11cm or more is input, an upper limit value exceeded message is generated.

4C is defined as four action patterns related to chest LED control.

The orange LED turns on when you are happy, the blue LED turns on when you are sad, and the red LED turns on when you are anxious. The chest LED is OFF when it is normal and calm feeling.

4D is a total of 10 types of action patterns related to eye LED control.

The default hold time and repeat count are defined as one blink per second, just like any other action. In order to control the blinking speed of the eye according to the mood of the robot, the number of repetitions per second may be added or decreased.

For example, in order to express the boring emotional state of a robot, if you define 10 seconds of holding time and 3 times of repetition, it can express the blinking of an eye slower than the default blinking once per second. In addition, it can be expressed in three colors: blue, orange, and red.

4H is a diagram illustrating an example of a frame in which the eye LED is operated, and shows a frame showing an LEW0 (wink) action. In the case of the eye, it shows 3 frames of LED actions per action.

4E, 4F, and 4G are 44 types of action patterns related to input LED control. Like the chest and eye LEDs, they blink at a rate of once per second, and the LEDs are designed to express emotions in three colors: green, yellow, and red.

FIG. 4I is a diagram illustrating an example of a frame in which an input LED is operated, and is a frame showing an LMR0 (mock 3) action. FIG. In the case of mouth, it shows 10 frames of LED action per action, and the play time can be adjusted by parameter.

5A to 5C show equations for checking the total number of repetitions by the robot action parameter input value.

The number of repetitions of the robot action is affected by the holding time and the angle and distance, and the upper limit of the angle and distance is set as the limit of the value that can be processed in hardware. By calculating the upper and lower limits of the parameter values assigned to each action pattern according to the holding time, the robot error caused by the exceeding value is prevented.

In addition to the holding time and the number of repetitions, there are various types of robot action parameters such as the property corresponding to the direction control such as the moving direction and the end point, the property corresponding to the angle control such as the swing angle and the bowing angle, and the short and long lengths.

For example, suppose that you can move the wheel back and forth 10 cm in a holding time of 1 second. If you enter 10 cm twice a second, you will get a message that the upper limit of the number of repetitions is exceeded. If defined, it will prevent the error from exceeding the upper limit. By applying the upper and lower limit formulas to the authoring tool, it is possible to prevent errors in hardware operation that may occur in the robot.

6A is a voice recognition window containing voice recognition information of a robot.

There are two types of speech recognition, a single recognition word and a section recognition word. The single recognition word corresponds to a file recognition word and a page recognition word, and the section recognition word includes a frame recognition word. The characteristics are roughly as follows.

The file and page recognition words define the recognition words used globally in a specific robot content file and a specific page of a specific file, respectively. Section recognition words are frame recognition words that are activated only in the start and end frame sections of a specific range. It can be defined.

The frame identifier has the advantage of defining different link paths of two or more identical identifiers within the same content page. However, in the same section, different link paths cannot be processed by overlapping the same recognized word.

6B is a window for registering file and page global recognition words. 6A is a screen that appears when a recognition type is selected in the voice recognition window of FIG. 6A and moved to a registration window, and includes items for setting a recognition word, a connection type, and a path.

For example, if the robot recognizes the word 'apple', it is possible to link to a specific frame of a specified page of a specific content file. When the recognition result is successful, there is an advantage that interactive robot contents can be produced between the robot and the user by changing the link path.

6C is a window for registering a frame section recognition word. Unlike file and page identifiers, frame identifiers are not recognized globally from the first frame to the end frame. Instead, frame identifiers may need to use localized identifiers within a specific frame or process the same identifier as different link paths within the same page. When used.

For example, it is useful when it is necessary to ensure that the correct answer recognition word is recognized only in a frame requiring correct answer after presenting the problem, rather than making the correct answer word recognized throughout the page. Alternatively, this function is used when the same recognition word requires different processing for each frame section.

7 is a window defining the voice synthesis (TTS) of the robot. In the authoring tool, the same Korean and English speech synthesis engine as the robot is built in. When the content is produced in the authoring tool, the Korean-English mode can be selected and registered (701). When registering a voice synthesis, the voice synthesis registration window may express suitable voices according to various emotion changes by using the attributes of the pitch and speed parameters 702 and 703 of the robot voice synthesis. Reference numeral 704 denotes speech synthesis content.

8 is a player screen for playing content produced by the authoring tool. When the robot content is created using the authoring tool and played on the PC, the voice recognition, synthesis engine, and 3D robot simulator are driven together with the content player.

The content player 801 may be referred to as a playback program for viewing the robot content produced by the authoring tool on the display device of the robot. The player is structured to operate in the same way as the robot and PC, and the program is also configured as one. In a real robot, a command generated by a player is transmitted to the robot, and in a PC, a command is transmitted to the simulator.

When the player is driven in the PC environment, the same voice synthesis and speech recognition engine as the robot is executed together, so that the robot contents can be tested on the computer in the same manner as the input / output environment of the robot.

Therefore, after producing the robot content, the player can run the player and check the content screen, voice recognition, voice synthesis, and the operation status of the robot, and fully test the content, thereby minimizing the debugging stage of the content and the operation error of the robot. have.

The 3D simulator checks the driving state of the actual robot that operates in synchronization with the contents. It is designed to simulate the robot even without the robot. By utilizing the 3D simulator, many robot content creators have the advantage of being able to fully simulate the contents that are driven with the robot even if they do not own the actual robot.

The driving of the 3D simulator realistically expresses the movements of the head, wheels, eyes, mouth, and chest, which are the same as those of a real robot. At this time, the content player and the simulator receive input by Transmission Control Protocol / Internet Protocol (TCP / IP) communication and operate like a robot.

The player control bar 802 is a tool necessary for playing content on a PC. The player control bar allows you to pause and play, move to the first frame, move to the end frame, and move the page navigation of the content. This does not appear when the content player is in robot mode.

9 is a diagram showing the structure of the script used for the purpose of controlling the content for the robot in the authoring tool, the advantage of using the script, it is possible to produce content with a variety of processing forms that are not defined when designing the logic of the learning content have.

As shown in FIG. 9, the script interpreter includes a lexical analyzer 901 that separates each token into input of a script source input by a user, retrieves a grammar from the generated tokens, and generates a table to be used in execution. The parser 902 is configured to include an executor 903 that executes an action on an object using the execution table as an input.

After inputting the script with the authoring tool in the content production stage, the time when the script source is lexical analyzed and parsed in the produced content is before the content is executed.

If there is no grammatical error as a result of lexical analysis and parsing, a syntax tree is created, and the contents instructed by the script are executed using this syntax tree when executing content. Specific examples will be described below.

10A to 10E are script built-in keyword lists, FIG. 10A is a script sentence type, 10B is a script global function type, 10c is a script event type, 10D is a script system function type, and 10e is a script usage symbol type, respectively. Indicates.

The following embodiment will be described with reference to FIG.

The content screen area 1101 is a content screen area seen on a 12.1 "LCD liquid crystal display device attached to the body of the robot. An object can be created by using a vector drawing tool provided by the drawing toolbar or an external macro. You can also load Media Flash Shockwave files and add robot actions only.You can insert navigation buttons for content by inserting buttons and inserting them into the content.

The timeline window 1102 of the authoring tool shows that the actions necessary for robot control are inserted in a specific frame corresponding to each layer as shown in the figure. In order to insert a robot action into the timeline frame, select the desired robot action code from the robot action list of the robot information window and drag it to the frame to automatically find the layer and add it to the frame.

In the case of the voice synthesis, the robot inputs the contents to be spoken by the TTS in the voice synthesis window, registers the text, and drags the desired voice synthesis list to the desired frame. The voice synthesis layer is automatically added to the voice synthesis layer. When the player runs, voice synthesis transmission starts at the corresponding frame.

The voice recognition word is also displayed on the voice recognition layer of the timeline. When the recognition word and the link path to be processed are input in the recognition word registration window 1103, the voice recognition word is automatically added. As shown in the figure, the inserted actions are processed in the order of time as the frame progresses when the content is played, and actions for actions are generated in the robot or the simulator.
12 is a timeline structure diagram where CTimeLine shows time for an action on an object, a frame indicates a frame of time, and a layer shows the order in which the object shows.
Overlapping objects are not visible here.
CFrame has a vertical link to the block. A block is connected to an object. CLayer has a horizontal link to the block. Blocks have objects connected to them. In CActionBlock, a block represents an action section. This block is connected to only one object.
CVoiceRecognitioninfo shows information about speech recognition. CTTSinfo shows information about speech synthesis. CRobotCommandinfo shows information about the robot action. CPageVoiceRecognitioninfo contains information about page speech recognition. CFileVoiceRecognitioninfo contains information about file speech recognition.
13 is a structure diagram of a robot action, in which RBTObject stores basic information about an object. CRobotObject has a robot object. Robot objects exist only on the timeline, and unlike regular objects, they always have action sections.
CVoiceRecognition stores information about speech recognition. CTTs store information about speech synthesis. CRobotCommand stores information about robot actions. Parameter stores the parameters for robot action. Value stores the value of the parameter.
14 is a speech recognition structure diagram, where CVoiceRecognition is a speech recognition object.
CContentsLink is a link associated with speech recognition words. CSingleRecognition is a single speech recognition language, in which one word has only one link. WordList is a list of interval speech recognition words. In this recognition word, several words have one link.

As described above, the present invention provides a program environment in which even a person who does not have special knowledge on the control of the robot can produce multimedia contents linked with the robot through short-term learning. In addition, by applying different types of parameters for each body part and operation characteristics of the robot, there is an effect that can newly create a variety of undefined motion of the robot. In addition, it provides hardware control of the robot's facial expressions, emotions, head, wheels, etc. through the authoring tool to control voice synthesis and speech recognition. It can be explained that there is an effect of shortening the cost and reducing the cost. In addition, through the content player and 3D simulator, robot contents created on the PC can be fully simulated and reviewed and watched.

Claims (8)

It includes a complete robot control, such as robot action, speech recognition, speech synthesis, multimedia-based authoring function, and a content player and a three-dimensional simulator that enables a user without a robot to simulate the same form as a robot; In the fixed layer, the head action layer 103, the wheel action layer 104, the eye LED layer 105, the mouth LED layer 106, and the chest LED layer 107 are arranged in the lower left of the screenshot. Below is a layer list that shows the parts related to direct control with the robot, including the speech synthesis layer 108 and the speech recognition layer 109, and the drawing toolbar 110 on the upper left and the right on the screen shots required for content creation. Authoring tool for content creation of an intelligent robot comprising a general object information window 111 at the top, a robot action information and voice synthesis, and a static object information window 112 at the right center for voice recognition registration . The method of claim 1, wherein the fixed layer includes a layer for displaying multimedia objects, such as flash and sound, and script control, and the timeline includes a robot action that can be actually driven so as to control a variety of different objects and robot actions. Authoring tool for content creation of an intelligent robot, characterized in that the structure is executed in synchronization with the content when inserted into the frame of the timeline. The method according to claim 1, wherein a script used for the purpose of controlling the content for the robot in the authoring tool, The script interpreter includes a lexical analyzer 901 for separating each token into input of a script source input by a user, a parser 902 for retrieving a grammar from the generated tokens and generating a table to be used at execution time. And an executor 903 that executes an action on an object with an execution table as an input. Command type 301 for distinguishing robot control parts such as head, wheel, eyes, mouth and chest, command code 302 defined as a unique code for action patterns available for each controlled part, and given to each action code The robot action window of claim 1 includes a command name 303, a parameter type 304 for detailed control of the robot action, and an upper and lower limit value 305 indicating a limit range that can be processed by the hardware of the robot. Using the authoring tool, In the timeline of the authoring tool, the operation of the head 201 and the wheel 202 is controlled through the structure of the frame progress method, the effective face expression is controlled through the control of the eye 203 and the mouth LED 204, and the chest LED is controlled. The content creation method of the intelligent robot, characterized in that to transmit the emotional state of the robot to the end user through the (205). The method according to claim 4, after producing the robot content using the authoring tool, when playing on the PC, the voice recognition, synthesis engine, and the three-dimensional robot simulator together with the content player is driven together; In the content player 801, the robot content produced by the authoring tool can be simulated in advance, and in a real robot, a command generated in the player is transmitted to the robot, and in a PC, a command is transmitted to the simulator; The method of producing a content of an intelligent robot, characterized in that the driving of the three-dimensional simulator to realistically represent the movement of the head, wheels, eyes, mouth, chest, and the same as the actual robot motion. The method of claim 5, wherein the content player and the simulator receive an input by TCP / IP communication and operate in the same manner as the robot. The method according to claim 4 or 5, In the speech recognition, a file and a page recognition word define a recognition word used globally in a specific robot content file and a specific page of the specific file, respectively, and the section recognition word is activated only in a start and end frame section of a specific range. Define a frame identifier, and the frame identifier may define different link paths of two or more identical identifiers within the same content page; When the robot recognizes a certain word, it is possible to link to a specific frame of a specified page of a specific content file.When the recognition result is successful, interactive robot content can be produced between the robot and the user by changing the link path. ; Unlike the file and page recognition words, the frame recognition words are not recognized globally from the first frame to the end frame, and it is not necessary to use the recognition words localized in a specific frame or to process the same recognition words as different link paths in the same page. Content creation method of intelligent robot, characterized in that used when there is. The lexical analyzer 901 of claim 4, wherein each token is separated by input of a script source input by a user, and a parser 902 is configured to retrieve a grammar from the generated token and generate a table to be used at execution time. By using a script consisting of a launcher 903 that executes an action on an object with an execution table as input, the script is controlled by the authoring tool in the authoring step in the content creation step. When the source is lexical parsed and parsed before the content is executed, if there is no grammatical error as a result of lexical parsing and parsing, a syntax tree is created. Content creation method of intelligent robot, characterized in that the contents are executed.

Images