KR20190109653A - Apparatus and method for generating and managing knowledge for service robot based on situation - Google Patents

Abstract

Embodiments of the present invention relate to an apparatus for generating and managing knowledge of a situation-based service robot, and to a method thereof. The apparatus for generating and managing knowledge of a situation-based service robot comprises: a knowledge manager using at least one of recognition data recognized by a service robot, work data of the service robot, and external data obtained by the outside of the service robot to generate ontology-based semantic memory for the service robot; and a database storing at least one of the recognition data, the work data, and the external data and the ontology-based semantic memory for the service robot.

Description

{APPARATUS AND METHOD FOR GENERATING AND MANAGING KNOWLEDGE FOR SERVICE ROBOT BASED ON SITUATION}

The present invention relates to an apparatus and method for generating and managing knowledge of a service robot, and more particularly, to an apparatus and method for generating and updating knowledge for a service robot based on a situation and ontology.

The conventional service robot provides a predetermined service corresponding to the voice history content input from the user according to a predetermined rule. However, such a service robot has a problem in that it cannot provide the same service or provide a service unless there is a change in a user's life pattern or a user's special input behavior.

For example, you can order a menu or bring a menu if you sit at a table at a fixed time.However, if the user does not give orders or if you sit outside or at a different time, the service will bring you a menu. There is no problem.

Korean Patent Publication No. 10-2010-0077930

In order for a service robot to respond appropriately to a person, it is necessary to create and manage the knowledge structure of the object from the memory of the past as well as the memory of the past.

Technical problem of the present invention is not limited to those mentioned above, another technical problem that is not mentioned will be clearly understood by those skilled in the art from the following description.

An apparatus for generating and managing knowledge of a situation-based service robot according to an embodiment of the present invention may be configured to provide a service using at least one of recognition data recognized by a service robot, job data of the service robot, and external data acquired from the outside of the service robot. Knowledge manager for generating ontology-based semantic memory for robots; And a database storing at least one of the recognition data, the work data, and the external data, and the ontology-based semantic memory for the service robot.

In one embodiment, it may further include an ontology storage for storing the ontology for the service robot.

In one embodiment, the knowledge manager stores the recognition data, the work data, and the external data in the database as instant memories, generates a situation tag describing a single situation using the stored instant memories, An illustration memory including a tag and quantitative information constituting the situation tag may be stored in the database.

In one embodiment, the situation tag may include information about who did what, when, where.

In one embodiment, the context tag further includes additional description information, the additional description information may include at least one of the surrounding environment, weather, schedule, people together.

In one embodiment, the quantitative information constituting the situation tag may be inferred by using the ontology for the service robot to further include an inferencing device for replacing the quantitative information with semantic information.

According to an embodiment, the apparatus may further include a situation classifier classifying the illustration memory for each situation, and the knowledge manager may generate new semantic memory from the illustration memory classified into the same situation and newly generate the semantic memory in the database. It may add or update the semantic memory stored in the database.

In an embodiment, the knowledge manager may update an instance and a relationship of the ontology for the service robot based on the generated semantic memory.

The recognition data may include at least one of personal identification information, location information, voice information, biosignal information, and robot status information, and the job data may include a service target, a job content, and a result of the robot. And at least one of a working state, and the external data may include at least one of weather information, news information, and schedule information.

Knowledge generation and management method of a situation-based service robot according to an embodiment of the present invention is executed by a computing device, the recognition data recognized by the service robot, the work data of the service robot, and the external data obtained from the outside of the service robot Storing as an instant memory in a database, generating a situation tag depicting one situation using the stored instant memory, and storing the illustration memory including the situation tag and quantitative information constituting the situation tag in the database. Storing, dividing the illustration memory by the situation and generating new meaning memory from the illustration memory classified into the same situation, and adding newly created meaning memory to the database, or updating the meaning memory stored in the database. It may include a step.

In an embodiment, the method may further include updating an instance and a relationship of the ontology for the service robot based on the generated semantic memory.

In the computer-readable recording medium according to an embodiment of the present invention, instructions for executing the above-described method may be stored.

According to one embodiment of the present invention, by creating and managing the knowledge structure of people and situations from the past memory, the event memory as well as the recognized memory for the user, the robot service that is appropriately changed according to habit changes or environmental changes Can be provided.

The effects of the present invention are not limited to the above-mentioned effects, and other effects not mentioned will be clearly understood by those skilled in the art from the description of the claims.

1 is a block diagram of an apparatus 1000 for generating and managing knowledge of a situation-based service robot according to an embodiment of the present invention.
2 is a detailed block diagram of a database 200 according to an embodiment of the present invention.
3 is a relationship diagram for explaining a process in which instant memory stored in the database 200 is processed into artwork memory, and meaning memory is generated or modified from an artwork company, according to an embodiment of the present invention.
4 is a diagram illustrating a process of generating a situation tag according to an embodiment of the present invention.
5 illustrates a change in the form of a situation tag according to an embodiment of the present invention.
6 is a flow chart of a knowledge generation and management method of a context-based service robot according to an embodiment of the present invention.

Embodiments described herein may have aspects that are wholly hardware, partly hardware and partly software, or wholly software. As used herein, "unit", "module", "device" or "system" and the like refer to hardware, a combination of hardware and software, or a computer related entity such as software. For example, parts, modules, devices, or systems herein refer to running processes, processors, objects, executables, threads of execution, programs, and / or computers. computer, but is not limited thereto. For example, both an application running on a computer and a computer may correspond to a part, module, device, system, or the like herein.

Embodiments have been described with reference to the flowchart presented in the drawings. Although the method is shown and described in a series of blocks for the sake of simplicity, the invention is not limited to the order of the blocks, and some blocks may occur in different order or simultaneously with other blocks than those shown and described herein. Various other branches, flow paths, and blocks may be implemented in order to achieve the same or similar results. In addition, not all illustrated blocks may be required for the implementation of the methods described herein. Furthermore, the method according to an embodiment of the present invention may be implemented in the form of a computer program for performing a series of processes, which may be recorded on a computer-readable recording medium.

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

1 is a block diagram of an apparatus 1000 for generating and managing knowledge of a situation-based service robot according to an embodiment of the present invention. The service robot is a mechanical device for observing and assisting a user around the user, and may be installed in a movable or fixed position. In addition, the service robot may provide a service that analyzes the user's behavior, ensures safety, and manages a schedule by sensing the surrounding environment information while observing the user. For example, the service robot may be used to help people in various places such as homes, public institutions, and hospitals.

Referring to FIG. 1, the knowledge generation and management apparatus 1000 of a situation-based service robot may include a knowledge manager 100 and a database 200. In another embodiment, the apparatus 1000 for generating and managing knowledge of the context-based service robot may further include at least one of the ontology repository 300, the reasoning machine 400, and the situation classifier 500. That is, in various embodiments, the ontology repository 300, the reasoner 400 and the situation classifier 500 may exist as built-in or external devices of the knowledge generation and management apparatus 1000.

The knowledge manager 100 uses at least one of recognition data (Pd, Perception data) recognized by the service robot, work data (Wd, Work data) of the service robot, and external data (Ed, External data) obtained from the outside of the service robot. By creating ontology-based semantic memory for service robots.

The knowledge manager 100 recognizes the service robot from the outside through a sensor (not shown) or a communication interface included in the knowledge manager 100, work data Wd of the service robot, and an external device obtained from the outside of the service robot. At least one of the data Ed may be received.

The database 200 may receive or exchange recognition data, work data, and external data from the knowledge manager 100. That is, the knowledge manager 100 may transmit specific data to the database 200, and may also take specific data stored in the database 200 and store the data again in the database 200 after processing.

The database 200 may store at least one of recognition data, work data, and external data and semantic memory based on the ontology for the service robot.

2 is a detailed block diagram of a database 200 according to an embodiment of the present invention. Referring to FIG. 2, the database may include an instant memory DB 210, an artwork memory DB 220, and a semantic memory DB 230. Here, the instant memory stored in the instant memory DB is short-term memory (memory), and the illustration memory and semantic memory stored in the oxidative memory DB and the semantic memory DB are long-term memory (memory). Here, short-term memory means data that is temporarily stored and long-term memory means data that is stored temporarily.

The ontology storage 300 may store the ontology for the service robot. The ontology stored in the ontology repository 300 may be used by the knowledge manager 100 or the reasoner 400 or the situation classifier 500.

Ontology is a model that formally describes the concept and the relationship between the concepts belonging to a specific domain, the ontology for service robot described herein formally describes the relationship between words and words belonging to the service robot field One model. The ontology repository 300 stores the relationship between the instance and the path length and the path length, the relationship between the classes including the instances having an association relationship and the path length.

Ontology for the service robot is a relationship between words for each user, situation, and corresponding service according to the situation is stored separately, that is, the ontology for the service robot is an ontology that can learn the service method to care for the user. Such ontology for service robot may be composed of classes, instances, properties and relationships.

A class is a concept or thing that belongs to the service robot field. For example, classes may include (how to use objects, how to eat (general knowledge), preferred food, personal information, family relationships (user's knowledge), robotization, robot code of conduct (interaction knowledge)) It doesn't happen.

An instance is the actual form of a class. For example, an instance may include, but is not limited to, a substantial form of class such as (a cup of water, a handle, a meal, a food, a user, a Korean food (a user's preferred food), a speech method (politeness, equality, flat), etc.).

A property is the nature or nature of a class or instance. For example, attributes can include the nature or nature of a class or instance, such as (color, shape, affiliation).

A relationship is a relationship that exists between a class or an instance. Relationships between classes or instances may be included, for example (used, constructed, liked, said), but are not limited to such.

The instant memory is data acquired by the knowledge manager 100 or received from the outside, and may include recognition data, work data, and external data. In this case, the knowledge manager 100 may store the recognition data, the work data, and the external data in the database 200 as instant memories. The knowledge manager 100 may generate a situation tag that describes one situation by using the stored instant memory. In addition, the knowledge manager 100 may store the illustration memory including the situation tag and the quantitative information constituting the situation tag in the database 200.

3 is a relationship diagram for explaining a process in which instant memory stored in the database 200 is processed into artwork memory, and meaning memory is generated or modified from an artwork company, according to an embodiment of the present invention.

Referring to FIG. 1, the recognition data relates to a person and a situation recognized by the service robot, and may include recognition information, location information, voice information, and biosignal information. The job data is about the work of the service robot and may include the service type and goal, the job contents and the current progress of the resultant job. For example, if the service robot is cleaning the floor, the service type may be cleaning, the target may be a living room and a home, and the progress may be information such as a living room completed home. The external data is information that can be obtained from the web or an app through a communication network, and may be weather information, news information, schedule information, or the like.

The knowledge manager 100 may generate a situation tag that describes one situation using this instant memory. In addition, the knowledge manager 100 may generate an example memory including the situation tag and the quantitative information constituting the situation tag and store it in the database 200. That is, illustration memory can be composed of situation tags and quantitative information.

The situation tag is a sentence that describes at least one situation and may include information about who did what, when, where, and what. In another embodiment, the situation tag may further include additional description information. Here, the additional description information may include at least one of the surrounding environment, the weather, the schedule, and the people together.

4 is a diagram illustrating a process of generating a situation tag according to an embodiment of the present invention. Referring to FIG. 4, the knowledge generator (part of the knowledge manager) may generate a situation tag by obtaining information about who, when, where, what, how, and why (5W1H) from instant memory. To this end, information included in one situation tag may be determined in consideration of the relationship between words (instant memory data) with the inferring machine referring to the ontology stored in the ontology store. In addition, the situation tag may further include who is with, the environment, the weather, the schedule.

5 illustrates a change in the form of a situation tag according to an embodiment of the present invention. The situation tag generated by the knowledge generator described in FIG. 4 is composed of instant information, and the instant information may be composed of quantitative information. For example, the user's location may not be a kitchen or a bedroom, but may be a coordinate value based on GPS or indoor reference point.

Referring to Figure 5 a situation tag consisting of quantitative information appears. Here, the information indicating the place is represented as quantitative information, such as (10, 20, 15). The reasoning machine 400 may infer a situation tag composed of the quantitative information by using the ontology for the service robot so as to give meaning to the quantitative information and replace the quantitative information with the semantic information. Referring to FIG. 5, the quantitative information (10, 20, 15) representing a place is replaced with a meaningful place of 'kitchen' and the situation tag is converted into a sentence having a meaning.

In addition, referring to FIG. 5, in addition to 5W1H, as an additional description information, the situation tag may be extended to that Bob, a user, ate with Alice. In addition, the situation tag may include that the schedule is completed by referring to a predetermined schedule of the withdrawal user (9 to 10 o'clock). In addition, the situation tag may contain information that Rain and Rain have been outside while Bob is eating.

The situation tag generated in this way may be a standard for constituting the artwork units in the illustration memory. Referring back to Figure 3, the illustrations stored in the illustration memory DB (article 1, illustration 2 ,,,) includes a situation tag, and the quantitative information (eg start / end time, work record, Instant memories such as human location records, robot location records, etc. may be included in each illustration memory unit. Since the situation tags constituting each illustration memory are different, the kind of quantitative information included in each illustration memory may be different. That is, in FIG. 3, the illustration 1 and the illustration 2 have different situation tags, and may be composed of different types of quantitative information and values. For example, Illustration 1 may be "He ate in the kitchen at 9 am", and Illustration 2 might be "He fell down in the living room at 10 am".

In one embodiment, the situation classifier 500 may classify the illustration memory in the illustration memory DB by situation. The classification criteria may be determined by the service robot manager. The classification criteria are items determined to be necessary for the robot to provide a service based on the user's characteristic information of the ontology. For example, it may be used based on a user's meal time and place, falling time and place, coughing time place, or other factors besides time and place, and may vary from user to user. The situation classifier 500 may classify the accumulated illustration memory by comparing the situation tags of the illustration memory according to these criteria.

The knowledge manager 100 may create new semantic memory from the illustrated memory classified into the same situation, add new semantic memory to the database, or update semantic memory stored in the database. For example, when illustration memories 5, 12, 13, and 30 are classified in the same situation, the illustration memories may be as follows.

Illustration memory Status Tag 3 Withdrawal fell in the living room at 9 am 5 Evacuation fell in the kitchen at 9:15 am 12 Withdrawal fell in the living room at 3 pm 13 Withdrawal sat in the bedroom at 8:50 am 30 Withdrawal hesitated in the bathroom at 9 am

The knowledge manager 100 may generate new semantic memory based on the illustrated memories 3, 5, 12, 13, and 30 classified as described above. For example, the knowledge manager 100 can create a new meaning memory, "Focus falls frequently around 9 am."

The knowledge manager 100 may update an instance and a relationship of the ontology for the service robot based on the generated semantic memory. For example, if the current instance and relationship of the ontology for service robots had a time of 9 am for the withdrawal fall, the time for a fall for the fall could have a higher relevance or importance than 9 am at 10 pm. Therefore, from the service robot side, it is possible to prepare a service for observing the fall of the withdrawal at about 9 am, and if a similar situation is found at the fall around 9 am, it can be recognized as a fall and provide a corresponding service.

 Referring to FIG. 3 again, if the attention time for falling into the meaning memory for the existing emergency response was 10 pm, the knowledge manager 100 updates the attention time for the fall in the meaning memory for the corresponding emergency response to 9 am. can do.

6 is a flow chart of a knowledge generation and management method of a context-based service robot according to an embodiment of the present invention. The knowledge generation and management method of the context-based service robot may be implemented by the components of the knowledge generation and management apparatus of the context-based service robot described above. In addition, the knowledge generation and management method of the context-based service robot may be implemented by a computing device.

Referring to FIG. 6, a method of generating and managing knowledge of a situation-based service robot includes storing recognition data recognized by a service robot, work data of the service robot, and external data obtained from the outside of the service robot as instant memories in a database. (S100), generating a situation tag depicting one situation using the stored instant memory (S200), and storing the illustration memory including the situation tag and the quantitative information constituting the situation tag in the database. (S300), the step of classifying the illustration memory for each situation (S400) and generates a new meaning memory from the illustration memory classified in the same situation, and adds the generated semantic memory to the database, or stored in the database It may include the step (S500) to update.

In one embodiment, the knowledge generation and management method of the context-based service robot may further include updating the instance and the relationship of the ontology for the service robot based on the generated semantic memory.

In addition, the computer-readable recording medium according to the embodiment of the present invention may store instructions for executing the above-described method.

Although the present invention described above has been described with reference to the embodiments shown in the drawings, this is merely exemplary, and it will be understood by those skilled in the art that various modifications and variations may be made therefrom. However, such modifications should be considered to be within the technical protection scope of the present invention. Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

Claims (12)

As a knowledge generation and management device of a situation-based service robot,
A knowledge manager for generating ontology-based semantic memory for the service robot using at least one of recognition data recognized by the service robot, work data of the service robot, and external data obtained from the outside of the service robot; And
And a database storing at least one of the recognition data, the work data, and the external data, and a ontology-based semantic memory for the service robot.
The method of claim 1,
Apparatus for creating and managing the knowledge of the situation-based service robot, characterized in that it further comprises an ontology storage for storing the ontology for the service robot.
The method of claim 1,
The knowledge manager,
Storing the recognition data, working data and external data in the database as instant storage,
A situation-based service robot, comprising: generating a situation tag depicting a situation using stored instant memory; and storing an illustration memory including the situation tag and quantitative information constituting the situation tag in the database. Knowledge generation and management device.
The method of claim 3,
The situation tag, the information generation and management device of a situation-based service robot, characterized in that it includes information on who, when, where, what.
The method of claim 4, wherein
The context tag further includes additional description information, wherein the additional description information includes at least one of the environment, the weather, the schedule, and the person with whom the knowledge-based service robot.
The method of claim 4, wherein
And a reasoning device for inferring the quantitative information constituting the situation tag by using the ontology for the service robot and replacing the quantitative information with semantic information.
The method of claim 6,
Further comprising a situation classifier for classifying the illustration memory by situation,
The knowledge manager,
An apparatus for creating and managing knowledge of a situation-based service robot, characterized by generating new semantic memory from the illustration memory classified into the same situation, adding newly created semantic memory to the database, or updating semantic memory stored in the database. .
The method of claim 7, wherein
The knowledge manager is a device for knowledge generation and management of a situation-based service robot, characterized in that for updating the instance and the relationship of the ontology for the service robot based on the created semantic memory.
The method of claim 1,
The recognition data includes at least one of human identification information, location information, voice information, biosignal information, and robot status information.
The job data includes at least one of a service target of the robot, job content and results, and job status,
The external data may include at least one of weather information, news information, and schedule information.
A method of creating and managing knowledge of a situation based service robot, executed by a computing device, the method comprising:
Storing the recognition data recognized by the service robot, the work data of the service robot, and external data obtained from the outside of the service robot as instant memories in a database;
Generating a situation tag depicting a situation using the stored instant memory;
Storing an illustration memory including the situation tag and quantitative information constituting the situation tag in the database;
Classifying the illustration memory by situation; And
Knowledge of the situation-based service robot, characterized in that it comprises the step of creating a new meaning memory from the illustration memory classified in the same situation, and adding the generated semantic memory to the database, or updating the semantic memory stored in the database How to create and manage.
The method of claim 10,
And updating the instance and the relationship of the ontology for the service robot based on the created semantic memory.
A computer-readable recording medium having stored thereon instructions for executing a method according to claim 10.

Images