A knowledge base self-updating method, device and robot for robot socialization

Abstract

The application relates to a knowledge base self-updating method for robot socialization, comprising the following steps: constructing a basic knowledge base containing various social information; reasoning the social relationship between different social objects and the social attribute of each social object through a preset common sense rule table, and complementing the basic knowledge base; querying the social attitude of a current social object from the complemented basic knowledge base according to a social behavior to be executed by a robot and the social attribute of the current social object; analyzing the social attitude into a social behavior parameter according to a preset mapping rule table, and acquiring an actual moving speed and an actual social distance according to the social behavior parameter and a preset corresponding relationship; executing the social behavior, recognizing and acquiring the expression feedback and the voice feedback of the current social object after the social behavior occurs by using an expression recognition algorithm and a text classification algorithm; and updating the social attitude of the current social object in the basic knowledge base according to the expression feedback and the voice feedback and according to a preset behavior parameter updating table.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent robots, in particular to a knowledge base self-updating method and device for robot socialization and a robot. BACKGROUND

[0002] With the development of artificial intelligence technology, service robots are applied to various scenarios due to their advantages in realizing natural and smooth human-robot interaction. Most of the current service robots do not have structured social knowledge, and they do not consider complex and comprehensive information such as environment, current task and object state when interacting with humans. Their behavior and attitude may be the same in the face of completely different situations; that is, although the robots are in different environments, performing different tasks, or facing different objects, they always interact with humans using the same fixed mode. Such social strategy not only appears rigid, but sometimes even offends users.

[0003] Currently, the social knowledge in the existing social knowledge expression method is often scattered and only realizes certain social norms in a specific scenario. At the same time, the existing technology lacks autonomous learning ability, hindering the personalized social knowledge acquisition of robots, making it difficult to meet the social interaction needs of service robots in different social backgrounds. In order to realize the structured expression of robot social knowledge, there are currently two main methods: a method based on artificial rules and a method based on demonstration learning, but both methods have certain limitations; the method based on artificial rules constructs a knowledge module through artificially defined reasoning rules, simulates a module to generate the actions of different characters, and transmits image data to a perception module; the perception module performs human body detection and tracking and face recognition, and transmits the perception data to the knowledge module; the knowledge module reasons out the social interaction state through the perception data; the method based on artificial rules strictly follows the behavior mode of "judgment-decision-execution", and the social behavior appears rigid and stiff; and the method based on demonstration learning needs a large amount of demonstration data set to train the model, which is costly. SUMMARY

[0004] Therefore, the technical problem to be solved by the present application is to overcome the problems of complex implementation of structured expression of robot social knowledge and inability to realize self-updating in the prior art.

[0005] To solve the above technical problems, the present application provides a knowledge base self-updating method for robot socialization, comprising:

[0006] constructing a basic knowledge base containing social objects, basic attributes, social attributes, social relationships, social behaviors and social attitudes;

[0007] Infer the social relationship between different social objects and the social attribute of each social object through a preset common sense rule table, and complete the basic knowledge base;

[0008] Query the social attitude of the current social object from the completed basic knowledge base through the social behavior to be performed by the robot and the social attribute of the current social object;

[0009] According to the preset mapping rule table, the social attitude of the current social object is analyzed into a social behavior parameter, and according to the preset corresponding relationship, the actual moving speed and the actual social distance are obtained;

[0010] According to the actual moving speed and the actual social distance, the social behavior is performed, and the expression feedback and the voice feedback of the current social object after the social behavior occurs are recognized and obtained by using an expression recognition algorithm and a text classification algorithm;

[0011] According to the expression feedback and the voice feedback, the social attitude of the current social object in the basic knowledge base is updated according to a preset behavior parameter update table.

[0012] In an embodiment of the present application, the basic knowledge base is constructed based on a Neo4j graph database.

[0013] In an embodiment of the present application, the construction of the basic knowledge base comprises:

[0014] Establish a character node for all social objects;

[0015] Fill in the basic attributes including name, gender, scene and identity for each character node;

[0016] Fill in the social relationship and the social attribute according to the basic attributes and a preset common sense rule table, wherein the social attribute includes a plurality of social behaviors and a plurality of social attitudes corresponding to the plurality of social behaviors;

[0017] The social attitude includes a social behavior parameter when the social behavior is performed on the current social object.

[0018] In an embodiment of the present application, the social attitude of the current social object is queried from the updated basic knowledge base through the social behavior to be performed by the robot and the social attribute of the current social object, comprising:

[0019] Obtain the face data of the current social object through the robot depth camera data;

[0020] Analyze the face data by using a face recognition network constructed by MTCNN and FaceNet to obtain the basic attributes of the current social object;

[0021] According to the basic attribute of the current social object and the basic knowledge base queried by the robot to perform the social behavior, the social attitude of the current social object is obtained.

[0022] In an embodiment of the present application, the social attitude of the current social object is queried from the basic knowledge base by the social behavior to be performed by the robot and the social attribute of the current social object.

[0023] If the current social object cannot be queried from the basic knowledge base, a person node is established for the current social object in the basic knowledge base.

[0024] In an embodiment of the present application, the actual moving speed and the actual social distance are obtained according to the social behavior parameters and the preset corresponding relationship, comprising:

[0025] The social behavior parameters include moving speed parameters and social distance parameters.

[0026] The actual moving speed of the robot is calculated according to the moving speed parameters.

[0027] ,

[0028] wherein, the moving speed parameter value is represented by v, the speed control signal issued by the dynamic window method in the local path planning algorithm is represented by v, and the maximum moving speed that the robot can reach is represented by v max.

[0029] The actual social distance between the robot and the current social object is calculated according to the social distance parameters.

[0030] ,

[0031] wherein, the social distance parameter value is represented by d, and the minimum social distance allowed by the robot is represented by d min.

[0032] In an embodiment of the present application, the expression feedback comprises:

[0033] The face data of the current social object is obtained by the depth camera data of the robot.

[0034] The expression recognition network constructed by the MTNCC and the convolutional neural network is used to analyze the face data, and the expression feedback of the current social object is obtained.

[0035] In an embodiment of the present application, the voice feedback comprises:

[0036] convert a voice instruction issued by a current social object into voice text data through a voice recognition algorithm;

[0037] a text classification network constructed by using a BERT model and a full connection network is used to identify the voice text information, so as to obtain voice feedback of the current social object.

[0038] The application further provides a knowledge base self-updating device for robot socialization, which comprises:

[0039] a knowledge base construction module, which is used to construct a basic knowledge base containing social objects, basic attributes, social attributes, social relationships, social behaviors and social attitudes;

[0040] a knowledge base updating module, which is used to infer social relationships between different social objects and social attributes of each social object through a preset common sense rule table, and to complete the basic knowledge base;

[0041] a social attitude query module, which is used to query a social attitude of a current social object from the completed basic knowledge base through a social behavior to be executed by a robot and a social attribute of the current social object;

[0042] a social attitude analysis module, which is used to analyze the social attitude of the current social object into a social behavior parameter according to a preset mapping rule table, and to obtain an actual moving speed and an actual social distance according to the social behavior parameter and a preset corresponding relationship;

[0043] a feedback analysis module, which is used to execute a social behavior according to the actual moving speed and the actual social distance, and to obtain an expression feedback and a voice feedback of the current social object after the social behavior occurs by using an expression recognition algorithm and a text classification algorithm;

[0044] a social attitude updating module, which is used to update the social attitude of the current social object in the basic knowledge base according to the expression feedback and the voice feedback and according to a preset behavior parameter updating table.

[0045] The application further provides a service robot, which comprises:

[0046] a depth camera, which is used to collect facial data of a current social object;

[0047] a sound sensor, which is used to collect a voice instruction issued by the current social object;

[0048] The knowledge base self-updating device for robot socialization described above is in communication connection with the depth camera and the sound sensor, is used for identifying a current social object according to the face data, and obtaining a social attitude of the current social object, calculating an actual moving speed and an actual social distance; obtaining an expression feedback according to the face data, obtaining a voice feedback according to the voice instruction, and updating the social attitude of the current social object in the basic knowledge base according to the expression feedback and the voice feedback, and according to a preset parameter update table;

[0049] The driving device is in communication connection with the knowledge base self-updating device for robot socialization, is used for driving the service robot to perform a social behavior according to the actual moving speed and the actual social distance.

[0050] The above technical solution of the present application has the following advantages compared with the prior art:

[0051] The robot socialization knowledge base self-updating method described in the present application takes a social object as a node, constructs a graphical social knowledge base, performs knowledge reasoning through a preset common sense rule table, and completes the social knowledge base; the behavior parameters when performing a social behavior on the current social object are updated according to a preset behavior parameter update table through the expression feedback and the voice feedback of the current social object, realizing the self-updating of the social knowledge base based on the current social object, making the social behavior parameters of the robot more in line with the social preferences of the current social object; different preset rules are set to meet the social interaction needs of the robot in different scenes. BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to make the content of the present application more easily understood, the present application will be further described in detail below according to specific embodiments of the present application and in conjunction with the drawings, in which:

[0053] Figure 1 is a step flow chart of the knowledge base self-updating method for robot socialization provided by the present application;

[0054] Figure 2 is a basic knowledge base schematic diagram for robot socialization provided by the present application;

[0055] Figure 3 is an actual moving speed calculation formula function image of the knowledge base self-updating method for robot socialization provided by the present application;

[0056] Figure 4 is an actual social distance calculation formula function image of the knowledge base self-updating method for robot socialization provided by the present application;

[0057] Figure 5 is an expression feedback identification schematic diagram of the knowledge base self-updating method for robot socialization provided by the present application;

[0058] Figure 6 is a voice feedback recognition schematic diagram of the knowledge base self-updating method for robot socialization provided by the present application;

[0059] Figure 7 is a social behavior parameter updating schematic diagram of the knowledge base self-updating method for robot socialization provided by the present application. DETAILED DESCRIPTION

[0060] The present application will be further described below in conjunction with the drawings and specific embodiments, so that those skilled in the art can better understand the present application and implement it. The embodiments are not limiting to the present application.

[0061] Referring to Figure 1 the present application provides a knowledge base self-updating method for robot socialization, comprising:

[0062] S1: constructing a basic knowledge base containing social objects, social behaviors, social relationships, social attitudes and social attributes;

[0063] S11: referring to Figure 2 based on a Neo4j graph database, a basic knowledge base containing social objects, social behaviors, social relationships, social attitudes and social attributes is constructed;

[0064] S12: in the basic knowledge base, a person node is established for all social objects in different scenarios;

[0065] S13: filling the basic attributes for each person node; the basic attributes include the name, gender, age, identity and other basic information of the person, and the values of the basic attributes will also be different in different scenarios;

[0066] S14: filling the social attributes for each person node according to the social rules; the social attributes refer to the social attitude that the robot should take when performing social behaviors on the current social object, and the movement speed parameter and the social distance parameter corresponding to the social attitude are obtained according to a preset rule mapping table;

[0067] S15: expressing the social relationships between different social objects by creating the edges between the person nodes;

[0068] S2: reasoning the social relationships between the social objects and the social attributes of each social object through common sense rules, and completing the basic knowledge base;

[0069] S21: reasoning the potential social relationships between the social objects according to a preset common sense rule table;

[0070] S22: reasoning the social attributes of the social objects according to their identity attributes;

[0071] S3: Query the social attitude of the current social object from the completed basic knowledge base by comparing the robot's social behavior with the social attributes of the current social object;

[0072] S31: Obtain facial data of the current social object through the robot's depth camera data;

[0073] S32: Construct a face recognition network using MTCNN+FaceNet;

[0074] S33: Use the face recognition network to identify the face data and obtain the name of the current social object;

[0075] S34: Based on the name of the current social contact and the robot's social behavior, query the basic knowledge base to obtain the social attitude;

[0076] S4: Parse the social attitude of the current social object into social behavior parameters according to the preset mapping rule table, update the social attitude in the basic knowledge base to the social behavior parameters, and execute social behavior according to the social behavior parameters;

[0077] S41: Parse the social attitude of the current social object into social behavior parameters according to the preset mapping rule table; the social behavior parameters include movement speed parameters and social distance parameters.

[0078] S42: Reference Figure 3 As shown, the robot's actual moving speed is calculated based on the moving speed parameters:

[0079] ,

[0080] in, This indicates the value of the movement speed parameter. This represents the speed control signal issued by the dynamic window method in the local path planning algorithm. This indicates the maximum speed the robot can move.

[0081] S43: Reference Figure 4 As shown, the actual social distance between the robot and the current social object is calculated based on the social distance parameter:

[0082] ,

[0083] in, Indicates the social distancing parameter value. Indicates the minimum social distance allowed for robots;

[0084] S44: Generate navigation target points based on the current location of the social object and the actual social distance;

[0085] S45: controlling the robot to reach the navigation behavior target point according to the actual moving speed, and performing the social behavior;

[0086] S5: recognizing the expression feedback and the voice feedback of the current social object after the social behavior occurs by using an expression recognition algorithm and a text classification algorithm;

[0087] S51: referring to Figure 5 as shown, obtaining face data of the current social object through a robot depth camera data; using MTCNN+convolutional neural network to construct an expression recognition network; and recognizing the face data by using the expression recognition network to obtain the expression feedback of the current social object;

[0088] S52: referring to Figure 6 as shown, converting the voice of the current social object into text information by using a voice recognition algorithm; using a BERT model+fully connected network to construct a text classification network; and recognizing the text information by using the text classification network to obtain the voice feedback of the current social object;

[0089] S6: updating the social attitude of the current social object in the basic knowledge base according to the expression feedback and the voice feedback, and according to a preset behavior parameter update table;

[0090] S61: referring to Figure 7 as shown, updating the social behavior parameter of the current social object by comparing the expression feedback and the voice feedback with the preset behavior parameter update table;

[0091] S62: using the updated social behavior parameter to replace the social attitude of the current social object in the basic knowledge base to obtain the social attitude of the current social object.

[0092] Specifically, in the basic knowledge base, a person node is established for all social objects in different scenes, and a node basic attribute and a social attribute are added to construct a social relationship. The scenes designed in this embodiment include a school scene, a hospital scene and a family scene; in the school scene, the identity attribute in the basic attribute has four values, which are student, parent, teacher and principal; in the hospital scene, the identity attribute in the basic attribute has four values, which are patient, doctor, nurse and family member; in the family scene, the identity attribute in the basic attribute has three values, which are father, mother and child; the value of the identity attribute of the person node will affect his secondary attribute, for example, the node with the identity of student will have a class attribute, and the node with the identity of doctor will have a department attribute.

[0093] In the embodiment of the application, the preset common sense rule table is shown in Table 1, which is part of the common sense rules, and the robot derives the social relationship between different social objects according to the common sense rules.

[0094] Table 1 preset common sense rule representation example

[0095]

[0096] Referring to Table 2, which is a preset mapping rule table, the robot parses the social behavior parameters, i.e. the moving speed parameter and the social distance parameter, according to the acquired social attitude of the current social object, so as to calculate the actual moving speed and the actual social distance when the current social behavior is performed to the current social object.

[0097] Table 2 preset mapping rule representation example

[0098]

[0099] Referring to Table 3, which is a preset behavior parameter update table, the robot updates the current moving speed parameter and the social distance parameter according to the facial feedback and the voice feedback of the current social object, and further changes the actual moving speed and the actual social distance when the robot performs the social behavior according to the preset behavior parameter update table.

[0100] Table 3 preset behavior parameter update representation example

[0101]

[0102] The robot social knowledge base self-updating method provided by the application constructs the social knowledge base through a graphical database, constructs the graphical social knowledge base with the social object as a node, performs knowledge reasoning through a preset common sense rule table, and supplements the social knowledge base; the behavior parameters when the social behavior is performed to the current social object are updated according to a preset behavior parameter update table through the facial feedback and the voice feedback of the current social object, the social knowledge base based on the current social object is self-updated, the social behavior parameters of the robot are more in line with the social preferences of the current social object, different preset rules are set, and the social interaction demand of the robot in different scenes is met.

[0103] The embodiment of the present application also provides a device for updating a knowledge base for robot socialization, comprising: a knowledge base construction module for constructing a basic knowledge base containing social objects, basic attributes, social attributes, social relationships, social behaviors and social attitudes; a knowledge base updating module for reasoning the social relationships between different social objects and the social attributes of each social object by a preset common sense rule table, and completing the basic knowledge base; a social attitude query module for querying the social attitude of a current social object from the completed basic knowledge base by a social behavior to be performed by a robot and the social attributes of the current social object; a social attitude analysis module for analyzing the social attitude of the current social object into a social behavior parameter according to a preset mapping rule table, and obtaining an actual moving speed and an actual social distance according to the social behavior parameter and a preset corresponding relationship; a feedback analysis module for performing a social behavior according to the actual moving speed and the actual social distance, and obtaining facial feedback and voice feedback of the current social object after the social behavior is performed by using an expression recognition algorithm and a text classification algorithm; and a social attitude updating module for updating the social attitude of the current social object in the basic knowledge base according to the facial feedback and the voice feedback, and obtaining a social attitude of the current social object according to a preset behavior parameter updating table.

[0104] The embodiment of the present application also provides a service robot, comprising: a depth camera for collecting facial data of a current social object; a sound sensor for collecting voice instructions issued by the current social object; the device for updating a knowledge base for robot socialization as described above, which is in communication connection with the depth camera and the sound sensor, and is used for identifying the current social object according to the facial data, obtaining a social attitude of the current social object, calculating an actual moving speed and an actual social distance, obtaining facial feedback according to the facial data, obtaining voice feedback according to the voice instructions, and updating the social attitude of the current social object in the basic knowledge base according to the facial feedback and the voice feedback according to a preset parameter updating table; and a driving device in communication connection with the device for updating a knowledge base for robot socialization, and used for driving the service robot to perform a social behavior according to the actual moving speed and the actual social distance. The service robot provided by the embodiment of the present application also comprises a display screen for providing a plurality of social behavior selection buttons, so that the current social object can select a social behavior, and so that the device for updating a knowledge base for robot socialization can obtain a social attitude of the current social object according to the social behavior. When the service robot is applied to different scenes, different preset common sense rule tables, preset mapping rule tables and preset behavior parameter rule tables suitable for the scenes are set for the service robot, so as to meet the service robot social interaction requirements in different scenes.

[0105] Those skilled in the art will appreciate that embodiments of the present application can be readily used as a method, a system or a computer program product. Accordingly, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) embodying computer readable program code.

[0106] The present application is described in reference to the flowchart illustrations and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams.

[0107] These computer program instructions can also be loaded into a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams.

[0108] Obviously, the above-described embodiments are only examples for clarity of description and are not limiting on the implementation. For those of ordinary skill in the art, other different forms and / or changes can be made on the basis of the above description. There is no need and it is also impossible to enumerate all the implementations. The obvious changes or modifications made on the basis of the above description are still within the scope of the present application.

Claims

1. A knowledge base self-updating method for robot socialization, characterized by, The application relates to a method for constructing a basic knowledge base of social objects, basic attributes, social attributes, social relationships, social behaviors and social attitudes. The method comprises the following steps: reasoning the social relationships between different social objects and the social attributes of each social object by a preset common sense rule table, and supplementing the basic knowledge base; querying the social attitude of a current social object from the supplemented basic knowledge base according to a social behavior to be executed by a robot and the social attribute of the current social object; analyzing the social attitude of the current social object into a social behavior parameter according to a preset mapping rule table, and acquiring an actual moving speed and an actual social distance according to the social behavior parameter and a preset corresponding relationship, which comprises the following steps: the social behavior parameter comprises a moving speed parameter and a social distance parameter; calculating the actual moving speed of the robot according to the moving speed parameter; calculating the actual social distance between the robot and the current social object according to the social distance parameter; executing the social behavior according to the actual moving speed and the actual social distance; recognizing and acquiring the facial feedback and the voice feedback of the current social object after the social behavior is executed by using an expression recognition algorithm and a text classification algorithm; and updating the social attitude of the current social object in the basic knowledge base according to the facial feedback and the voice feedback and according to a preset behavior parameter updating table. The basic knowledge base is constructed based on a Neo4j graph database. The construction of the basic knowledge base comprises the following steps: establishing a person node for all social objects; filling the person node with basic attributes including a name, a gender, a scene and an identity; filling the social relationships and the social attributes according to the basic attributes and a preset common sense rule table, wherein the social attributes comprise a plurality of social behaviors and a plurality of social attitudes corresponding to the plurality of social behaviors; and the social attitude comprises a social behavior parameter when a social behavior is executed on a current social object. The method for querying the social attitude of the current social object from the supplemented basic knowledge base according to the social behavior to be executed by the robot and the social attribute of the current social object comprises the following steps: acquiring facial data of the current social object by using a robot depth camera; analyzing the facial data by using a face recognition network constructed by MTCNN and FaceNet to acquire the basic attributes of the current social object; and querying the social attitude of the current social object from the supplemented basic knowledge base according to the basic attributes of the current social object and the social behavior to be executed by the robot. The method for querying the social attitude of the current social object from the supplemented basic knowledge base according to the social behavior to be executed by the robot and the social attribute of the current social object further comprises the following steps: if the current social object cannot be queried from the supplemented basic knowledge base, a person node is established for the current social object in the supplemented basic knowledge base. The method for acquiring the facial feedback of the current social object comprises the following steps: acquiring facial data of the current social object by using a robot depth camera; and analyzing the facial data by using an expression recognition network constructed by MTNCC and a convolutional neural network to acquire the facial feedback of the current social object. ， wherein, denotes a mobile speed parameter value, denotes a speed control signal issued by the dynamic window method in the local path planning algorithm, denotes a maximum mobile speed achievable by the robot; The method for acquiring the voice feedback of the current social object comprises the following steps: converting the voice of the current social object into voice text data by using a voice recognition algorithm. ， wherein, represents a social distancing parameter value, represents a minimum social distancing allowed by the robot; ​ ​ 2. The knowledge base self-updating method for robot socialization of claim 1, wherein, ​ 3. The knowledge base self-updating method for robot socialization of claim 2, wherein, ​ ​ ​ ​ ​ 4. The knowledge base self-updating method for robot socialization of claim 1, wherein, ​ ​ ​ ​ 5. The knowledge base self-updating method for robot socialization according to claim 4, wherein, ​ ​ 6. The knowledge base self-updating method for robot socialization of claim 1, wherein, ​ ​ ​ 7. The knowledge base self-updating method for robot socialization of claim 1, wherein, ​ ​ The text classification network constructed by the BERT model and the full connection network is used to identify the voice text data to obtain voice feedback of the current social object.

8. A knowledge base self-updating apparatus for robot socialization, characterized by, The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The social attitude analysis module analyzes the social attitude of the current social object into a social behavior parameter according to a preset mapping rule table, and obtains the actual moving speed and the actual social distance according to the social behavior parameter and a preset corresponding relationship. The social behavior parameter includes a moving speed parameter and a social distance parameter. The actual moving speed of the robot is calculated according to the moving speed parameter: wherein, represents the moving speed parameter value, represents a speed control signal issued by a dynamic window method in a local path planning algorithm, represents the maximum moving speed that the robot can reach; and the actual social distance between the robot and the current social object is calculated according to the social distance parameter: wherein, represents the social distance parameter value, represents the minimum social distance allowed by the robot; The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization.

9. A service robot, characterized in that The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-updating device for robot socialization. The application relates to a knowledge base self-up

Images