WO2022114263A1 - Action correction system and action correction method for education robot - Google Patents

Action correction system and action correction method for education robot Download PDF

Info

Publication number
WO2022114263A1
WO2022114263A1 PCT/KR2020/016933 KR2020016933W WO2022114263A1 WO 2022114263 A1 WO2022114263 A1 WO 2022114263A1 KR 2020016933 W KR2020016933 W KR 2020016933W WO 2022114263 A1 WO2022114263 A1 WO 2022114263A1
Authority
WO
WIPO (PCT)
Prior art keywords
emotion
correction
value
type
unit
Prior art date
Application number
PCT/KR2020/016933
Other languages
French (fr)
Korean (ko)
Inventor
정지우
Original Assignee
주식회사 제페토로보틱스
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 주식회사 제페토로보틱스 filed Critical 주식회사 제페토로보틱스
Publication of WO2022114263A1 publication Critical patent/WO2022114263A1/en

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1679—Programme controls characterised by the tasks executed
    • B25J9/1692—Calibration of manipulator
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J11/00—Manipulators not otherwise provided for
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J11/00—Manipulators not otherwise provided for
    • B25J11/0005—Manipulators having means for high-level communication with users, e.g. speech generator, face recognition means
    • B25J11/001—Manipulators having means for high-level communication with users, e.g. speech generator, face recognition means with emotions simulating means
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1602—Programme controls characterised by the control system, structure, architecture
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1628—Programme controls characterised by the control loop
    • B25J9/163—Programme controls characterised by the control loop learning, adaptive, model based, rule based expert control
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
    • G09B19/00—Teaching not covered by other main groups of this subclass
    • G09B19/04—Speaking
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
    • G09B7/00—Electrically-operated teaching apparatus or devices working with questions and answers
    • G09B7/02—Electrically-operated teaching apparatus or devices working with questions and answers of the type wherein the student is expected to construct an answer to the question which is presented or wherein the machine gives an answer to the question presented by a student

Definitions

  • the present invention relates to a motion correction system and a motion correction method for an educational robot. Specifically, it relates to a technology for correcting the body motion of an educational robot through emotion correction.
  • Natural language refers to the language we use in our daily life. Natural language processing means analyzing the meaning of such natural language so that a computer can process it.
  • Natural language processing is a field used in speech recognition, content summarization, translation, user sentiment analysis, text classification tasks (spam mail classification, news article category classification), question and answer systems, and chatbots.
  • a motion correction system and a motion correction method of an educational robot according to the present invention have the following problems.
  • the present invention has a server and a database including an answer information generating unit, a context correction determining unit, an emotion correction determining unit and a body motion correction calculating unit, and is a motion correction system for an educational robot performed by a computing means.
  • Question information receiving unit to receive; an answer information generating unit for generating answer information for the question information received from the question information receiving unit; an emotion correction determining unit for determining the emotion and strength of the answer information calculated by the answer information generating unit according to a preset criterion; a body motion correction calculation unit for correcting the body motion of the educational robot according to the emotion and the degree of emotion calculated by the emotion correction determining unit; and a driving unit for driving the body by applying the correction value calculated by the body motion correction operation unit.
  • the answer information generating unit may further include a context correction determining unit for calculating the answer information of the question information by grasping the context from the question information received before the question information.
  • the answer information generating unit, the context correction determining unit and the emotion correction determining unit may analyze information using a natural language processing technique or a deep learning technique.
  • the answer information generating unit determines that there is dialogue information received before the question information
  • the question information or the answer information is sent to the context correction determining unit
  • the context correction determining unit receiving the question information before Compare the conversation information with the context information stored in the database, and when it is determined that the context of the question information and the previous conversation information falls within a preset context range, the context correction determining unit includes a context keyword within a preset context range You can create response information.
  • the answer information generating unit determines that there is no dialogue information received before the question information, or when the answer information generation unit determines that there is dialogue information received before the question information, the question information is converted into a context correction plate
  • the answer information generating unit may generate answer information including a keyword of a keyword group to which the keyword of the corresponding question information belongs in the database.
  • the emotion correction determination unit may be provided as an emotion type determination unit for quantifying the type of emotion and an emotion strength determination unit for quantifying the intensity of the emotion.
  • the emotion type determination unit may analyze the answer information to classify the emotion types into positive types, neutral types and negative types, and calculate the emotion type values within a preset range.
  • the emotional strength determining unit may analyze the answer information to calculate the intensity of each type as an emotional intensity value within a preset range.
  • the body motion correction calculation unit may calculate a body motion correction value according to the emotion type value and the emotion intensity value calculated by the emotion correction determining unit.
  • the body motion correction calculation unit increases the correction value to a (+) value if the emotion type is a positive type, and if the emotion type is a neutral type, the correction value is not given, and if the emotion type is a negative type, the correction value is It can be increased to a (-) value.
  • the body motion correction calculator may have a measured value within the range of a value when there is no emotional intensity and a value when the emotional intensity is the strongest.
  • the driving unit keeps the eyebrows of the educational robot in a horizontal state, and the correction value is a (+) value as the positive type. If it increases, the inner end of the eyebrow of the educational robot is driven correspondingly to rise, and when the correction value increases to a (-) value in the negative type, the outer end of the eyebrow of the educational robot can be driven correspondingly to rise.
  • the correction value in the positive type and the negative type further increases correspondingly, and the driving range of the inner end of the eyebrow of the educational robot can also be increased correspondingly. have.
  • the driving unit keeps the arm of the educational robot in a normal state, and the correction value is a (+) value as the positive type. If it increases, the arm of the educational robot is driven to move in front of the body, and when the correction value increases to a (-) value in the negative type, the arm of the educational robot can be driven to move toward the rear of the body.
  • the present invention has a server and a database including an answer information generating unit, a context correction determining unit, an emotion correction determining unit and a body motion correction calculation unit, and is a method for correcting a motion of an educational robot performed by a computer
  • the question information receiving unit is the user's Step S1 of receiving question information
  • S2 step of generating, by the answer information generating unit, answer information for the question information received from the question information receiving unit
  • step S5 in which the driving unit drives the body by applying the correction value calculated by the body motion correction operation unit.
  • the emotion correction determination unit of step S3 may be provided as an emotion type determination unit for quantifying the type of emotion and an emotion strength determination unit for quantifying the intensity of the emotion.
  • the emotion type determining unit in step S3 may analyze the answer information to classify the emotion types into positive types, neutral types and negative types, and calculate the emotion type values within a preset range.
  • the emotional strength determination unit in step S3 may analyze the answer information and calculate the strength of each type as an emotional strength value within a preset range.
  • the body motion correction calculation unit in step S4 may calculate a body motion correction value according to the emotion type value and the emotion intensity value calculated by the emotion correction determining unit.
  • the body motion correction operation unit in step S4 increases the correction value to a (+) value if the emotion type is a positive type, and if the emotion type is a neutral type, the correction value is not given, and if the emotion type is a negative type, the correction value is corrected
  • the value can be incremented to a (-) value.
  • the body motion correction calculator in step S4 may have a measured value within the range of a value when there is no emotional intensity and a value when the emotional intensity is the strongest.
  • the driving unit in step S5 causes the eyebrows of the educational robot to maintain a horizontal state when the correction value is not given as the neutral type in the emotion type value, and the correction value is (+) as the positive type
  • the inner end of the eyebrow of the educational robot is driven correspondingly to rise, and when the correction value increases to a (-) value in the negative type, the outer end of the eyebrow of the educational robot can be driven correspondingly to rise.
  • step S5 when the driving unit of step S5 increases the emotional intensity value, the correction value in the positive type and negative type increases correspondingly, and the driving range of the inner end of the eyebrow of the educational robot is also increased correspondingly. can do.
  • the driving unit in step S5 causes the arm of the educational robot to maintain a normal state, and the correction value is (+) in the positive type
  • the arm of the educational robot is driven to move in front of the body, and when the correction value increases to a (-) value in the negative type, the arm of the educational robot can be driven to move toward the rear of the body.
  • the present invention may be implemented as a computer program stored in a computer-readable recording medium in order to execute the method for correcting the motion of the educational robot according to claim 15 by a computer in combination with hardware.
  • the motion compensation system and motion compensation method of the educational robot according to the present invention have the following effects.
  • the educational robot when the educational robot responds, it is intended to achieve non-verbal communication with the user through verbal communication and gestures by correcting the educational robot's facial expression and body motion.
  • FIG. 1 is a schematic diagram of a motion correction system for an educational robot according to the present invention.
  • FIG. 2 is a main configuration of the motion correction system of the educational robot according to the present invention according to the present invention.
  • 3 and 4 are flowcharts of a motion correction method of an educational robot according to the present invention.
  • 5 and 6 are an embodiment showing that the eyebrows of the educational robot are driven in the present invention.
  • 7 to 9 are one embodiment showing that the arm of the educational robot is driven in the present invention.
  • FIG. 10 is a diagram illustrating a computing device according to the present invention.
  • the present invention has a server and a database including an answer information generating unit, a context correction determining unit, an emotion correction determining unit and a body motion correction calculating unit, and is a motion correction system for an educational robot performed by a computing means.
  • Question information receiving unit to receive; an answer information generating unit for generating answer information for the question information received from the question information receiving unit; an emotion correction determining unit for determining the emotion and strength of the answer information calculated by the answer information generating unit according to a preset criterion; a body motion correction calculation unit for correcting the body motion of the educational robot according to the emotion and the degree of emotion calculated by the emotion correction determining unit; and a driving unit for driving the body by applying the correction value calculated by the body motion correction operation unit.
  • the present invention has a server 30 and a database 40 including an answer information generation unit, a context correction determination unit, an emotion correction determination unit and a body motion correction operation unit, and the operation of the educational robot 10 performed by a computing means It is about a calibration system.
  • FIG. 1 is a schematic diagram of a motion correction system for an educational robot according to the present invention.
  • the educational robot 10 and the user 20 can have a conversation, and the educational robot 10 that has received the user 2 's conversation information is a natural language processing algorithm through the server 30 . etc. may be performed.
  • the server 30 performs a calculation function as a computing means, and may communicate with the educational robot through a wireless network or a wired network.
  • the server 30 may be disposed inside the educational robot 10 or may be separately disposed in an independent space in the form of a cloud server or the like.
  • the database 40 stores information necessary for the operation of the server 30 and also stores newly generated information.
  • the database 40 communicates with the server 30 . Each of these actions may be performed by a computer.
  • FIG. 2 is a main configuration of the motion correction system of the educational robot according to the present invention according to the present invention.
  • a motion correction system for an educational robot includes: a question information receiving unit 100 for receiving question information from a user 20; an answer information generating unit 200 for generating answer information for the question information received from the question information receiving unit 100; an emotion correction determining unit 300 for determining the emotion and strength of the answer information calculated by the answer information generating unit 200 according to a preset criterion; a body motion correction calculation unit 400 for calculating motion correction values of at least a part of the body of the educational robot according to the emotion type value and the emotion intensity value calculated by the emotion correction determining unit 300; and a driving unit 500 for driving the body by applying the correction value calculated by the body motion correction operation unit 400 .
  • the answer information generating unit 200 may include a context correction determining unit 210 for calculating the answer information of the question information by grasping the context from the question information received before the question information.
  • the answer information generating unit 200 may analyze information using a natural language processing technique and a deep learning technique.
  • the answer information generating unit 200 may generate answer information for the question information received by the question information receiving unit 100 .
  • the answer information generating unit 200 means that the educational robot 10 answers the question of the user 20 .
  • the answer information of the educational robot 10 may be divided into a case where the user 20 asks the question for the first time, and a case where there is conversational information related to the question before the question.
  • the dialogue information includes question information and answer information.
  • the user asks a question first and the educational robot answers the question, but in some cases, an embodiment in which the educational robot asks a question first may be implemented.
  • the answer information generating unit 200 determines that there is conversation information received before the question information, the question information or the answer information may be sent to the context correction determining unit 210 .
  • the context correction determining unit 210 compares the question information and the previously received dialogue information with the context information stored in the database, and determines that the context of the question information and the previous dialogue information falls within a preset context range. At one time, the context correction determining unit 210 may generate answer information including a context keyword within a preset textual range.
  • the answer information generating unit 200 determines that there is no conversation information received before the question information, or when the answer information generating unit 200 determines that there is conversation information received before the question information to send the question information to the context correction determining unit 210, and the context correction determining unit 210 compares the question information with the previously received conversation information and the textual information stored in the database,
  • the answer information generating unit 200 may generate answer information including a keyword of a keyword group to which the keyword of the question information belongs in the database.
  • the emotion correction determining unit 300 may determine the emotion and strength of the answer information calculated by the answer information generating unit 200 according to a preset criterion.
  • the emotion and intensity determined in this way may be utilized for correcting facial expressions or body motions of the educational robot, which will be described later.
  • the emotion correction determination unit 300 digitizes the emotions contained in the answer, and uses it as the expression of the educational robot. If it is reflected in the movement of the body, the user will be able to interact more with the educational robot.
  • the emotion correction determination unit 300 may be provided with an emotion type determination unit 310 that digitizes the type of emotion and an emotion strength determination unit 320 that quantifies the intensity of the emotion.
  • the emotion type determining unit 310 may analyze the answer information to classify the emotion types into positive types, neutral types and negative types, and calculate the emotion type values within a preset range.
  • the emotional strength determining unit 320 may analyze the answer information and calculate each type of strength as an emotional strength value within a preset range.
  • the body motion correction calculation unit 400 may correct the motion motion of the body of the educational robot according to the emotion and the degree of emotion calculated by the emotion correction determination unit 300 .
  • the body motion correction calculation unit 400 may calculate a body motion correction value according to the emotion type value and the emotion intensity value calculated by the emotion correction determination unit 300 .
  • the body motion correction calculation unit 400 increases the correction value to a (+) value if the emotion type is a positive type, a correction value is not given if the emotion type is a neutral type, and the correction value is (-) if the emotion type is a negative type value can be increased.
  • the body motion correction calculator 400 may have a measured value within the range of a value when there is no emotional intensity and a value when the emotional intensity is the strongest.
  • the emotional intensity may have a value within the range of 0 to 1, with 0 being the absence of emotional intensity and 1 being the strongest emotional intensity.
  • the driving unit 500 may drive the body by applying the correction value calculated by the body motion correction operation unit 400 .
  • the movement of the body may be exemplified by driving the facial expressions and gestures of the educational robot.
  • 'eyebrows' can be selected as a part that can easily express the emotions of the educational robot through facial expressions.
  • the 'arm' can be selected as a part that easily expresses the emotions of the educational robot through gestures.
  • 5 and 6 are an embodiment showing that the eyebrows of the educational robot are driven in the present invention.
  • the driving unit 500 keeps the eyebrows of the educational robot in a horizontal state, and the correction value is a (+) value as the positive type. If it increases, the inner end 12 of the eyebrow 11 of the educational robot is driven to rise, and when the correction value increases to a (-) value in the negative type, the outer end of the eyebrow 11 of the educational robot ( 12) can be driven correspondingly to rise.
  • the user 20 When the inner end 12 of the eyebrow 11 of the educational robot rises together with the positive type answer information, the user 20 will be able to further sympathize with the positive expression of the educational robot.
  • the user 20 When the inner end 12 of the eyebrow 11 of the educational robot is lowered together with the negative type answer information, the user 20 will be able to further sympathize with the negative expression of the educational robot.
  • the driving unit 500 when the emotional intensity value increases, the correction value in the positive type and the negative type further increases correspondingly, and the driving range of the inner end 12 of the eyebrow of the educational robot also corresponds to can increase
  • 7 to 9 are one embodiment showing that the arm of the educational robot is driven in the present invention.
  • the driving unit 500 in the emotion type value, when the correction value is not given as a neutral type, the arm 13 of the educational robot maintains a normal state, and the correction value is ( When it increases to a +) value, the arm 13 of the educational robot is driven to move in front of the body 14, and when the correction value increases to a (-) value in the negative type, the arm 13 of the educational robot is It can be driven correspondingly to move to the rear of the body 14 .
  • the present invention may be implemented as a method of correcting the motion of an educational robot.
  • Substantial contents of the above-described motion correction system are the same, and the categories of the invention are different.
  • 3 and 4 are flowcharts of a motion correction method of an educational robot according to the present invention.
  • the present invention has a server 30 and a database 40 including an answer information generation unit, a context correction determination unit, an emotion correction determination unit, and a body motion correction calculation unit, and motion correction of an educational robot 10 performed by a computer it's about how
  • the method for correcting the motion of the educational robot includes the step S1 in which the question information receiving unit 100 receives the question information of the user 20; S2 step of generating, by the answer information generating unit 200, answer information for the question information received from the question information receiving unit 100; S3 step in which the emotion correction determining unit 300 determines the emotion and strength of the answer information calculated by the answer information generating unit 200 according to a preset criterion; S4 step of calculating, by the body motion correction calculation unit 400, a motion correction value of at least a part of the body of the educational robot according to the emotion type value and the emotion intensity value calculated by the emotion correction determination unit 300; and a step S5 in which the driving unit 500 drives the body by applying the correction value calculated by the body motion correction operation unit 400 .
  • the emotion correction determination unit 300 of step S3 may be provided with an emotion type determination unit 310 that digitizes the type of emotion and an emotion strength determination unit 320 that quantifies the intensity of the emotion.
  • the emotion type determination unit 310 of step S3 may analyze the answer information to classify the emotion types into positive types, neutral types and negative types, and calculate the emotion type values within a preset range.
  • the emotional strength determining unit 320 of step S3 may analyze the answer information and calculate the strength of each type as an emotional strength value within a preset range.
  • the body motion correction calculation unit 400 in step S4 may calculate a body motion correction value according to the emotion type value and the emotion intensity value calculated by the emotion correction determination unit 300 .
  • step S4 if the emotion type is a positive type, the correction value is increased to a (+) value, if the emotion type is a neutral type, the correction value is not given, and if the emotion type is a negative type, the correction value is It can be increased to a (-) value.
  • the body motion correction calculator 400 in step S4 may have a measured value within the range of a value when there is no emotional intensity and a value when the emotional intensity is the strongest.
  • the driving unit 500 in step S5 causes the eyebrows of the educational robot to maintain a horizontal state when the correction value is not given as the neutral type, and the correction value is set to a (+) value as the positive type. If it increases, the inner end 12 of the eyebrow 11 of the educational robot is driven to rise, and when the correction value increases to a (-) value in the negative type, the outer end of the eyebrow 11 of the educational robot ( 12) can be driven correspondingly to rise.
  • step S5 when the emotional intensity value increases, the correction value in the positive type and negative type further increases correspondingly, and the driving range of the inner end 12 of the eyebrow of the educational robot also corresponds to can increase
  • the driving unit 500 in step S5 causes the arm 13 of the educational robot to maintain a normal state when the correction value is not given to the neutral type, and the correction value is (+) to the positive type ), the arm 13 of the educational robot is driven to move to the front of the body 14, and when the correction value increases to a (-) value in the negative type, the arm 13 of the educational robot moves to the front of the body (14) It can be driven to move backwards.
  • the present invention may be implemented as a computer program. Specifically, in combination with hardware, it may be implemented as a computer program stored in a computer-readable recording medium in order to execute the method for correcting the motion of the educational robot according to the present invention by a computer.
  • a correction value can be added to [expression motor, body motor, led driving] by referring to the relevant driving correction information in the database.
  • the robot answers 'A2'.
  • the context keyword can be referred to.
  • the context keyword is “information transfer”
  • the overall conversational atmosphere is information transfer-oriented atmosphere.
  • the preset information set in advance eg, "information transfer”: [number of gestures +1]
  • the number of gestures to shake the arm was originally 2 times, but by adding 1, the trajectory can be modified by moving the arm up and down for a total of 3 times.
  • the level of emotion [negative] -1 to 1 [positive] in the dialogue that the robot will talk about can be corrected, and the correction value can be added to [expression motor, body motor, led drive].
  • the emotion determination result may be performed in two ways: an emotion type and an emotion intensity.
  • the angle of the eyebrows in the expression will be reduced by 3 (maximum correction)*-07, 21 degrees. Therefore, the robot's eyebrows will go down and the robot's expression will give an angry feeling.
  • the trajectory of the face motor can be calculated. By checking the trajectory of the face motor, if it is in a specific section, a correction value can be added to [body motor, led driving].
  • the blue color gives a sense of calm and reduces the total trajectory value by 10%.
  • FIG. 10 is a diagram illustrating a computing device according to the present invention.
  • the computing device TN100 may include at least one processor TN110 , a transceiver device TN120 , and a memory TN130 .
  • the computing device TN100 may further include a storage device TN140 , an input interface device TN150 , an output interface device TN160 , and the like.
  • Components included in the computing device TN100 may be connected by a bus TN170 to communicate with each other.
  • the processor TN110 may execute a program command stored in at least one of the memory TN130 and the storage device TN140.
  • the processor TN110 may mean a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor on which methods according to an embodiment of the present invention are performed.
  • the processor TN110 may be configured to implement procedures, functions, and methods described in connection with an embodiment of the present invention.
  • the processor TN110 may control each component of the computing device TN100.
  • Each of the memory TN130 and the storage device TN140 may store various information related to the operation of the processor TN110.
  • Each of the memory TN130 and the storage device TN140 may be configured as at least one of a volatile storage medium and a nonvolatile storage medium.
  • the memory TN130 may include at least one of a read only memory (ROM) and a random access memory (RAM).
  • the transceiver TN120 may transmit or receive a wired signal or a wireless signal.
  • the transceiver TN120 may be connected to a network to perform communication.
  • the methods according to the embodiment of the present invention described above may be implemented in the form of a program readable by various computer means and recorded in a computer readable recording medium.
  • the recording medium may include a program command, a data file, a data structure, etc. alone or in combination.
  • the program instructions recorded on the recording medium may be specially designed and configured for the present invention, or may be known and available to those skilled in the art of computer software.
  • the recording medium includes magnetic media such as hard disks, floppy disks and magnetic tapes, optical recording media such as CDROMs and DVDs, and magneto-optical media such as floppy disks. optical media), and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like.
  • Examples of program instructions may include high-level languages that can be executed by a computer using an interpreter or the like as well as machine language such as generated by a compiler.
  • Such hardware devices may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

Landscapes

  • Engineering & Computer Science (AREA)
  • Robotics (AREA)
  • Mechanical Engineering (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Educational Technology (AREA)
  • Educational Administration (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Automation & Control Theory (AREA)
  • Human Computer Interaction (AREA)
  • Manipulator (AREA)
  • Toys (AREA)

Abstract

The present invention relates to an action correction system for an education robot (10), having a server (30), which comprises an answer information generation unit, a context correction determination unit, an emotion correction determination unit, and a body action correction calculation unit, and a database (40), and being performed by a computing means. The action correction system for an education robot comprises: a question information reception unit (100) for receiving question information about a user (20); the answer information generation unit (200) for generating answer information to the question information received by the question information reception unit (100); the emotion correction determination unit (300) for determining, according to preset standards, the emotion and strength of the answer information calculated by the answer information generation unit (200); the body action correction calculation unit (400) for correcting the motion of a body action of the education robot according to the emotion and the degree of emotion calculated by the emotion correction determination unit (300); and a driving unit (500) for driving a body by applying a correction value calculated by the body action correction calculation unit (400).

Description

교육용 로봇의 동작보정 시스템 및 동작보정 방법Educational robot motion compensation system and motion compensation method
본 발명은 교육용 로봇의 동작보정 시스템 및 동작보정 방법에 관한 것이다. 구체적으로 감정보정을 통해 교육용 로봇의 신체 동작을 보정하는 기술에 관한 것이다.The present invention relates to a motion correction system and a motion correction method for an educational robot. Specifically, it relates to a technology for correcting the body motion of an educational robot through emotion correction.
종래의 교육용 로봇은 미리 정해진 대화문에 따라서만 대화가 가능한 경우가 일반적이었다.In the conventional educational robot, it is common that a conversation is possible only according to a predetermined dialogue text.
이에, 자연어 처리(natural language processing) 방법에 의해 미리 정해진 대화문이 아니라, 미지정된 대화문이 생성되는 방법이 교육용 로봇에 도입되는 추세에 있다.Accordingly, a method of generating an unspecified dialogue rather than a predetermined dialogue by a natural language processing method is being introduced into educational robots.
자연어(natural language)란 우리가 일상 생활에서 사용하는 언어를 의미한다. 자연어 처리(natural language processing)란 이러한 자연어의 의미를 분석하여 컴퓨터가 처리할 수 있도록 하는 것을 의미한다.Natural language refers to the language we use in our daily life. Natural language processing means analyzing the meaning of such natural language so that a computer can process it.
자연어 처리는 음성 인식, 내용 요약, 번역, 사용자의 감성 분석, 텍스트 분류 작업(스팸 메일 분류, 뉴스 기사 카테고리 분류), 질의 응답 시스템, 챗봇과 같은 곳에서 사용되는 분야이다.Natural language processing is a field used in speech recognition, content summarization, translation, user sentiment analysis, text classification tasks (spam mail classification, news article category classification), question and answer systems, and chatbots.
하지만, 자연어 처리에 의해 대화내용은 감정이 표현된 문장이 생성되는 장점이 있었지만, 감정과 무관하게 대화만 주고 받는 형태로 운용되었기에, 사용자 특히 어린 학생 사용자의 경우 교육용 로봇과 감정이입이 되지 못하여 교육 효과가 저하되는 문제점이 있었다.However, although natural language processing had the advantage of generating sentences expressing emotions, it was operated in the form of exchanging only conversations regardless of emotions. There was a problem that the effect was lowered.
본 발명에 따른 교육용 로봇의 동작보정 시스템 및 동작보정 방법은 다음과 같은 해결과제를 가진다.A motion correction system and a motion correction method of an educational robot according to the present invention have the following problems.
첫째, 당해 문장과 직전 문장의 관계를 고려하여, 교육용 로봇의 답변을 생성하고자 한다.First, in consideration of the relationship between the sentence and the previous sentence, an answer of the educational robot is to be generated.
둘째, 교육용 로봇이 답변을 할 때, 교육용 로봇의 표정과 신체 변화를 통해 사용자와 교감을 이루고자 한다.Second, when the educational robot gives an answer, it is intended to achieve sympathy with the user through the facial expression and body changes of the educational robot.
본 발명의 해결과제는 이상에서 언급한 것들에 한정되지 않으며, 언급되지 아니한 다른 해결과제들은 아래의 기재로부터 당업자에게 명확하게 이해될 수 있을 것이다. The problems to be solved of the present invention are not limited to those mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the following description.
본 발명은 답변정보 생성부, 문맥보정 판정부, 감정보정 판정부 및 신체동작 보정연산부를 포함하는 서버 및 데이터베이스를 가지며, 컴퓨팅 수단에 의해 수행되는 교육용 로봇의 동작보정 시스템으로서, 사용자의 질문정보를 수신하는 질문정보 수신부; 상기 질문정보 수신부에서 수신된 질문정보에 대한 답변정보를 생성하는 답변정보 생성부; 기 설정된 기준에 따라, 상기 답변정보 생성부에서 산출된 답변정보의 감정 및 강도를 판정하는 감정보정 판정부; 상기 감정보정 판정부에서 산출된 감정 및 감정의 정도에 따라, 교육용 로봇의 신체의 동작 움직임을 보정하는 신체동작 보정연산부; 및 상기 신체동작 보정연산부에서 산출된 보정값을 적용하여 신체를 구동시키는 구동부를 포함할 수 있다.The present invention has a server and a database including an answer information generating unit, a context correction determining unit, an emotion correction determining unit and a body motion correction calculating unit, and is a motion correction system for an educational robot performed by a computing means. Question information receiving unit to receive; an answer information generating unit for generating answer information for the question information received from the question information receiving unit; an emotion correction determining unit for determining the emotion and strength of the answer information calculated by the answer information generating unit according to a preset criterion; a body motion correction calculation unit for correcting the body motion of the educational robot according to the emotion and the degree of emotion calculated by the emotion correction determining unit; and a driving unit for driving the body by applying the correction value calculated by the body motion correction operation unit.
본 발명에 있어서, 상기 답변정보 생성부는 당해 질문정보 이전에 수신된 질문정보로부터 문맥을 파악하여 당해 질문정보의 답변정보를 산출하는 문맥보정 판정부를 더 구비할 수 있다.In the present invention, the answer information generating unit may further include a context correction determining unit for calculating the answer information of the question information by grasping the context from the question information received before the question information.
본 발명에 있어서, 답변정보 생성부, 문맥보정 판정부 및 감정보정 판정부는 자연어처리 기법 또는 딥러닝 기법으로 정보를 분석할 수 있다.In the present invention, the answer information generating unit, the context correction determining unit and the emotion correction determining unit may analyze information using a natural language processing technique or a deep learning technique.
본 발명에 있어서, 상기 답변정보 생성부가 당해 질문정보 이전에 수신된 대화정보가 있다고 판정하면, 당해 질문정보 또는 답변정보를 문맥보정 판정부로 보내며, 상기 문맥보정 판정부가 당해 질문정보와 이전에 수신된 대화정보를 상기 데이터베이스에 저장된 문맥정보와 대비하여, 당해 질문정보와 이전 대화정보의 문맥이 기 설정된 문맥범위 내에 속한다고 판정한 때에는, 상기 문맥보정 판정부는 기 설정된 문백범위 내의 문맥 키워드를 포함하는 답변정보를 생성할 수 있다.In the present invention, when the answer information generating unit determines that there is dialogue information received before the question information, the question information or the answer information is sent to the context correction determining unit, and the context correction determining unit receiving the question information before Compare the conversation information with the context information stored in the database, and when it is determined that the context of the question information and the previous conversation information falls within a preset context range, the context correction determining unit includes a context keyword within a preset context range You can create response information.
본 발명에 있어서, 상기 답변정보 생성부가 당해 질문정보 이전에 수신된 대화정보가 없다고 판정한 때 또는 상기 답변정보 생성부가 당해 질문정보 이전에 수신된 대화정보가 있다고 판정하여 당해 질문정보를 문맥보정 판정부로 보내고, 상기 문맥보정 판정부가 당해 질문정보와 이전에 수신된 대화정보 및 상기 데이터베이스에 저장된 문백정보를 대비하여, 당해 질문정보와 이전 질문정보의 문맥이 기 설정된 문맥범위내에 속한다고 판정한 때에는, 상기 답변정보 생성부는 데이터베이스에서 당해 질문정보의 키워드가 속한 키워드 그룹의 키워드를 포함하는 답변정보를 생성할 수 있다.In the present invention, when the answer information generating unit determines that there is no dialogue information received before the question information, or when the answer information generation unit determines that there is dialogue information received before the question information, the question information is converted into a context correction plate When it is sent to the government and the context correction determining unit compares the question information with the previously received conversation information and the textual information stored in the database, and determines that the context of the question information and the previous question information falls within the preset context range , the answer information generating unit may generate answer information including a keyword of a keyword group to which the keyword of the corresponding question information belongs in the database.
본 발명에 있어서, 상기 감정보정 판정부는 감정의 유형을 수치화하는 감정유형 판정부 및 감정의 강도를 수치화하는 감정강도 판정부로 구비될 수 있다.In the present invention, the emotion correction determination unit may be provided as an emotion type determination unit for quantifying the type of emotion and an emotion strength determination unit for quantifying the intensity of the emotion.
본 발명에 있어서, 상기 감정유형 판정부는 상기 답변정보를 분석하여 감정유형을 긍정유형, 중립유형 및 부정유형으로 구분하고, 기 설정된 범위내의 감정유형값으로 산출할 수 있다.In the present invention, the emotion type determination unit may analyze the answer information to classify the emotion types into positive types, neutral types and negative types, and calculate the emotion type values within a preset range.
본 발명에 있어서, 상기 감정강도 판정부는 상기 답변정보를 분석하여 상기 각 유형의 강도를 기 설정된 범위내의 감정강도값으로 산출할 수 있다.In the present invention, the emotional strength determining unit may analyze the answer information to calculate the intensity of each type as an emotional intensity value within a preset range.
본 발명에 있어서, 상기 신체동작 보정연산부는 감정보정 판정부가 산출한 감정유형값 및 감정강도값에 따른 신체동작 보정값을 산출할 수 있다.In the present invention, the body motion correction calculation unit may calculate a body motion correction value according to the emotion type value and the emotion intensity value calculated by the emotion correction determining unit.
본 발명에 있어서, 상기 신체동작 보정연산부는 감정유형이 긍정유형이면 보정값은 (+)값으로 증가되고, 감정유형이 중립유형이면 보정값은 미부여되고, 감정유형이 부정유형이면 보정값은 (-)값으로 증가될 수 있다.In the present invention, the body motion correction calculation unit increases the correction value to a (+) value if the emotion type is a positive type, and if the emotion type is a neutral type, the correction value is not given, and if the emotion type is a negative type, the correction value is It can be increased to a (-) value.
본 발명에 있어서, 상기 신체동작 보정연산부는 감정강도가 없을때의 값과 감정강도가 가장 강할때의 값의 범위 내에서 측정값을 가질 수 있다.In the present invention, the body motion correction calculator may have a measured value within the range of a value when there is no emotional intensity and a value when the emotional intensity is the strongest.
본 발명에 있어서, 상기 구동부는 상기 감정유형값에 있어서, 상기 중립유형으로 보정값이 미부여되면, 교육용 로봇의 눈썹이 수평상태를 유지하도록 하며, 상기 긍정유형으로 보정값이 (+)값으로 증가하면, 교육용 로봇의 눈썹의 안쪽 단부가 상승하도록 대응 구동시키며, 상기 부정유형으로 보정값이 (-)값으로 증가하면, 교육용 로봇의 눈썹의 바깥쪽 단부가 상승하도록 대응 구동시킬 수 있다.In the present invention, in the emotion type value, when the correction value is not given as the neutral type, the driving unit keeps the eyebrows of the educational robot in a horizontal state, and the correction value is a (+) value as the positive type. If it increases, the inner end of the eyebrow of the educational robot is driven correspondingly to rise, and when the correction value increases to a (-) value in the negative type, the outer end of the eyebrow of the educational robot can be driven correspondingly to rise.
본 발명에 있어서, 상기 구동부는 상기 감정강도값이 증가하면, 긍정유형 및 부정유형에서의 상기 보정값이 대응하여 더욱 증가하며, 상기 교육용 로봇의 눈썹의 안쪽 단부의 구동 범위도 대응하여 증가할 수 있다.In the present invention, when the emotional intensity value of the driving unit increases, the correction value in the positive type and the negative type further increases correspondingly, and the driving range of the inner end of the eyebrow of the educational robot can also be increased correspondingly. have.
본 발명에 있어서, 상기 구동부는 상기 감정유형값에 있어서, 상기 중립유형으로 보정값이 미부여되면, 교육용 로봇의 팔이 일반 상태를 유지하도록 하며, 상기 긍정유형으로 보정값이 (+)값으로 증가하면, 교육용 로봇의 팔이 몸체 앞쪽으로 이동하도록 대응 구동시키며, 상기 부정유형으로 보정값이 (-)값으로 증가하면, 교육용 로봇의 팔이 몸체 뒷쪽으로 이동하도록 대응 구동시킬 수 있다.In the present invention, in the emotion type value, when the correction value is not given as the neutral type, the driving unit keeps the arm of the educational robot in a normal state, and the correction value is a (+) value as the positive type. If it increases, the arm of the educational robot is driven to move in front of the body, and when the correction value increases to a (-) value in the negative type, the arm of the educational robot can be driven to move toward the rear of the body.
본 발명은 답변정보 생성부, 문맥보정 판정부, 감정보정 판정부 및 신체동작 보정연산부를 포함하는 서버 및 데이터베이스를 가지며, 컴퓨터에 의해 수행되는 교육용 로봇의 동작보정 방법으로서, 질문정보 수신부가 사용자의 질문정보를 수신하는 S1 단계; 답변정보 생성부가 상기 질문정보 수신부에서 수신된 질문정보에 대한 답변정보를 생성하는 S2 단계; 감정보정 판정부가 기 설정된 기준에 따라, 상기 답변정보 생성부에서 산출된 답변정보의 감정 및 강도를 판정하는 S3 단계; 신체동작 보정연산부가 상기 감정보정 판정부에서 산출된 감정유형값과 감정강도값에 따라, 교육용 로봇의 신체의 적어도 일부의 동작보정값을 계산하는 S4 단계; 및 구동부가 상기 신체동작 보정연산부에서 산출된 보정값을 적용하여 신체를 구동시키는 S5 단계를 포함할 수 있다.The present invention has a server and a database including an answer information generating unit, a context correction determining unit, an emotion correction determining unit and a body motion correction calculation unit, and is a method for correcting a motion of an educational robot performed by a computer, wherein the question information receiving unit is the user's Step S1 of receiving question information; S2 step of generating, by the answer information generating unit, answer information for the question information received from the question information receiving unit; S3 step of determining the emotion and strength of the answer information calculated by the answer information generating unit according to the emotion correction determining unit preset criteria; S4 step of calculating, by the body motion correction calculation unit, a motion correction value of at least a part of the body of the educational robot according to the emotion type value and the emotion intensity value calculated by the emotion correction determining unit; and step S5 in which the driving unit drives the body by applying the correction value calculated by the body motion correction operation unit.
본 발명에 있어서, S3 단계의 감정보정 판정부는 감정의 유형을 수치화하는 감정유형 판정부 및 감정의 강도를 수치화하는 감정강도 판정부로 구비될 수 있다.In the present invention, the emotion correction determination unit of step S3 may be provided as an emotion type determination unit for quantifying the type of emotion and an emotion strength determination unit for quantifying the intensity of the emotion.
본 발명에 있어서, S3 단계의 감정유형 판정부는 상기 답변정보를 분석하여 감정유형을 긍정유형, 중립유형 및 부정유형으로 구분하고, 기 설정된 범위내의 감정유형값으로 산출할 수 있다.In the present invention, the emotion type determining unit in step S3 may analyze the answer information to classify the emotion types into positive types, neutral types and negative types, and calculate the emotion type values within a preset range.
본 발명에 있어서, S3 단계의 감정강도 판정부는 상기 답변정보를 분석하여 상기 각 유형의 강도를 기 설정된 범위내의 감정강도값으로 산출할 수 있다.In the present invention, the emotional strength determination unit in step S3 may analyze the answer information and calculate the strength of each type as an emotional strength value within a preset range.
본 발명에 있어서, S4 단계의 신체동작 보정연산부는 감정보정 판정부가 산출한 감정유형값 및 감정강도값에 따른 신체동작 보정값을 산출할 수 있다.In the present invention, the body motion correction calculation unit in step S4 may calculate a body motion correction value according to the emotion type value and the emotion intensity value calculated by the emotion correction determining unit.
본 발명에 있어서, S4 단계의 신체동작 보정연산부는 감정유형이 긍정유형이면 보정값은 (+)값으로 증가되고, 감정유형이 중립유형이면 보정값은 미부여되고, 감정유형이 부정유형이면 보정값은 (-)값으로 증가될 수 있다.In the present invention, the body motion correction operation unit in step S4 increases the correction value to a (+) value if the emotion type is a positive type, and if the emotion type is a neutral type, the correction value is not given, and if the emotion type is a negative type, the correction value is corrected The value can be incremented to a (-) value.
본 발명에 있어서, S4 단계의 신체동작 보정연산부는 감정강도가 없을 때의 값과 감정강도가 가장 강할 때의 값의 범위 내에서 측정값을 가질 수 있다.In the present invention, the body motion correction calculator in step S4 may have a measured value within the range of a value when there is no emotional intensity and a value when the emotional intensity is the strongest.
본 발명에 있어서, S5 단계의 구동부는 상기 감정유형값에 있어서, 상기 중립유형으로 보정값이 미부여되면, 교육용 로봇의 눈썹이 수평상태를 유지하도록 하며, 상기 긍정유형으로 보정값이 (+)값으로 증가하면, 교육용 로봇의 눈썹의 안쪽 단부가 상승하도록 대응 구동시키며, 상기 부정유형으로 보정값이 (-)값으로 증가하면, 교육용 로봇의 눈썹의 바깥쪽 단부가 상승하도록 대응 구동시킬 수 있다.In the present invention, the driving unit in step S5 causes the eyebrows of the educational robot to maintain a horizontal state when the correction value is not given as the neutral type in the emotion type value, and the correction value is (+) as the positive type When the value increases, the inner end of the eyebrow of the educational robot is driven correspondingly to rise, and when the correction value increases to a (-) value in the negative type, the outer end of the eyebrow of the educational robot can be driven correspondingly to rise. .
본 발명에 있어서, S5 단계의 구동부는 상기 감정강도값이 증가하면, 긍정유형 및 부정유형에서의 상기 보정값이 대응하여 더욱 증가하며, 상기 교육용 로봇의 눈썹의 안쪽 단부의 구동 범위도 대응하여 증가할 수 있다.In the present invention, when the driving unit of step S5 increases the emotional intensity value, the correction value in the positive type and negative type increases correspondingly, and the driving range of the inner end of the eyebrow of the educational robot is also increased correspondingly. can do.
본 발명에 있어서, S5 단계의 구동부는 상기 감정유형값에 있어서, 상기 중립유형으로 보정값이 미부여되면, 교육용 로봇의 팔이 일반 상태를 유지하도록 하며, 상기 긍정유형으로 보정값이 (+)값으로 증가하면, 교육용 로봇의 팔이 몸체 앞쪽으로 이동하도록 대응 구동시키며, 상기 부정유형으로 보정값이 (-)값으로 증가하면, 교육용 로봇의 팔이 몸체 뒷쪽으로 이동하도록 대응 구동시킬 수 있다.In the present invention, when the correction value is not given to the neutral type in the emotion type value, the driving unit in step S5 causes the arm of the educational robot to maintain a normal state, and the correction value is (+) in the positive type When the value increases, the arm of the educational robot is driven to move in front of the body, and when the correction value increases to a (-) value in the negative type, the arm of the educational robot can be driven to move toward the rear of the body.
본 발명은 하드웨어와 결합되어, 청구항 15에 따른 교육용 로봇의 동작보정 방법을 컴퓨터에 의해 실행시키기 위하여 컴퓨터가 판독 가능한 기록매체에 저장된 컴퓨터 프로그램으로 구현될 수 있다.The present invention may be implemented as a computer program stored in a computer-readable recording medium in order to execute the method for correcting the motion of the educational robot according to claim 15 by a computer in combination with hardware.
본 발명에 따른 교육용 로봇의 동작보정 시스템 및 동작보정 방법은 다음과 같은 효과를 가진다.The motion compensation system and motion compensation method of the educational robot according to the present invention have the following effects.
첫째, 자연어 처리 및 문맥 보정을 통해, 당해 문장과 직전 문장의 문맥을 고려한 교육용 로봇의 답변을 생성하는 효과가 있다.First, through natural language processing and context correction, there is an effect of generating the answer of the educational robot in consideration of the context of the sentence and the previous sentence.
둘째, 교육용 로봇이 답변을 할 때, 교육용 로봇의 표정과 신체의 동작 보정을 하여, 사용자와 언어적 교감 및 제스처 등을 통한 비언어적 교감을 이루고자 한다.Second, when the educational robot responds, it is intended to achieve non-verbal communication with the user through verbal communication and gestures by correcting the educational robot's facial expression and body motion.
본 발명의 효과는 이상에서 언급된 것들에 한정되지 않으며, 언급되지 아니한 다른 효과들은 아래의 기재로부터 당업자에게 명확하게 이해될 수 있을 것이다.Effects of the present invention are not limited to those mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.
도 1은 본 발명에 따른 교육용 로봇의 동작보정 시스템의 개요도이다.1 is a schematic diagram of a motion correction system for an educational robot according to the present invention.
도 2는 본 발명에 따른 본 발명에 따른 교육용 로봇의 동작보정 시스템의 주요 구성부이다.2 is a main configuration of the motion correction system of the educational robot according to the present invention according to the present invention.
도 3 및 도 4는 본 발명에 따른 교육용 로봇의 동작보정 방법의 순서도이다.3 and 4 are flowcharts of a motion correction method of an educational robot according to the present invention.
도 5 및 도 6은 본 발명에서 교육용 로봇의 눈썹이 구동되는 것을 나타내는 일 실시예이다.5 and 6 are an embodiment showing that the eyebrows of the educational robot are driven in the present invention.
도 7 내지 도 9는 본 발명에서 교육용 로봇의 팔이 구동되는 것을 나타내는 일 실시예이다.7 to 9 are one embodiment showing that the arm of the educational robot is driven in the present invention.
도 10은 본 발명에 따른 컴퓨팅 장치를 나타내는 도면이다.10 is a diagram illustrating a computing device according to the present invention.
10 : 교육용 로봇 11 : 눈썹10: educational robot 11: eyebrows
12 : 눈썹 안쪽 단부 13 : 팔12: inner end of eyebrow 13: arm
14 : 몸체14: body
20 : 사용자 30 : 서버20: user 30: server
40 : 데이터베이스40: database
100 : 질문정보 수신부 200 : 답변정보 생성부100: question information receiving unit 200: answer information generating unit
210 : 문맥보정 판정부 300 : 감정보정 판정부210: context correction determination unit 300: emotion correction determination unit
310 : 감정유형 판정부 320 : 감정강도 판정부310: emotion type determination unit 320: emotion strength determination unit
400 : 신체동작 보정연산부 500 : 구동부400: body motion correction calculation unit 500: drive unit
본 발명은 답변정보 생성부, 문맥보정 판정부, 감정보정 판정부 및 신체동작 보정연산부를 포함하는 서버 및 데이터베이스를 가지며, 컴퓨팅 수단에 의해 수행되는 교육용 로봇의 동작보정 시스템으로서, 사용자의 질문정보를 수신하는 질문정보 수신부; 상기 질문정보 수신부에서 수신된 질문정보에 대한 답변정보를 생성하는 답변정보 생성부; 기 설정된 기준에 따라, 상기 답변정보 생성부에서 산출된 답변정보의 감정 및 강도를 판정하는 감정보정 판정부; 상기 감정보정 판정부에서 산출된 감정 및 감정의 정도에 따라, 교육용 로봇의 신체의 동작 움직임을 보정하는 신체동작 보정연산부; 및 상기 신체동작 보정연산부에서 산출된 보정값을 적용하여 신체를 구동시키는 구동부를 포함할 수 있다.The present invention has a server and a database including an answer information generating unit, a context correction determining unit, an emotion correction determining unit and a body motion correction calculating unit, and is a motion correction system for an educational robot performed by a computing means. Question information receiving unit to receive; an answer information generating unit for generating answer information for the question information received from the question information receiving unit; an emotion correction determining unit for determining the emotion and strength of the answer information calculated by the answer information generating unit according to a preset criterion; a body motion correction calculation unit for correcting the body motion of the educational robot according to the emotion and the degree of emotion calculated by the emotion correction determining unit; and a driving unit for driving the body by applying the correction value calculated by the body motion correction operation unit.
이하, 첨부한 도면을 참조하여, 본 발명이 속하는 기술분야에서 통상의 지식을 가진 자가 용이하게 실시할 수 있도록 본 발명의 실시예를 설명한다. 본 발명이 속하는 기술분야에서 통상의 지식을 가진 자가 용이하게 이해할 수 있는 바와 같이, 후술하는 실시예는 본 발명의 개념과 범위를 벗어나지 않는 한도 내에서 다양한 형태로 변형될 수 있다. 가능한 한 동일하거나 유사한 부분은 도면에서 동일한 도면부호를 사용하여 나타낸다.Hereinafter, with reference to the accompanying drawings, embodiments of the present invention will be described so that those of ordinary skill in the art can easily carry out the present invention. As can be easily understood by those of ordinary skill in the art to which the present invention pertains, the embodiments described below may be modified in various forms without departing from the concept and scope of the present invention. Wherever possible, identical or similar parts are denoted by the same reference numerals in the drawings.
본 명세서에서 사용되는 전문용어는 단지 특정 실시예를 언급하기 위한 것이며, 본 발명을 한정하는 것을 의도하지는 않는다. 여기서 사용되는 단수 형태들은 문구들이 이와 명백히 반대의 의미를 나타내지 않는 한 복수 형태들도 포함한다.The terminology used herein is for the purpose of referring to specific embodiments only, and is not intended to limit the present invention. As used herein, the singular forms also include the plural forms unless the phrases clearly indicate the opposite.
본 명세서에서 사용되는 "포함하는"의 의미는 특정 특성, 영역, 정수, 단계, 동작, 요소 및/또는 성분을 구체화하며, 다른 특정 특성, 영역, 정수, 단계, 동작, 요소, 성분 및/또는 군의 존재나 부가를 제외시키는 것은 아니다.The meaning of "comprising," as used herein, specifies a particular characteristic, region, integer, step, operation, element and/or component, and other specific characteristic, region, integer, step, operation, element, component and/or It does not exclude the presence or addition of groups.
본 명세서에서 사용되는 기술용어 및 과학용어를 포함하는 모든 용어들은 본 발명이 속하는 기술분야에서 통상의 지식을 가진 자가 일반적으로 이해하는 의미와 동일한 의미를 가진다. 사전에 정의된 용어들은 관련기술문헌과 현재 개시된 내용에 부합하는 의미를 가지는 것으로 추가 해석되고, 정의되지 않는 한 이상적이거나 매우 공식적인 의미로 해석되지 않는다.All terms including technical terms and scientific terms used in this specification have the same meaning as those commonly understood by those of ordinary skill in the art to which the present invention belongs. Terms defined in the dictionary are additionally interpreted as having a meaning consistent with the related art literature and the presently disclosed content, and unless defined, are not interpreted in an ideal or very formal meaning.
이하에서는 도면을 참고하여 본 발명을 설명하고자 한다. 참고로, 도면은 본 발명의 특징을 설명하기 위하여, 일부 과장되게 표현될 수도 있다. 이 경우, 본 명세서의 전 취지에 비추어 해석되는 것이 바람직하다.Hereinafter, the present invention will be described with reference to the drawings. For reference, the drawings may be partially exaggerated in order to explain the features of the present invention. In this case, it is preferable to be interpreted in light of the whole meaning of this specification.
본 발명은 답변정보 생성부, 문맥보정 판정부, 감정보정 판정부 및 신체동작 보정연산부를 포함하는 서버(30) 및 데이터베이스(40)를 가지며, 컴퓨팅 수단에 의해 수행되는 교육용 로봇(10)의 동작보정 시스템에 관한 것이다.The present invention has a server 30 and a database 40 including an answer information generation unit, a context correction determination unit, an emotion correction determination unit and a body motion correction operation unit, and the operation of the educational robot 10 performed by a computing means It is about a calibration system.
도 1은 본 발명에 따른 교육용 로봇의 동작보정 시스템의 개요도이다. 1 is a schematic diagram of a motion correction system for an educational robot according to the present invention.
도 1에 도시된 바와 같이, 교육용 로봇(10)과 사용자(20)는 대화를 할 수 있으며, 사용자(2)의 대화정보를 수신한 교육용 로봇(10)은 서버(30)를 통해 자연어 처리 알고리즘 등이 수행될 수 있다.As shown in FIG. 1 , the educational robot 10 and the user 20 can have a conversation, and the educational robot 10 that has received the user 2 's conversation information is a natural language processing algorithm through the server 30 . etc. may be performed.
여기서, 서버(30)는 컴퓨팅 수단으로 연산기능을 수행하며, 무선 네트워크 또는 유선 네트워크로 교육용 로봇과 교신할 수 있다. 서버(30)는 교육용 로봇(10) 내부에 배치될 수도 있고, 별도로 클라우드 서버 형태 등으로 독립된 공간에 배치될 수도 있다. 데이터베이스(40)는 서버(30)의 연산에 필요한 정보를 저장하며, 또한 새로 생성되는 정보도 저장하게 된다. 데이터베이스(40)는 서버(30)와 교신한다. 이러한 각 행위는 컴퓨터에 의해 수행될 수 있다.Here, the server 30 performs a calculation function as a computing means, and may communicate with the educational robot through a wireless network or a wired network. The server 30 may be disposed inside the educational robot 10 or may be separately disposed in an independent space in the form of a cloud server or the like. The database 40 stores information necessary for the operation of the server 30 and also stores newly generated information. The database 40 communicates with the server 30 . Each of these actions may be performed by a computer.
도 2는 본 발명에 따른 본 발명에 따른 교육용 로봇의 동작보정 시스템의 주요 구성부이다.2 is a main configuration of the motion correction system of the educational robot according to the present invention according to the present invention.
본 발명에 따른 교육용 로봇의 동작보정 시스템은 사용자(20)의 질문정보를 수신하는 질문정보 수신부(100); 상기 질문정보 수신부(100)에서 수신된 질문정보에 대한 답변정보를 생성하는 답변정보 생성부(200); 기 설정된 기준에 따라, 상기 답변정보 생성부(200)에서 산출된 답변정보의 감정 및 강도를 판정하는 감정보정 판정부(300); 상기 감정보정 판정부(300)에서 산출된 감정유형값과 감정강도값에 따라, 교육용 로봇의 신체의 적어도 일부의 동작보정값을 계산하는 신체동작 보정연산부(400); 및 상기 신체동작 보정연산부(400)에서 산출된 보정값을 적용하여 신체를 구동시키는 구동부(500)를 포함한다.A motion correction system for an educational robot according to the present invention includes: a question information receiving unit 100 for receiving question information from a user 20; an answer information generating unit 200 for generating answer information for the question information received from the question information receiving unit 100; an emotion correction determining unit 300 for determining the emotion and strength of the answer information calculated by the answer information generating unit 200 according to a preset criterion; a body motion correction calculation unit 400 for calculating motion correction values of at least a part of the body of the educational robot according to the emotion type value and the emotion intensity value calculated by the emotion correction determining unit 300; and a driving unit 500 for driving the body by applying the correction value calculated by the body motion correction operation unit 400 .
본 발명에 따른 답변정보 생성부(200)는 당해 질문정보 이전에 수신된 질문정보로부터 문맥을 파악하여 당해 질문정보의 답변정보를 산출하는 문맥보정 판정부(210)를 구비할 수 있다.The answer information generating unit 200 according to the present invention may include a context correction determining unit 210 for calculating the answer information of the question information by grasping the context from the question information received before the question information.
본 발명에 있어서, 답변정보 생성부(200), 문맥보정 판정부(210) 및 감정보정 판정부(300)는 자연어처리 기법 및 딥러닝 기법으로 정보를 분석할 수 있다.In the present invention, the answer information generating unit 200 , the context correction determining unit 210 , and the emotion correction determining unit 300 may analyze information using a natural language processing technique and a deep learning technique.
이하에서는, 답변정보 생성부(200)에 관하여 설명하고자 한다.Hereinafter, the answer information generating unit 200 will be described.
본 발명에 따른 답변정보 생성부(200) 질문정보 수신부(100)에서 수신된 질문정보에 대한 답변정보를 생성할 수 있다.The answer information generating unit 200 according to the present invention may generate answer information for the question information received by the question information receiving unit 100 .
본 발명에 따른 답변정보 생성부(200)는 사용자(20)의 질문에 대하여 교육용 로봇(10)이 답변을 하는 것을 의미한다.The answer information generating unit 200 according to the present invention means that the educational robot 10 answers the question of the user 20 .
교육용 로봇(10)의 답변정보는 사용자(20)의 당해 질문이 최초인 경우와 당해 질문 전에 관련된 대화정보가 있는 경우로 구분될 수 있다. The answer information of the educational robot 10 may be divided into a case where the user 20 asks the question for the first time, and a case where there is conversational information related to the question before the question.
대화정보는 질문정보와 답변정보를 포함한다. 일반적으로 사용자가 먼저 질문을 하고 교육용 로봇이 답변을 하지만, 경우에 따라 교육용 로봇이 먼저 질문을 하는 실시예도 구현될 수 있을 것이다. The dialogue information includes question information and answer information. In general, the user asks a question first and the educational robot answers the question, but in some cases, an embodiment in which the educational robot asks a question first may be implemented.
본 발명에 있어서, 상기 답변정보 생성부(200)가 당해 질문정보 이전에 수신된 대화정보가 있다고 판정하면, 당해 질문정보 또는 답변정보를 문맥보정 판정부(210)로 보낼 수 있다.In the present invention, if the answer information generating unit 200 determines that there is conversation information received before the question information, the question information or the answer information may be sent to the context correction determining unit 210 .
이때, 상기 문맥보정 판정부(210)가 당해 질문정보와 이전에 수신된 대화정보를 상기 데이터베이스에 저장된 문맥정보와 대비하여, 당해 질문정보와 이전 대화정보의 문맥이 기 설정된 문맥범위 내에 속한다고 판정한 때에는, 상기 문맥보정 판정부(210)는 기 설정된 문백범위 내의 문맥 키워드를 포함하는 답변정보를 생성할 수 있다.In this case, the context correction determining unit 210 compares the question information and the previously received dialogue information with the context information stored in the database, and determines that the context of the question information and the previous dialogue information falls within a preset context range. At one time, the context correction determining unit 210 may generate answer information including a context keyword within a preset textual range.
본 발명에 있어서, 상기 답변정보 생성부(200)가 당해 질문정보 이전에 수신된 대화정보가 없다고 판정한 때 또는 상기 답변정보 생성부(200)가 당해 질문정보 이전에 수신된 대화정보가 있다고 판정하여 당해 질문정보를 문맥보정 판정부(210)로 보내고, 상기 문맥보정 판정부(210)가 당해 질문정보와 이전에 수신된 대화정보 및 상기 데이터베이스에 저장된 문백정보를 대비하여, 당해 질문정보와 이전 질문정보의 문맥이 기 설정된 문맥범위내에 속한다고 판정한 때에는, 상기 답변정보 생성부(200)는 데이터베이스에서 당해 질문정보의 키워드가 속한 키워드 그룹의 키워드를 포함하는 답변정보를 생성할 수 있다.In the present invention, when the answer information generating unit 200 determines that there is no conversation information received before the question information, or when the answer information generating unit 200 determines that there is conversation information received before the question information to send the question information to the context correction determining unit 210, and the context correction determining unit 210 compares the question information with the previously received conversation information and the textual information stored in the database, When it is determined that the context of the question information falls within a preset context range, the answer information generating unit 200 may generate answer information including a keyword of a keyword group to which the keyword of the question information belongs in the database.
이하에서는, 감정보정 판정부(300)에 관하여 설명하고자 한다.Hereinafter, the emotion correction determination unit 300 will be described.
본 발명에 따른 감정보정 판정부(300)는 기 설정된 기준에 따라, 상기 답변정보 생성부(200)에서 산출된 답변정보의 감정 및 강도를 판정할 수 있다.The emotion correction determining unit 300 according to the present invention may determine the emotion and strength of the answer information calculated by the answer information generating unit 200 according to a preset criterion.
이와 같이 판정되는 감정 및 강도는 후술할 교육용 로봇의 표정 내지 신체의 동작 보정에 활용될 수 있다. The emotion and intensity determined in this way may be utilized for correcting facial expressions or body motions of the educational robot, which will be described later.
전술한 답변정보 생성부(200)에서 생성된 답변만 제시하면, 사용자와의 감정교감이 부족하므로, 본 감정보정 판정부(300)에서 답변에 내포된 감정을 수치화하여, 이를 교육용 로봇의 표정 내지 신체의 동작 움직임에 반영하면, 사용자는 교육용 로봇과 더욱 교감이 가능할 것이다.If only the answer generated by the above-described answer information generating unit 200 is presented, emotional sympathy with the user is insufficient, so the emotion correction determination unit 300 digitizes the emotions contained in the answer, and uses it as the expression of the educational robot. If it is reflected in the movement of the body, the user will be able to interact more with the educational robot.
본 발명에 따른 감정보정 판정부(300)는 감정의 유형을 수치화하는 감정유형 판정부(310) 및 감정의 강도를 수치화하는 감정강도 판정부(320)로 구비될 수 있다.The emotion correction determination unit 300 according to the present invention may be provided with an emotion type determination unit 310 that digitizes the type of emotion and an emotion strength determination unit 320 that quantifies the intensity of the emotion.
감정유형 판정부(310)는 답변정보를 분석하여 감정유형을 긍정유형, 중립유형 및 부정유형으로 구분하고, 기 설정된 범위내의 감정유형값으로 산출할 수 있다.The emotion type determining unit 310 may analyze the answer information to classify the emotion types into positive types, neutral types and negative types, and calculate the emotion type values within a preset range.
감정강도 판정부(320)는 답변정보를 분석하여 상기 각 유형의 강도를 기 설정된 범위내의 감정강도값으로 산출할 수 있다.The emotional strength determining unit 320 may analyze the answer information and calculate each type of strength as an emotional strength value within a preset range.
이하에서는, 신체동작 보정연산부(400)에 관하여 설명하고자 한다.Hereinafter, the body motion correction operation unit 400 will be described.
본 발명에 따른 신체동작 보정연산부(400)는 감정보정 판정부(300)에서 산출된 감정 및 감정의 정도에 따라, 교육용 로봇의 신체의 동작 움직임을 보정할 수 있다.The body motion correction calculation unit 400 according to the present invention may correct the motion motion of the body of the educational robot according to the emotion and the degree of emotion calculated by the emotion correction determination unit 300 .
신체동작 보정연산부(400)는 감정보정 판정부(300)가 산출한 감정유형값 및 감정강도값에 따른 신체동작 보정값을 산출할 수 있다.The body motion correction calculation unit 400 may calculate a body motion correction value according to the emotion type value and the emotion intensity value calculated by the emotion correction determination unit 300 .
신체동작 보정연산부(400)는 감정유형이 긍정유형이면 보정값은 (+)값으로 증가되고, 감정유형이 중립유형이면 보정값은 미부여되고, 감정유형이 부정유형이면 보정값은 (-)값으로 증가될 수 있다.The body motion correction calculation unit 400 increases the correction value to a (+) value if the emotion type is a positive type, a correction value is not given if the emotion type is a neutral type, and the correction value is (-) if the emotion type is a negative type value can be increased.
신체동작 보정연산부(400)는 감정강도가 없을 때의 값과 감정강도가 가장 강할 때의 값의 범위 내에서 측정값을 가질 수 있다. 예를 들어, 감정강도가 없을때를 0점으로 하고, 감정강도가 가장 강할 때를 1점으로 하여, 감정강도는 0~1의 범위내의 값을 가질 수 있다.The body motion correction calculator 400 may have a measured value within the range of a value when there is no emotional intensity and a value when the emotional intensity is the strongest. For example, the emotional intensity may have a value within the range of 0 to 1, with 0 being the absence of emotional intensity and 1 being the strongest emotional intensity.
이하에서는, 신체동작 보정연산부(400)에 관하여 설명하고자 한다.Hereinafter, the body motion correction operation unit 400 will be described.
본 발명에 따른 구동부(500)는 신체동작 보정연산부(400)에서 산출된 보정값을 적용하여 신체를 구동시킬 수 있다.The driving unit 500 according to the present invention may drive the body by applying the correction value calculated by the body motion correction operation unit 400 .
본 발명에 있어서, 신체구동은 교육용 로봇의 표정과 몸짓을 구동하는 것이 예시될 수 있다. In the present invention, the movement of the body may be exemplified by driving the facial expressions and gestures of the educational robot.
본 발명에서는 얼굴 표정을 통해 교육용 로봇의 감정을 용이하게 표현할 수 있는 부분으로 '눈썹'을 선택할 수 있다. 또한, 몸짓을 통해 교육용 로봇의 감정을 용이하게 표현하는 부분으로 '팔'을 선택할 수 있다.In the present invention, 'eyebrows' can be selected as a part that can easily express the emotions of the educational robot through facial expressions. In addition, the 'arm' can be selected as a part that easily expresses the emotions of the educational robot through gestures.
도 5 및 도 6은 본 발명에서 교육용 로봇의 눈썹이 구동되는 것을 나타내는 일 실시예이다.5 and 6 are an embodiment showing that the eyebrows of the educational robot are driven in the present invention.
본 발명에 따른 구동부(500)는 감정유형값에 있어서, 상기 중립유형으로 보정값이 미부여되면, 교육용 로봇의 눈썹이 수평상태를 유지하도록 하며, 상기 긍정유형으로 보정값이 (+)값으로 증가하면, 교육용 로봇의 눈썹(11)의 안쪽 단부(12)가 상승하도록 대응 구동시키며, 상기 부정유형으로 보정값이 (-)값으로 증가하면, 교육용 로봇의 눈썹(11)의 바깥쪽 단부(12)가 상승하도록 대응 구동시킬 수 있다.In the emotion type value, if the correction value is not given to the neutral type, the driving unit 500 according to the present invention keeps the eyebrows of the educational robot in a horizontal state, and the correction value is a (+) value as the positive type. If it increases, the inner end 12 of the eyebrow 11 of the educational robot is driven to rise, and when the correction value increases to a (-) value in the negative type, the outer end of the eyebrow 11 of the educational robot ( 12) can be driven correspondingly to rise.
긍정유형인 답변정보와 함께, 교육용 로봇의 눈썹(11)의 안쪽 단부(12)가 상승하면, 사용자(20)는 교육용 로봇의 긍정적 표현을 더욱 교감할 수 있을 것이다.When the inner end 12 of the eyebrow 11 of the educational robot rises together with the positive type answer information, the user 20 will be able to further sympathize with the positive expression of the educational robot.
부정유형인 답변정보와 함께, 교육용 로봇의 눈썹(11)의 안쪽 단부(12)가 하강하면, 사용자(20)는 교육용 로봇의 부정적 표현을 더욱 교감할 수 있을 것이다.When the inner end 12 of the eyebrow 11 of the educational robot is lowered together with the negative type answer information, the user 20 will be able to further sympathize with the negative expression of the educational robot.
본 발명에 따른 구동부(500)는 감정강도값이 증가하면, 긍정유형 및 부정유형에서의 상기 보정값이 대응하여 더욱 증가하며, 상기 교육용 로봇의 눈썹의 안쪽 단부(12)의 구동 범위도 대응하여 증가할 수 있다.In the driving unit 500 according to the present invention, when the emotional intensity value increases, the correction value in the positive type and the negative type further increases correspondingly, and the driving range of the inner end 12 of the eyebrow of the educational robot also corresponds to can increase
예로, 긍정감정 또는 부정감정의 강도값이 다른 경우, 이를 눈썹의 구동범위에 반영하는 구성이다. 사용자는 이러한 구성을 통한 눈썹의 구동과 교육용 로봇의 답변을 함께 받게 되므로, 사용자의 교감은 더욱 증가될 수 있을 것이다.For example, when the intensity values of positive emotions or negative emotions are different, this is reflected in the driving range of the eyebrows. Since the user receives both the driving of the eyebrows and the answer of the educational robot through this configuration, the user's sympathy may be further increased.
도 7 내지 도 9는 본 발명에서 교육용 로봇의 팔이 구동되는 것을 나타내는 일 실시예이다.7 to 9 are one embodiment showing that the arm of the educational robot is driven in the present invention.
본 발명에 따른 상기 구동부(500)는 상기 감정유형값에 있어서, 중립유형으로 보정값이 미부여되면, 교육용 로봇의 팔(13)이 일반 상태를 유지하도록 하며, 상기 긍정유형으로 보정값이 (+)값으로 증가하면, 교육용 로봇의 팔(13)이 몸체(14) 앞쪽으로 이동하도록 대응 구동시키며, 상기 부정유형으로 보정값이 (-)값으로 증가하면, 교육용 로봇의 팔(13)이 몸체(14) 뒷쪽으로 이동하도록 대응 구동시킬 수 있다.The driving unit 500 according to the present invention, in the emotion type value, when the correction value is not given as a neutral type, the arm 13 of the educational robot maintains a normal state, and the correction value is ( When it increases to a +) value, the arm 13 of the educational robot is driven to move in front of the body 14, and when the correction value increases to a (-) value in the negative type, the arm 13 of the educational robot is It can be driven correspondingly to move to the rear of the body 14 .
한편, 본 발명은 교육용 로봇의 동작보정 방법으로 구현될 수도 있다. 전술한 동작보정 시스템과 실질적 내용은 동일하며, 발명의 카테고리가 상이한 것에 해당된다. On the other hand, the present invention may be implemented as a method of correcting the motion of an educational robot. Substantial contents of the above-described motion correction system are the same, and the categories of the invention are different.
이에, 발명의 내용은 공통되므로, 공통되는 설명은 생략하고, 동작보정 방법에 대한 주요 단계를 설명하고자 한다.Accordingly, since the contents of the invention are common, the common description will be omitted and the main steps of the motion correction method will be described.
도 3 및 도 4는 본 발명에 따른 교육용 로봇의 동작보정 방법의 순서도이다.3 and 4 are flowcharts of a motion correction method of an educational robot according to the present invention.
본 발명은 답변정보 생성부, 문맥보정 판정부, 감정보정 판정부 및 신체동작 보정연산부를 포함하는 서버(30) 및 데이터베이스(40)를 가지며, 컴퓨터에 의해 수행되는 교육용 로봇(10)의 동작보정 방법에 관한 것이다.The present invention has a server 30 and a database 40 including an answer information generation unit, a context correction determination unit, an emotion correction determination unit, and a body motion correction calculation unit, and motion correction of an educational robot 10 performed by a computer it's about how
본 발명에 따른 는 교육용 로봇의 동작보정 방법은 질문정보 수신부(100)가 사용자(20)의 질문정보를 수신하는 S1 단계; 답변정보 생성부(200)가 상기 질문정보 수신부(100)에서 수신된 질문정보에 대한 답변정보를 생성하는 S2 단계; 감정보정 판정부(300)가 기 설정된 기준에 따라, 상기 답변정보 생성부(200)에서 산출된 답변정보의 감정 및 강도를 판정하는 S3 단계; 신체동작 보정연산부(400)가 상기 감정보정 판정부(300)에서 산출된 감정유형값과 감정강도값에 따라, 교육용 로봇의 신체의 적어도 일부의 동작보정값을 계산하는 S4 단계; 및 구동부(500)가 상기 신체동작 보정연산부(400)에서 산출된 보정값을 적용하여 신체를 구동시키는 S5 단계를 포함한다.According to the present invention, the method for correcting the motion of the educational robot includes the step S1 in which the question information receiving unit 100 receives the question information of the user 20; S2 step of generating, by the answer information generating unit 200, answer information for the question information received from the question information receiving unit 100; S3 step in which the emotion correction determining unit 300 determines the emotion and strength of the answer information calculated by the answer information generating unit 200 according to a preset criterion; S4 step of calculating, by the body motion correction calculation unit 400, a motion correction value of at least a part of the body of the educational robot according to the emotion type value and the emotion intensity value calculated by the emotion correction determination unit 300; and a step S5 in which the driving unit 500 drives the body by applying the correction value calculated by the body motion correction operation unit 400 .
각 세부 단계의 구성은 동작보정 시스템과 공통된다. 주된 특징은 다음과 같다.The configuration of each detailed step is common with the motion compensation system. The main features are as follows.
S3 단계의 감정보정 판정부(300)는 감정의 유형을 수치화하는 감정유형 판정부(310) 및 감정의 강도를 수치화하는 감정강도 판정부(320)로 구비될 수 있다.The emotion correction determination unit 300 of step S3 may be provided with an emotion type determination unit 310 that digitizes the type of emotion and an emotion strength determination unit 320 that quantifies the intensity of the emotion.
S3 단계의 감정유형 판정부(310)는 상기 답변정보를 분석하여 감정유형을 긍정유형, 중립유형 및 부정유형으로 구분하고, 기 설정된 범위내의 감정유형값으로 산출할 수 있다.The emotion type determination unit 310 of step S3 may analyze the answer information to classify the emotion types into positive types, neutral types and negative types, and calculate the emotion type values within a preset range.
S3 단계의 감정강도 판정부(320)는 상기 답변정보를 분석하여 상기 각 유형의 강도를 기 설정된 범위내의 감정강도값으로 산출할 수 있다.The emotional strength determining unit 320 of step S3 may analyze the answer information and calculate the strength of each type as an emotional strength value within a preset range.
S4 단계의 신체동작 보정연산부(400)는 감정보정 판정부(300)가 산출한 감정유형값 및 감정강도값에 따른 신체동작 보정값을 산출할 수 있다.The body motion correction calculation unit 400 in step S4 may calculate a body motion correction value according to the emotion type value and the emotion intensity value calculated by the emotion correction determination unit 300 .
S4 단계의 신체동작 보정연산부(400)는 감정유형이 긍정유형이면 보정값은 (+)값으로 증가되고, 감정유형이 중립유형이면 보정값은 미부여되고, 감정유형이 부정유형이면 보정값은 (-)값으로 증가될 수 있다.In the body motion correction operation unit 400 of step S4, if the emotion type is a positive type, the correction value is increased to a (+) value, if the emotion type is a neutral type, the correction value is not given, and if the emotion type is a negative type, the correction value is It can be increased to a (-) value.
S4 단계의 신체동작 보정연산부(400)는 감정강도가 없을 때의 값과 감정강도가 가장 강할 때의 값의 범위 내에서 측정값을 가질 수 있다.The body motion correction calculator 400 in step S4 may have a measured value within the range of a value when there is no emotional intensity and a value when the emotional intensity is the strongest.
S5 단계의 구동부(500)는 상기 감정유형값에 있어서, 상기 중립유형으로 보정값이 미부여되면, 교육용 로봇의 눈썹이 수평상태를 유지하도록 하며, 상기 긍정유형으로 보정값이 (+)값으로 증가하면, 교육용 로봇의 눈썹(11)의 안쪽 단부(12)가 상승하도록 대응 구동시키며, 상기 부정유형으로 보정값이 (-)값으로 증가하면, 교육용 로봇의 눈썹(11)의 바깥쪽 단부(12)가 상승하도록 대응 구동시킬 수 있다.In the emotion type value, the driving unit 500 in step S5 causes the eyebrows of the educational robot to maintain a horizontal state when the correction value is not given as the neutral type, and the correction value is set to a (+) value as the positive type. If it increases, the inner end 12 of the eyebrow 11 of the educational robot is driven to rise, and when the correction value increases to a (-) value in the negative type, the outer end of the eyebrow 11 of the educational robot ( 12) can be driven correspondingly to rise.
S5 단계의 구동부(500)는 상기 감정강도값이 증가하면, 긍정유형 및 부정유형에서의 상기 보정값이 대응하여 더욱 증가하며, 상기 교육용 로봇의 눈썹의 안쪽 단부(12)의 구동 범위도 대응하여 증가할 수 있다.In the driving unit 500 of step S5, when the emotional intensity value increases, the correction value in the positive type and negative type further increases correspondingly, and the driving range of the inner end 12 of the eyebrow of the educational robot also corresponds to can increase
S5 단계의 구동부(500)는 상기 감정유형값에 있어서, 상기 중립유형으로 보정값이 미부여되면, 교육용 로봇의 팔(13)이 일반 상태를 유지하도록 하며, 상기 긍정유형으로 보정값이 (+)값으로 증가하면, 교육용 로봇의 팔(13)이 몸체(14) 앞쪽으로 이동하도록 대응 구동시키며, 상기 부정유형으로 보정값이 (-)값으로 증가하면, 교육용 로봇의 팔(13)이 몸체(14) 뒷쪽으로 이동하도록 대응 구동시킬 수 있다.In the emotion type value, the driving unit 500 in step S5 causes the arm 13 of the educational robot to maintain a normal state when the correction value is not given to the neutral type, and the correction value is (+) to the positive type ), the arm 13 of the educational robot is driven to move to the front of the body 14, and when the correction value increases to a (-) value in the negative type, the arm 13 of the educational robot moves to the front of the body (14) It can be driven to move backwards.
한편, 본 발명은 컴퓨터프로그램으로 구현될 수도 있다. 구체적으로, 하드웨어와 결합되어, 본 발명에 따른 교육용 로봇의 동작보정 방법을 컴퓨터에 의해 실행시키기 위하여 컴퓨터가 판독 가능한 기록매체에 저장된 컴퓨터 프로그램으로 구현될 수 있다.Meanwhile, the present invention may be implemented as a computer program. Specifically, in combination with hardware, it may be implemented as a computer program stored in a computer-readable recording medium in order to execute the method for correcting the motion of the educational robot according to the present invention by a computer.
이하에서는, 가상의 질의와 답변을 통해 본 발명의 특징을 더욱 구체적으로 설명하고자 한다.Hereinafter, the features of the present invention will be described in more detail through hypothetical questions and answers.
먼저, 문맥보정을 설명한다.First, context correction will be described.
대화가 여러마디 이상 진행되고 있을 경우, 그전 대화의 문맥에 관한 정보를 수신하는것이 가능하다. 대화의 유형에 따라서, 데이터베이스의 연관구동 보정정보를 참조해서, [표정 모터, 몸 모터, led 구동 ] 에 보정치가 더해질 수 있다.If a conversation is in progress for more than several words, it is possible to receive information about the context of the previous conversation. Depending on the type of conversation, a correction value can be added to [expression motor, body motor, led driving] by referring to the relevant driving correction information in the database.
예를 들어, 아래 표 1과 같은 질문정보와 답변정보를 가진 대화정보가 있다고 가정하면, 'Q1' 의 질의에 대해서, 로봇은 'A1' 의 답변을 시작하면서, 앞으로 추가적으로 대화를 이어나가기 위한 문맥 키워드를 생성을 하게 된다. 표 1과 같이 "산책", "긍정"이라는 문맥 키워드가 생성될 수 있다.For example, assuming that there is dialogue information with question information and answer information as shown in Table 1 below, for the query of 'Q1', the robot starts to answer 'A1', and the context for further continuing the conversation in the future create keywords. As shown in Table 1, context keywords such as "walking" and "positive" may be generated.
그리고 사용자가 'Q2'의 질의를 했을때, 로봇은 'A2'의 답변을 하게되는데, 이때 상기 문맥 키워드를 참조할 수 있다.And when the user makes a query of 'Q2', the robot answers 'A2'. At this time, the context keyword can be referred to.
[Q1] (사용자 질문)[Q1] (User Question) 오늘 날씨 어때? how's the weather today
[A1] (로봇의 답변)[A1] (Robot's answer) 아주 좋은 날이야, 산책을 나가고 싶어~!
: 문맥 키워드(context)-산책, 긍정, 정보전달It's a great day, I want to go for a walk!
: Context keywords - walking, affirmation, information transfer
[Q2] (사용자 질문)[Q2] (User Question) 어디로 산책 나가면 좋을까? Where would you like to go for a walk?
[A2] (로봇의 답변)[A2] (Robot's answer) 근처 공원이든 어디든 좋지 않을까? Wouldn't it be nice to have a nearby park or somewhere?
문맥 키워드가 "긍정"이므로, 전체적인 대화 분위기가 긍정적임을 알게 된다. 미리 설정된 프리셋 정보에 따라(예로, "긍정" : [표정+1]) 표정에서 눈썹의 각도에 +1 도를 추가할 수 있다. 따라서 로봇의 눈썹은 올라가고 로봇의 표정은 더 웃는상이 될 수 있다.Since the context keyword is "positive", we know that the overall conversational atmosphere is positive. According to preset preset information (eg, "positive": [expression+1]), +1 degree may be added to the angle of the eyebrows in the facial expression. Therefore, the robot's eyebrows are raised and the robot's facial expression can become a more smiling image.
또한, 문맥 키워드가 "정보전달"이므로, 전체적인 대화 분위기가 정보전달정 분위기임을 알게 된다. 미리 설정된 프리셋 정보에 따라(예로, "정보전달" : [제스처 회수 +1] ) 팔을 흔드는 제스처 회수가 원래 2회이었는데 1회를 추가해서 총 3회로 팔을 위아래로 왔다갔다하게, 궤적을 수정할 수 있다.In addition, since the context keyword is “information transfer”, it is known that the overall conversational atmosphere is information transfer-oriented atmosphere. According to the preset information set in advance (eg, "information transfer": [number of gestures +1] ), the number of gestures to shake the arm was originally 2 times, but by adding 1, the trajectory can be modified by moving the arm up and down for a total of 3 times. can
다음으로, 감정보정을 설명한다.Next, emotion correction will be described.
현재 로봇이 이야기할 대화문에서 감정의 정도 [부정]-1~1[긍정] 를 받아서 보정할 수 있고, [표정 모터, 몸 모터, led 구동] 에 보정치가 더해질 수 있다.The level of emotion [negative] -1 to 1 [positive] in the dialogue that the robot will talk about can be corrected, and the correction value can be added to [expression motor, body motor, led drive].
Figure PCTKR2020016933-appb-img-000001
Figure PCTKR2020016933-appb-img-000001
또한, 아래 표 3의 대화문의 경우, 로봇이 대답할 " 아주 좋은날이야, 산책을 나가고싶어~! " 에 대해서 딥러닝 기술을 통해서 감정평가를 진행하게 된다. In addition, in the case of the dialogue question in Table 3 below, the robot's response to "It's a very nice day, I want to go for a walk~!" is evaluated through deep learning technology.
[Q] (사용자 질문)[Q] (User Question) 오늘 날씨 어때? how's the weather today
[A] (로봇의 답변)[A] (Robot's answer) 아주 좋은 날이야, 산책을 나가고 싶어~! It's a great day, I want to go for a walk!
감정판정 결과는 감정 유형과 감정 강도의 2가지로 수행될 수 있다.The emotion determination result may be performed in two ways: an emotion type and an emotion intensity.
감정 유형의 경우, 다음 표 4와 같이 구분될 수 있다.In the case of emotion types, they can be classified as shown in Table 4 below.
[0.25 ~ +1][0.25 to +1] 긍정Positive
[-0.25 ~ +0.25][-0.25 to +0.25] 중립neutrality
[-1 ~ -0.25][-1 to -0.25] 부정denial
예를 들어, 표 3에서 아주 좋은날이야, 산책을 나가고 싶어~! " 같은 경우, 긍정에 가까운 감정이며, 0.9 의 값이 나올 수 있다.For example, in Table 3, it's a very good day, I want to go for a walk~! " In this case, it is an emotion close to positive, and a value of 0.9 can be obtained.
감정 강도의 경우, 위와 같은 감정의 긍정과 부정을 얼마나 강하게 느끼고 있는지 그 강도가 측정되는 것이다. 예를 들어, 0~1 사이 값으로 측정될 수 있다.In the case of emotional intensity, how strongly one feels the affirmation and negation of the above emotions is measured. For example, it may be measured as a value between 0 and 1.
표 3에서 " 아주 좋은날이야, 산책을 나가고 싶어~! " 같은 경우, 0.9라는 매우 강렬한 감정으로 평가될 수 있을 것이다.In Table 3, "It's a very nice day, I want to go for a walk!", it can be evaluated as a very strong emotion of 0.9.
먼저, 감정유형값과 감정강도값이 다음과 같이 예시되는 경우를 설명하고자 한다.First, a case in which the emotion type value and the emotion intensity value are exemplified as follows will be described.
[감정유형값: +0.9점, 감정강도값: +0.9점][Emotion type value: +0.9 points, Emotion intensity value: +0.9 points]
감정유형값이 +0.9점인 경우, 표정에서 눈썹의 각도에 3(최대보정)*09, +27도를 추가하게 될 것이다. 따라서 로봇의 눈썹은 올라가고 로봇의 표정은 더 웃는상이 될 것이다.If the emotion type value is +0.9 points, 3 (maximum correction) * 09, +27 degrees will be added to the angle of the eyebrows in the expression. Therefore, the robot's eyebrows will rise and the robot's facial expression will be more smiling.
또한, 감정강도값이 +0.9점인 경우, 구동부(모터)에 의해 눈썹과 팔 목 등이 20% * 09 = 18%, 만큼 더 많이 움직이게 될 것이다.In addition, when the emotional intensity value is +0.9 points, the eyebrows and wrists and neck will move more by 20% * 09 = 18% by the driving unit (motor).
다음으로, 감정유형값과 감정강도값이 다음과 같이 예시되는 경우를 설명하고자 한다.Next, a case in which the emotion type value and the emotion intensity value are exemplified as follows will be described.
[감정유형값: -0.7점, 감정강도값: +0.5점][Emotion type value: -0.7 points, Emotion intensity value: +0.5 points]
감정유형값이 -0.7점인 경우, 표정에서 눈썹의 각도에 3(최대보정)*-07, 21도를 삭감하게 될 것이다. 따라서 로봇의 눈썹은 내려가고 로봇의 표정은 화난 느낌을 주게 될 것이다.If the emotion type value is -0.7 points, the angle of the eyebrows in the expression will be reduced by 3 (maximum correction)*-07, 21 degrees. Therefore, the robot's eyebrows will go down and the robot's expression will give an angry feeling.
또한, 감정강도값이 +0.5점인 경우, 구동부(모터)에 의해 눈썹과 팔, 목 등 이 20% * 05 = 10%, 만큼 더 많이 움직이게 될 것이다.Also, when the emotional intensity value is +0.5, the eyebrows, arms, and neck will be moved more by 20% * 05 = 10% by the driving unit (motor).
다음으로, 얼굴궤적 보정에 관하여 설명한다.Next, face trajectory correction will be described.
궤적 계산전 물리적 한계를 확인한 다음, 얼굴모터의 궤적을 계산할 수 있다. 얼굴 모터의 궤적을 확인해서, 특정구간에 있을 경우 [몸체 모터, led 구동 ] 에 보정치가 더해질 수 있다.After checking the physical limits before calculating the trajectory, the trajectory of the face motor can be calculated. By checking the trajectory of the face motor, if it is in a specific section, a correction value can be added to [body motor, led driving].
예를 들어, 눈썹의 각도가 0도 이하인 구간에 진입했을 경우, 눈썹의 각도가 0도 이하이면, 부정적이고 화난 표정을 가지게 될 것이다. 이때 몸의 동작에서 팔이 가만히 있는 상황이었다면, 화난 감정을 표시하기 위해서 팔을 앞으로 흔드는 제스처를 한번 추가해줄 수 있다. 또한, led 램프의 원래 색상이 약한 노란색이엇다면, 빨간색을 추가하여, 약한 주황색으로 변경할 수 있을 것이다.For example, if you enter a section where the angle of the eyebrows is 0 degrees or less, if the angle of the eyebrows is less than 0 degrees, you will have a negative and angry expression. At this time, if the arm is still in the body motion, you can add a gesture of waving the arm forward to show the angry emotion. Also, if the original color of the led lamp was weak yellow, it could be changed to a weak orange color by adding red.
다음으로, 몸체 궤적 보정을 설명한다.Next, body trajectory correction will be described.
궤적 계산전 물리적 한계를 확인한 다음, 몸체 모터의 궤적을 계산한다. 몸체 모터의 궤적을 확인해서, 특정 구간에 있을 경우, [ led 구동 ] 에 보정치가 더해지고, [얼굴모터]는 이미 만들어진 궤적을 보정하게 된다.Before calculating the trajectory, check the physical limits and then calculate the trajectory of the body motor. After checking the trajectory of the body motor, if it is in a specific section, the correction value is added to [ led drive ], and the [face motor] corrects the already made trajectory.
예를 들어, 팔이 회전을 하다가 등 뒤쪽으로 넘어간 구간에 진입했을 경우(0도 이하), 불편한 자세임으로 표정을 찡그리게 보정을 추가(눈썹의 각도에 -3도를 추가)할 수 있다. 또한, led 램프의 원래 색상이 약한 흰색이었다면, 빨간색을 추가해서, 약한 빨간색으로 변경할 수 있다.For example, if you enter a section where your arm goes over your back while rotating (0 degrees or less), you can add a correction (add -3 degrees to the angle of your eyebrows) to make your face frown due to an uncomfortable posture. Also, if the original color of the led lamp was weak white, it can be changed to weak red by adding red.
다음으로, LED 구동 보정을 설명한다.Next, LEDs Driving correction will be described.
궤적 계산전 물리적 한계를 확인한 다음, LED 램프의 동작 정보를 계산한다. LED 램프의 동작 정보를 확인해서, 특정 구간에 있을 경우, [얼굴모터,및 몸 모터에]는 이미 만들어진 궤적을 보정하게 된다.Before calculating the trajectory, check the physical limit and then calculate the operation information of the LED lamp. After checking the operation information of the LED lamp, if it is in a specific section, [for face motor and body motor] will correct the already created trajectory.
예를 들어, LED 램프가 파란색 계열( R G B, LED 값중 R+G <B 값인 경우)으로 동작하는 구간에서는 파란색은 차분한 느낌을 줌으로, 전체 궤적값을 10% 감소해서 움직이게 된다.For example, in the section where the LED lamp operates in the blue series (R G B, if R+G < B among the LED values), the blue color gives a sense of calm and reduces the total trajectory value by 10%.
다음으로, 최종 궤적 검사를 설명한다.Next, the final trajectory inspection will be described.
모든 궤적을 계산한 이후 최종적으로 물리적 한계를 넘어서 모터를 손상시키거나 충돌하는 구간이 있는지 검사한 뒤에, 구동부 모터 및 LED 램프를 동작시킬 수 있다.After calculating all the trajectories, it is finally checked whether there is a section where the motor is damaged or collided beyond the physical limit, and then the drive motor and the LED lamp can be operated.
도 10은 본 발명에 따른 컴퓨팅 장치를 나타내는 도면이다.10 is a diagram illustrating a computing device according to the present invention.
본 발명에 따른 컴퓨팅 장치(TN100)는 적어도 하나의 프로세서(TN110), 송수신 장치(TN120), 및 메모리(TN130)를 포함할 수 있다. 또한, 컴퓨팅 장치(TN100)는 저장 장치(TN140), 입력 인터페이스 장치(TN150), 출력 인터페이스 장치(TN160) 등을 더 포함할 수 있다. 컴퓨팅 장치(TN100)에 포함된 구성 요소들은 버스(bus)(TN170)에 의해 연결되어 서로 통신을 수행할 수 있다.The computing device TN100 according to the present invention may include at least one processor TN110 , a transceiver device TN120 , and a memory TN130 . In addition, the computing device TN100 may further include a storage device TN140 , an input interface device TN150 , an output interface device TN160 , and the like. Components included in the computing device TN100 may be connected by a bus TN170 to communicate with each other.
프로세서(TN110)는 메모리(TN130) 및 저장 장치(TN140) 중에서 적어도 하나에 저장된 프로그램 명령(program command)을 실행할 수 있다. 프로세서(TN110)는 중앙 처리 장치(CPU: central processing unit), 그래픽 처리 장치(GPU: graphics processing unit), 또는 본 발명의 실시예에 따른 방법들이 수행되는 전용의 프로세서를 의미할 수 있다. 프로세서(TN110)는 본 발명의 실시예와 관련하여 기술된 절차, 기능, 및 방법 등을 구현하도록 구성될 수 있다. 프로세서(TN110)는 컴퓨팅 장치(TN100)의 각 구성 요소를 제어할 수 있다.The processor TN110 may execute a program command stored in at least one of the memory TN130 and the storage device TN140. The processor TN110 may mean a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor on which methods according to an embodiment of the present invention are performed. The processor TN110 may be configured to implement procedures, functions, and methods described in connection with an embodiment of the present invention. The processor TN110 may control each component of the computing device TN100.
메모리(TN130) 및 저장 장치(TN140) 각각은 프로세서(TN110)의 동작과 관련된 다양한 정보를 저장할 수 있다. 메모리(TN130) 및 저장 장치(TN140) 각각은 휘발성 저장 매체 및 비휘발성 저장 매체 중에서 적어도 하나로 구성될 수 있다. 예를 들어, 메모리(TN130)는 읽기 전용 메모리(ROM: read only memory) 및 랜덤 액세스 메모리(RAM: random access memory) 중에서 적어도 하나로 구성될 수 있다. Each of the memory TN130 and the storage device TN140 may store various information related to the operation of the processor TN110. Each of the memory TN130 and the storage device TN140 may be configured as at least one of a volatile storage medium and a nonvolatile storage medium. For example, the memory TN130 may include at least one of a read only memory (ROM) and a random access memory (RAM).
송수신 장치(TN120)는 유선 신호 또는 무선 신호를 송신 또는 수신할 수 있다. 송수신 장치(TN120)는 네트워크에 연결되어 통신을 수행할 수 있다. The transceiver TN120 may transmit or receive a wired signal or a wireless signal. The transceiver TN120 may be connected to a network to perform communication.
한편, 앞서 설명된 본 발명의 실시예에 따른 방법들은 다양한 컴퓨터수단을 통하여 판독 가능한 프로그램 형태로 구현되어 컴퓨터로 판독 가능한 기록매체에 기록될 수 있다. 여기서, 기록매체는 프로그램 명령, 데이터 파일, 데이터구조 등을 단독으로 또는 조합하여 포함할 수 있다. 기록매체에 기록되는 프로그램 명령은 본 발명을 위하여 특별히 설계되고 구성된 것들이거나 컴퓨터 소프트웨어 당업자에게 공지되어 사용 가능한 것일 수도 있다. 예컨대 기록매체는 하드 디스크, 플로피 디스크 및 자기 테이프와 같은 자기 매체(magnetic media), CDROM, DVD와 같은 광 기록 매체(optical media), 플롭티컬 디스크(floptical disk)와 같은 자기-광 매체(magneto-optical media), 및 롬(ROM), 램(RAM), 플래시 메모리 등과 같은 프로그램 명령을 저장하고 수행하도록 특별히 구성된 하드웨어 장치를 포함한다. 프로그램 명령의 예에는 컴파일러에 의해 만들어지는 것과 같은 기계어뿐만 아니라 인터프리터 등을 사용해서 컴퓨터에 의해서 실행될 수 있는 고급 언어를 포함할 수 있다. 이러한 하드웨어 장치는 본 발명의 동작을 수행하기 위해 하나 이상의 소프트웨어 모듈로서 작동하도록 구성될 수 있으며, 그 역도 마찬가지이다.On the other hand, the methods according to the embodiment of the present invention described above may be implemented in the form of a program readable by various computer means and recorded in a computer readable recording medium. Here, the recording medium may include a program command, a data file, a data structure, etc. alone or in combination. The program instructions recorded on the recording medium may be specially designed and configured for the present invention, or may be known and available to those skilled in the art of computer software. For example, the recording medium includes magnetic media such as hard disks, floppy disks and magnetic tapes, optical recording media such as CDROMs and DVDs, and magneto-optical media such as floppy disks. optical media), and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions may include high-level languages that can be executed by a computer using an interpreter or the like as well as machine language such as generated by a compiler. Such hardware devices may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.
본 명세서에서 설명되는 실시예와 첨부된 도면은 본 발명에 포함되는 기술적 사상의 일부를 예시적으로 설명하는 것에 불과하다. 따라서, 본 명세서에 개시된 실시예들은 본 발명의 기술적 사상을 한정하기 위한 것이 아니라 설명하기 위한 것이므로, 이러한 실시예에 의하여 본 발명의 기술 사상의 범위가 한정되는 것은 아님은 자명하다. 본 발명의 명세서 및 도면에 포함된 기술적 사상의 범위 내에서 당업자가 용이하게 유추할 수 있는 변형 예와 구체적인 실시 예는 모두 본 발명의 권리범위에 포함되는 것으로 해석되어야 할 것이다.The embodiments described in this specification and the accompanying drawings are merely illustrative of some of the technical ideas included in the present invention. Accordingly, since the embodiments disclosed in the present specification are for explanation rather than limitation of the technical spirit of the present invention, it is obvious that the scope of the technical spirit of the present invention is not limited by these embodiments. Modifications and specific embodiments that can be easily inferred by those skilled in the art within the scope of the technical spirit included in the specification and drawings of the present invention should be interpreted as being included in the scope of the present invention.

Claims (25)

  1. 답변정보 생성부, 문맥보정 판정부, 감정보정 판정부 및 신체동작 보정연산부를 포함하는 서버 및 데이터베이스를 가지며, 컴퓨팅 수단에 의해 수행되는 교육용 로봇의 동작보정 시스템으로서, A motion correction system for an educational robot that has a server and a database including an answer information generating unit, a context correction determining unit, an emotion correction determining unit, and a body motion correction calculating unit, and performed by a computing means,
    사용자의 질문정보를 수신하는 질문정보 수신부;a question information receiving unit for receiving the user's question information;
    상기 질문정보 수신부에서 수신된 질문정보에 대한 답변정보를 생성하는 답변정보 생성부;an answer information generating unit for generating answer information for the question information received from the question information receiving unit;
    기 설정된 기준에 따라, 상기 답변정보 생성부에서 산출된 답변정보의 감정 및 강도를 판정하는 감정보정 판정부;an emotion correction determining unit for determining the emotion and strength of the answer information calculated by the answer information generating unit according to a preset criterion;
    상기 감정보정 판정부에서 산출된 감정 및 감정의 정도에 따라, 교육용 로봇의 신체의 동작 움직임을 보정하는 신체동작 보정연산부; 및a body motion correction calculation unit for correcting the body motion of the educational robot according to the emotion and the degree of emotion calculated by the emotion correction determining unit; and
    상기 신체동작 보정연산부에서 산출된 보정값을 적용하여 신체를 구동시키는 구동부를 포함하는 것을 특징으로 하는 교육용 로봇의 동작보정 시스템.and a driving unit for driving the body by applying the correction value calculated by the body motion correction operation unit.
  2. 청구항 1에 있어서,The method according to claim 1,
    상기 답변정보 생성부는 당해 질문정보 이전에 수신된 질문정보로부터 문맥을 파악하여 당해 질문정보의 답변정보를 산출하는 문맥보정 판정부를 더 구비하는 것을 특징으로 하는 교육용 로봇의 동작보정 시스템.The response information generating unit further comprises a context correction determining unit for calculating the answer information of the question information by grasping the context from the question information received before the question information.
  3. 청구항 2에 있어서,3. The method according to claim 2,
    답변정보 생성부, 문맥보정 판정부 및 감정보정 판정부는 자연어처리 기법 또는 딥러닝 기법으로 정보를 분석하는 것을 특징으로 하는 교육용 로봇의 동작보정 시스템.An answer information generating unit, a context correction determining unit, and an emotion correction determining unit are an educational robot motion correction system, characterized in that it analyzes information using a natural language processing technique or a deep learning technique.
  4. 청구항 2에 있어서,3. The method according to claim 2,
    상기 답변정보 생성부가 당해 질문정보 이전에 수신된 대화정보가 있다고 판정하면, 당해 질문정보 또는 답변정보를 문맥보정 판정부로 보내며,If the answer information generation unit determines that there is conversation information received before the question information, it sends the question information or answer information to the context correction determination unit,
    상기 문맥보정 판정부가 당해 질문정보와 이전에 수신된 대화정보를 상기 데이터베이스에 저장된 문맥정보와 대비하여, 당해 질문정보와 이전 대화정보의 문맥이 기 설정된 문맥범위 내에 속한다고 판정한 때에는,When the context correction determining unit compares the question information and the previously received conversation information with the context information stored in the database, and determines that the context of the question information and the previous conversation information falls within a preset context range,
    상기 문맥보정 판정부는 기 설정된 문백범위 내의 문맥 키워드를 포함하는 답변정보를 생성하는 것을 특징으로 하는 교육용 로봇의 동작보정 시스템.The context correction determining unit is an educational robot motion correction system, characterized in that it generates answer information including a context keyword within a preset textual range.
  5. 청구항 2에 있어서,3. The method according to claim 2,
    상기 답변정보 생성부가 당해 질문정보 이전에 수신된 대화정보가 없다고 판정한 때 또는 When the answer information generating unit determines that there is no conversation information received before the question information, or
    상기 답변정보 생성부가 당해 질문정보 이전에 수신된 대화정보가 있다고 판정하여 당해 질문정보를 문맥보정 판정부로 보내고, 상기 문맥보정 판정부가 당해 질문정보와 이전에 수신된 대화정보 및 상기 데이터베이스에 저장된 문백정보를 대비하여, 당해 질문정보와 이전 질문정보의 문맥이 기 설정된 문맥범위내에 속한다고 판정한 때에는,The answer information generating unit determines that there is dialogue information received before the question information, and sends the question information to the context correction determining unit, and the context correction determining unit includes the question information, the previously received dialogue information, and the text stored in the database. In comparison with the information, when it is determined that the context of the question information and the previous question information falls within the preset context range,
    상기 답변정보 생성부는 데이터베이스에서 당해 질문정보의 키워드가 속한 키워드 그룹의 키워드를 포함하는 답변정보를 생성하는 것을 특징으로 하는 교육용 로봇의 동작보정 시스템.The response information generating unit generates response information including a keyword of a keyword group to which the keyword of the question information belongs in a database.
  6. 청구항 4 또는 청구항 5에 있어서,6. The method according to claim 4 or 5,
    상기 감정보정 판정부는 감정의 유형을 수치화하는 감정유형 판정부 및 감정의 강도를 수치화하는 감정강도 판정부로 구비되는 것을 특징으로 하는 교육용 로봇의 동작보정 시스템.The emotion correction determination unit is an educational robot motion correction system, characterized in that it is provided with an emotion type determination unit to quantify the type of emotion and an emotion strength determination unit to quantify the intensity of the emotion.
  7. 청구항 6에 있어서,7. The method of claim 6,
    상기 감정유형 판정부는 상기 답변정보를 분석하여 감정유형을 긍정유형, 중립유형 및 부정유형으로 구분하고, 기 설정된 범위내의 감정유형값으로 산출하는 것을 특징으로 하는 교육용 로봇의 동작보정 시스템.The emotion type determination unit analyzes the answer information to classify the emotion type into a positive type, a neutral type and a negative type, and calculates the emotion type value within a preset range.
  8. 청구항 7에 있어서,8. The method of claim 7,
    상기 감정강도 판정부는 상기 답변정보를 분석하여 상기 각 유형의 강도를 기 설정된 범위내의 감정강도값으로 산출하는 것을 특징으로 하는 교육용 로봇의 동작보정 시스템.The emotional strength determination unit analyzes the answer information and calculates each type of intensity as an emotional intensity value within a preset range.
  9. 청구항 8에 있어서,9. The method of claim 8,
    상기 신체동작 보정연산부는 감정보정 판정부가 산출한 감정유형값 및 감정강도값에 따른 신체동작 보정값을 산출하는 것을 특징으로 하는 교육용 로봇의 동작보정 시스템.The body motion correction calculation unit calculates a body motion correction value according to the emotion type value and the emotion intensity value calculated by the emotion correction determination unit.
  10. 청구항 9에 있어서,10. The method of claim 9,
    상기 신체동작 보정연산부는 감정유형이 긍정유형이면 보정값은 (+)값으로 증가되고, 감정유형이 중립유형이면 보정값은 미부여되고, 감정유형이 부정유형이면 보정값은 (-)값으로 증가되는 것을 특징으로 하는 교육용 로봇의 동작보정 시스템.If the emotion type is a positive type, the correction value is increased to a (+) value, if the emotion type is a neutral type, the correction value is not given, and if the emotion type is a negative type, the correction value is a (-) value The motion correction system of the educational robot, characterized in that increased.
  11. 청구항 10에 있어서,11. The method of claim 10,
    상기 신체동작 보정연산부는 감정강도가 없을때의 값과 감정강도가 가장 강할때의 값의 범위 내에서 측정값을 가지는 것을 특징으로 하는 교육용 로봇의 동작보정 시스템.The body motion correction calculation unit has a measurement value within the range of a value when there is no emotional intensity and a value when the emotional intensity is the strongest.
  12. 청구항 11에 있어서,12. The method of claim 11,
    상기 구동부는 상기 감정유형값에 있어서,In the emotion type value, the driving unit
    상기 중립유형으로 보정값이 미부여되면, 교육용 로봇의 눈썹이 수평상태를 유지하도록 하며,When the correction value is not given as the neutral type, the eyebrows of the educational robot are maintained in a horizontal state,
    상기 긍정유형으로 보정값이 (+)값으로 증가하면, 교육용 로봇의 눈썹의 안쪽 단부가 상승하도록 대응 구동시키며,When the correction value increases to a (+) value in the positive type, the inner end of the eyebrow of the educational robot is driven correspondingly to rise,
    상기 부정유형으로 보정값이 (-)값으로 증가하면, 교육용 로봇의 눈썹의 바깥쪽 단부가 상승하도록 대응 구동시키는 것을 특징으로 하는 교육용 로봇의 동작보정 시스템.When the correction value increases to a (-) value in the negative type, the educational robot motion correction system, characterized in that it is driven so that the outer end of the eyebrow of the educational robot rises.
  13. 청구항 12에 있어서,13. The method of claim 12,
    상기 구동부는 상기 감정강도값이 증가하면,When the driving unit increases the emotional intensity value,
    긍정유형 및 부정유형에서의 상기 보정값이 대응하여 더욱 증가하며, 상기 교육용 로봇의 눈썹의 안쪽 단부의 구동 범위도 대응하여 증가하는 것을 특징으로 하는 교육용 로봇의 동작보정 시스템.The correction value in the positive type and the negative type is further increased correspondingly, and the driving range of the inner end of the eyebrow of the educational robot is correspondingly increased.
  14. 청구항 11에 있어서,12. The method of claim 11,
    상기 구동부는 상기 감정유형값에 있어서,In the emotion type value, the driving unit
    상기 중립유형으로 보정값이 미부여되면, 교육용 로봇의 팔이 일반 상태를 유지하도록 하며,When the correction value is not given as the neutral type, the arm of the educational robot maintains a normal state,
    상기 긍정유형으로 보정값이 (+)값으로 증가하면, 교육용 로봇의 팔이 몸체 앞쪽으로 이동하도록 대응 구동시키며,When the correction value increases to a (+) value in the positive type, the arm of the educational robot is driven to move in front of the body,
    상기 부정유형으로 보정값이 (-)값으로 증가하면, 교육용 로봇의 팔이 몸체 뒷쪽으로 이동하도록 대응 구동시키는 것을 특징으로 하는 교육용 로봇의 동작보정 시스템.When the correction value increases to a (-) value in the negative type, the educational robot's motion correction system, characterized in that it drives the arm of the educational robot to move toward the back of the body.
  15. 답변정보 생성부, 문맥보정 판정부, 감정보정 판정부 및 신체동작 보정연산부를 포함하는 서버 및 데이터베이스를 가지며, 컴퓨터에 의해 수행되는 교육용 로봇의 동작보정 방법으로서, As a motion correction method of an educational robot performed by a computer having a server and a database including an answer information generating unit, a context correction determining unit, an emotion correction determining unit, and a body motion correction calculating unit, the method comprising:
    질문정보 수신부가 사용자의 질문정보를 수신하는 S1 단계;Step S1 of the question information receiving unit receiving the user's question information;
    답변정보 생성부가 상기 질문정보 수신부에서 수신된 질문정보에 대한 답변정보를 생성하는 S2 단계;S2 step of generating, by the answer information generating unit, answer information for the question information received from the question information receiving unit;
    감정보정 판정부가 기 설정된 기준에 따라, 상기 답변정보 생성부에서 산출된 답변정보의 감정 및 강도를 판정하는 S3 단계;S3 step of determining the emotion and strength of the answer information calculated by the answer information generating unit according to the emotion correction determining unit preset criteria;
    신체동작 보정연산부가 상기 감정보정 판정부에서 산출된 감정유형값과 감정강도값에 따라, 교육용 로봇의 신체의 적어도 일부의 동작보정값을 계산하는 S4 단계; 및S4 step of calculating, by the body motion correction calculation unit, a motion correction value of at least a part of the body of the educational robot according to the emotion type value and the emotion intensity value calculated by the emotion correction determining unit; and
    구동부가 상기 신체동작 보정연산부에서 산출된 보정값을 적용하여 신체를 구동시키는 S5 단계를 포함하는 것을 특징으로 하는 교육용 로봇의 동작보정 방법.and a step S5 in which a driving unit drives the body by applying the correction value calculated by the body motion correction operation unit.
  16. 청구항 15에 있어서,16. The method of claim 15,
    S3 단계의 감정보정 판정부는 감정의 유형을 수치화하는 감정유형 판정부 및 감정의 강도를 수치화하는 감정강도 판정부로 구비되는 것을 특징으로 하는 교육용 로봇의 동작보정 방법.The emotion correction determination unit of step S3 is an educational robot motion correction method, characterized in that it is provided with an emotion type determination unit that digitizes the type of emotion and an emotion strength determination unit that quantifies the intensity of emotion.
  17. 청구항 16에 있어서,17. The method of claim 16,
    S3 단계의 감정유형 판정부는 상기 답변정보를 분석하여 감정유형을 긍정유형, 중립유형 및 부정유형으로 구분하고, 기 설정된 범위내의 감정유형값으로 산출하는 것을 특징으로 하는 교육용 로봇의 동작보정 방법.The emotion type determination unit of step S3 analyzes the answer information, classifies the emotion types into positive types, neutral types and negative types, and calculates the emotion type values within a preset range.
  18. 청구항 17에 있어서,18. The method of claim 17,
    S3 단계의 감정강도 판정부는 상기 답변정보를 분석하여 상기 각 유형의 강도를 기 설정된 범위내의 감정강도값으로 산출하는 것을 특징으로 하는 교육용 로봇의 동작보정 방법.The emotional strength determination unit of step S3 analyzes the answer information and calculates the intensity of each type as an emotional intensity value within a preset range.
  19. 청구항 18에 있어서,19. The method of claim 18,
    S4 단계의 신체동작 보정연산부는 감정보정 판정부가 산출한 감정유형값 및 감정강도값에 따른 신체동작 보정값을 산출하는 것을 특징으로 하는 교육용 로봇의 동작보정 방법.The body motion correction calculation unit of step S4 calculates a body motion correction value according to the emotion type value and the emotion intensity value calculated by the emotion correction determination unit.
  20. 청구항 19에 있어서,20. The method of claim 19,
    S4 단계의 신체동작 보정연산부는 감정유형이 긍정유형이면 보정값은 (+)값으로 증가되고, 감정유형이 중립유형이면 보정값은 미부여되고, 감정유형이 부정유형이면 보정값은 (-)값으로 증가되는 것을 특징으로 하는 교육용 로봇의 동작보정 방법.The body motion correction operation unit of step S4 increases the correction value to a (+) value if the emotion type is a positive type. If the emotion type is a neutral type, the correction value is not given. If the emotion type is a negative type, the correction value is (-) A method of correcting the motion of an educational robot, characterized in that it increases with a value.
  21. 청구항 20에 있어서,21. The method of claim 20,
    S4 단계의 신체동작 보정연산부는 감정강도가 없을 때의 값과 감정강도가 가장 강할 때의 값의 범위 내에서 측정값을 가지는 것을 특징으로 하는 교육용 로봇의 동작보정 방법.The body motion correction calculation unit of step S4 has a measurement value within the range of a value when there is no emotional intensity and a value when the emotional intensity is the strongest.
  22. 청구항 21에 있어서,22. The method of claim 21,
    S5 단계의 구동부는 상기 감정유형값에 있어서,In the emotion type value, the driving unit of step S5,
    상기 중립유형으로 보정값이 미부여되면, 교육용 로봇의 눈썹이 수평상태를 유지하도록 하며,When the correction value is not given as the neutral type, the eyebrows of the educational robot are maintained in a horizontal state,
    상기 긍정유형으로 보정값이 (+)값으로 증가하면, 교육용 로봇의 눈썹의 안쪽 단부가 상승하도록 대응 구동시키며,When the correction value is increased to a (+) value in the positive type, the inner end of the eyebrow of the educational robot is driven correspondingly to rise,
    상기 부정유형으로 보정값이 (-)값으로 증가하면, 교육용 로봇의 눈썹의 바깥쪽 단부가 상승하도록 대응 구동시키는 것을 특징으로 하는 교육용 로봇의 동작보정 방법.When the correction value increases to a (-) value in the negative type, the operation correction method of the educational robot, characterized in that it is driven so that the outer end of the eyebrow of the educational robot rises.
  23. 청구항 22에 있어서,23. The method of claim 22,
    S5 단계의 구동부는 상기 감정강도값이 증가하면,When the driving unit of step S5 increases the emotional intensity value,
    긍정유형 및 부정유형에서의 상기 보정값이 대응하여 더욱 증가하며, 상기 교육용 로봇의 눈썹의 안쪽 단부의 구동 범위도 대응하여 증가하는 것을 특징으로 하는 교육용 로봇의 동작보정 방법.The correction value in the positive type and the negative type is further increased correspondingly, and the driving range of the inner end of the eyebrow of the educational robot is correspondingly increased.
  24. 청구항 21에 있어서,22. The method of claim 21,
    S5 단계의 구동부는 상기 감정유형값에 있어서,In the emotion type value, the driving unit of step S5,
    상기 중립유형으로 보정값이 미부여되면, 교육용 로봇의 팔이 일반 상태를 유지하도록 하며,When the correction value is not given as the neutral type, the arm of the educational robot maintains a normal state,
    상기 긍정유형으로 보정값이 (+)값으로 증가하면, 교육용 로봇의 팔이 몸체 앞쪽으로 이동하도록 대응 구동시키며,When the correction value increases to a (+) value in the positive type, the arm of the educational robot is driven to move in front of the body,
    상기 부정유형으로 보정값이 (-)값으로 증가하면, 교육용 로봇의 팔이 몸체 뒷쪽으로 이동하도록 대응 구동시키는 것을 특징으로 하는 교육용 로봇의 동작보정 방법.When the correction value increases to a (-) value in the negative type, the educational robot's motion correction method, characterized in that it is driven so that the arm of the educational robot moves to the rear side of the body.
  25. 하드웨어와 결합되어, 청구항 15에 따른 교육용 로봇의 동작보정 방법을 컴퓨터에 의해 실행시키기 위하여 컴퓨터가 판독 가능한 기록매체에 저장된 컴퓨터 프로그램.A computer program stored in a computer-readable recording medium in combination with hardware to execute the method for correcting the motion of the educational robot according to claim 15 by a computer.
PCT/KR2020/016933 2020-11-26 2020-11-26 Action correction system and action correction method for education robot WO2022114263A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR10-2020-0160784 2020-11-26
KR1020200160784A KR102480556B1 (en) 2020-11-26 2020-11-26 Operational Calibration System and Method of Training Robot

Publications (1)

Publication Number Publication Date
WO2022114263A1 true WO2022114263A1 (en) 2022-06-02

Family

ID=81755765

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2020/016933 WO2022114263A1 (en) 2020-11-26 2020-11-26 Action correction system and action correction method for education robot

Country Status (2)

Country Link
KR (1) KR102480556B1 (en)
WO (1) WO2022114263A1 (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006289508A (en) * 2005-04-05 2006-10-26 Sony Corp Robot device and its facial expression control method
JP2016193466A (en) * 2015-03-31 2016-11-17 シャープ株式会社 robot
JP2019000937A (en) * 2017-06-14 2019-01-10 トヨタ自動車株式会社 Communication device, communication robot, and communication control program
KR101997072B1 (en) * 2018-02-21 2019-10-02 한성대학교 산학협력단 Robot control system using natural language and operation method therefor
JP2019184694A (en) * 2018-04-04 2019-10-24 一般社団法人Ｉｔ＆診断支援センター・北九州 Conversation robot

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101882585B1 (en) 2016-11-29 2018-07-26 한양대학교 산학협력단 Method and system for classifying natural language sentence/paragraph readability in educational environment for hri

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006289508A (en) * 2005-04-05 2006-10-26 Sony Corp Robot device and its facial expression control method
JP2016193466A (en) * 2015-03-31 2016-11-17 シャープ株式会社 robot
JP2019000937A (en) * 2017-06-14 2019-01-10 トヨタ自動車株式会社 Communication device, communication robot, and communication control program
KR101997072B1 (en) * 2018-02-21 2019-10-02 한성대학교 산학협력단 Robot control system using natural language and operation method therefor
JP2019184694A (en) * 2018-04-04 2019-10-24 一般社団法人Ｉｔ＆診断支援センター・北九州 Conversation robot

Also Published As

Publication number Publication date
KR102480556B1 (en) 2022-12-23
KR20220073067A (en) 2022-06-03

Similar Documents

Publication Publication Date Title
WO2020213996A1 (en) Method and apparatus for interrupt detection
WO2012134180A2 (en) Emotion classification method for analyzing inherent emotions in a sentence, and emotion classification method for multiple sentences using context information
WO2021003930A1 (en) Quality inspection method, apparatus, and device for customer service audio, and computer readable storage medium
WO2019117466A1 (en) Electronic device for analyzing meaning of speech, and operation method therefor
WO2021112642A1 (en) Voice user interface
WO2020231209A1 (en) Method, apparatus, electronic device, and computer readable storage medium for voice translation
WO2021015308A1 (en) Robot and trigger word recognition method therefor
WO2018097439A1 (en) Electronic device for performing translation by sharing context of utterance and operation method therefor
WO2019000462A1 (en) Face image processing method and apparatus, storage medium, and electronic device
WO2020080846A1 (en) Electronic device, control method thereof, and sound output control system of the electronic device
WO2020091519A1 (en) Electronic apparatus and controlling method thereof
WO2019050137A1 (en) System and method of determining input characters based on swipe input
WO2022139155A1 (en) Electronic apparatus providing content-based care service and controlling method thereof
WO2021112631A1 (en) Device, method, and program for enhancing output content through iterative generation
WO2020168606A1 (en) Advertisement video optimising method, apparatus and device and computer readable storage medium
WO2021101051A1 (en) Electronic device for converting handwriting input to text and method of operating the same
WO2022114263A1 (en) Action correction system and action correction method for education robot
WO2019112117A1 (en) Method and computer program for inferring meta information of text content creator
WO2020141787A1 (en) Language correction system, method therefor, and language correction model learning method of system
WO2019190142A1 (en) Method and device for processing image
WO2021246812A1 (en) News positivity level analysis solution and device using deep learning nlp model
EP3707678A1 (en) Method and device for processing image
WO2023132657A1 (en) Device, method, and program for providing product trend prediction service
WO2023229094A1 (en) Method and apparatus for predicting actions
WO2022055107A1 (en) Electronic device for voice recognition, and control method therefor

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20963659

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20963659

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 27.10.2023)