WO2019169854A1

WO2019169854A1 - Human-computer interaction method, and interactive robot

Info

Publication number: WO2019169854A1
Application number: PCT/CN2018/106780
Authority: WO
Inventors: 乔倚松; 吴海周
Original assignee: 南京阿凡达机器人科技有限公司
Priority date: 2018-03-09
Filing date: 2018-09-20
Publication date: 2019-09-12
Also published as: CN108363492B; CN108363492A

Abstract

A human-computer interaction method, and an interactive robot. The method comprises: when a robot detects a user needing active interaction, acquiring user comprehensive information (S1), wherein the user comprehensive information comprises personal information of a current user, and environment information of a current robot system; generating active interaction content matching the user comprehensive information (S2); and performing active interaction with the user according to the active interaction content (S3). During the process of a robot performing active interaction, active personalized human-computer interaction can be achieved according to different user information of different users and current environment information.

Description

Human-computer interaction method and interactive robot

The present application claims priority to Chinese Patent Application No. 201101193098.5, entitled "A Human-Machine Interaction Method and Interactive Robot", the entire contents of which are incorporated herein by reference.

Technical field

The invention relates to the field of human-computer interaction, in particular to a human-computer interaction method and an interactive robot.

Background technique

Robot is an emerging comprehensive discipline developed in recent decades. It concentrates on the latest research results in various disciplines such as mechanical engineering, electronic engineering, information science, automatic control and artificial intelligence. It is one of the most active research fields in science and technology. . With the development of science and technology, service robots have been widely used.

For service robots, a good human-computer interaction experience is the key to the service performance of service-oriented robots, and it is also the most basic user demand for robots. At present, mainstream service robots generally have a certain degree of human-computer interaction. The usual human-computer interaction includes PC-type mouse-and-key human-computer interaction, touch-sliding human-computer interaction in tablet mobile phones, and voice human-computer interaction. Among them, the human-computer interaction mode of the voice interaction class has become one of the most important human-computer interaction modes of service-oriented robots due to the convenience and nature of the interaction mode and the low cost of interactive learning.

The mainstream voice human-computer interaction mode mainly uses passive voice activation mode to activate the whole human-computer interaction process. The robot continuously listens to the user's voice command and starts to perform voice recognition after receiving a specific voice command. Identify the specific content and give the user the corresponding answers and feedback. However, such human-computer interaction is relatively passive, the robot does not actively communicate with the user, and the attraction to the user is not strong enough; and the interactive content of the interaction is relatively rigid, not flexible enough, and the answer is different for different people. The content is the same, not personalized enough. Will reduce the user's experience.

Therefore, in order to solve the above drawbacks, the present invention provides an active personalized human-computer interaction.

Summary of the invention

The object of the present invention is to provide a human-computer interaction method and an interactive robot. In the process of actively interacting with a robot, a personalized interactive interaction can be realized according to different user information of different users and combined with information of the current environment.

The technical solution provided by the present invention is as follows:

The invention provides a method for human-computer interaction, comprising the steps of: S1 acquiring user comprehensive information when the robot detects a user who needs active interaction, the user comprehensive information including the personal information of the user, and the current robot system Environment information; S2 generates active interaction content that matches the user comprehensive information; S3 actively interacts with the user according to the active interaction content.

Preferably, before the step S1, the step includes: S0 setting an application scenario of the robot that matches the environment information according to the environment information of the current robot, where the application scenario includes an interaction rule of the robot interaction and an interaction used in the interaction. Interactive resources.

Preferably, step S2 specifically includes: S20, according to the interaction rule, acquiring an interaction resource that matches the user comprehensive information, thereby generating active interaction content.

Preferably, the interaction resource includes: voice content, action content, or multimedia content; the interaction rule includes multiple rule nodes, and each rule node embodies a mapping relationship between different user comprehensive information and different interaction resources.

Preferably, the personal information of the current user includes: gender, age, expression, face angle, face space location, number of face appearances, user name, number of faces currently detected, and voice information; The environmental information of the robot system includes: time, location, temperature, weather, network connection status, system language.

The present invention also provides an interactive robot, comprising: an information acquisition module, configured to acquire user comprehensive information when the robot detects a user who needs active interaction, the user comprehensive information including personal information of the user And the current environment information of the robot system; the processing module is electrically connected to the information acquisition module, and configured to generate active interaction content that matches the user comprehensive information; and the interaction module is configured to use the active interaction content and the user according to the Make active interactions.

Preferably, the scene setting module is further configured to set an application scenario of the robot that matches the environment information according to the environment information of the current robot, where the application scenario includes an interaction rule of the robot interaction and an interaction resource used in the interaction.

Preferably, the processing module is further configured to acquire, according to the interaction rule, an interaction resource that matches the user comprehensive information, thereby generating active interaction content.

The human-computer interaction method and the interactive robot provided by the invention can bring at least one of the following beneficial effects:

1. In the present invention, the interaction mode of the robot is different from the current passive interaction. After the robot recognizes the user, the robot can actively interact with the user, thereby attracting the customer to participate in the interaction and improving the interaction experience. At the same time, in the interaction, the robot acquires the personal information of the current user and the environmental information of the current robot system, and comprehensively forms the active interactive content, so that the active interactive content can better conform to the current environment and the personal characteristics of the user, and the user is more integrated into the human. In the machine interaction, improve the experience of human-computer interaction.

2. After the robot recognizes the current user, the user can obtain the personal information of the user, such as age, gender, expression, etc., and obtain the current facial expression of the user. If the user speaks, the voice information will also be obtained. This information is combined to form personalized user information. The robot can obtain the current date, time, location of the robot, weather and other environmental information through the network system or the current system of the robot. The robot will generate corresponding active interactive content based on the comprehensive information of the user with the user's personality, so that the interaction is closer to the user and the intelligence of the interaction is improved.

DRAWINGS

The preferred embodiment will be described in a clear and easy-to-understand manner with reference to the accompanying drawings, and a method for human-computer interaction and the above-mentioned characteristics, technical features, advantages and implementation manners of the interactive robot will be further described.

1 is a flow chart of an embodiment of a method for human-computer interaction of the present invention;

2 is a flow chart of another embodiment of a method for human-computer interaction of the present invention;

3 is a flow chart of still another embodiment of a method for human-computer interaction of the present invention;

4 is a schematic structural view of an embodiment of an interactive robot according to the present invention;

Description of the reference numerals:

1-Scenario setting module, 2-information acquiring module, 3-user integrated information module, 4-processing module, 41-matching sub-module, 42-interactive content generating sub-module, 5-interaction module.

Detailed ways

In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the specific embodiments of the present invention will be described below with reference to the accompanying drawings. Obviously, the drawings in the following description are only some embodiments of the present invention, and those skilled in the art can obtain other drawings according to the drawings without obtaining creative labor, and obtain Other embodiments.

In order to simplify the drawings, only the parts related to the present invention are schematically shown in the drawings, and they do not represent the actual structure of the product. In addition, in order to make the drawings simple and easy to understand, components having the same structure or function in some of the figures are only schematically illustrated, or only one of them is marked. In the present context, "a" means not only "only one" but also "more than one".

The present invention provides an embodiment of a human-computer interaction method, as shown in FIG. 1 , including:

S1, when the robot detects a user who needs active interaction, acquiring user comprehensive information, where the user comprehensive information includes personal information of the current user, and environment information of the current robot system;

S2 generates active interactive content that matches the user comprehensive information;

S3 actively interacts with the user according to the active interaction content.

Preferably, the step S1 includes the step of: S0 setting an application scenario of the robot that matches the environment information according to the environment information of the current robot, the application scenario includes an interaction rule of the robot interaction and an interaction used during the interaction. Resources.

In this embodiment, the application scenario includes interaction rules and interaction resources required for interaction when the robot interacts in the scenario. The user customizes the customized user scene by selecting or based on the webpage, application and other platforms, and deploys it to the robot in real time, so as to avoid the change of the system level, the robot can satisfy the rapid application in different environments.

Preferably, the step S2 generating the active interaction content that matches the user comprehensive information specifically includes: S20 acquiring, according to the interaction rule, an interaction resource that matches the user comprehensive information, thereby generating active interaction content.

Specifically, before the robot is used, the user can set an application scenario that is more in line with the current use environment according to the scenario of the robot application, and can better conform to the current environment in the process of human-computer interaction, so that the user is more integrated into the human-computer interaction, thereby improving The experience of human-computer interaction. Application scenarios include interaction rules and interaction resources. If the robot is used in a mall, it will increase the shopping-related interaction rules and interaction resources. For example, the interaction rule is to ask the user which products to purchase when the robot recognizes the user, or to ask the user mall which services are not enough, etc.; the interactive resources can set cheerful songs, or clapping hands, squatting actions, in the user When responding to purchased goods, you can make a clap action and encourage users to continue spending.

If the robot is used in a hospital, it will increase the interaction rules and interaction resources related to medical treatment, drugs, and epidemic prevention. For example, when the robot recognizes the user's sadness, the user will be asked why the user is not happy, and comfort the user, while making the action of patting the user's shoulder and playing some cheerful music; some cheerful music can be set in the interactive resource. Shoulder, refueling and other actions.

When the robot recognizes the user, it will obtain the current user information through face recognition, and obtain the environment information of the robot through the internal system of the robot, and finally form the user comprehensive information with the user's personalization. Then the robot will get the current interactions according to the interaction rules in the application scenario, retrieve the corresponding interaction resources in the interaction resources, form active interaction content, and actively interact with the user. Different from the current interaction mode, many current interaction methods are passive interactions. The robot continuously listens to the user's voice commands, and after receiving a specific voice command, starts voice recognition, and performs for the user according to the specific content of the recognition. Give the corresponding answers and feedback. Such human-computer interaction is relatively passive, and the robot does not actively communicate with the user, which is not attractive to the user. In the present invention, after detecting the user, the robot can form the active interactive content with the user personalized through the obtained comprehensive information of the user, and actively interact with the user, thereby attracting the user to participate in the human-computer interaction.

For example, if the robot is used in a mall, the user can add shopping-related interaction rules and interaction resources. The interaction rule is: after the robot recognizes the user, first identifies the user, obtains the current user information and the environment information, and if it is an adult, it will ask the user which products to purchase and respond, and then ask the user mall which services are not enough. In place, respond; finally say goodbye to the user. If it is a child, it will first greet the children and dance for the children.

The user can set a cheerful song in the interactive resource, or an action of clap, slap, wave, squat, etc., when the user responds to the purchased product, can make a clap action, and encourage the user to continue to consume.

When the user identified by the robot is a lady, the current time is 10:30 am, the robot will say: "Mrs. Good morning! What products did you buy today?" At the same time, make a wave of gestures to say hello to the lady. The lady responded: “Buy a piece of clothing.” The robot will respond: “Thank you for your support of the mall!” At the same time, he made awkward moves. Then continue to say: "There are not enough services in the mall," and then record what the user said. Finally, say "goodbye" to the user.

When the user recognized by the robot is a little girl, the current time is 5 pm, the current little girl's expression is a smiley face, the robot will say: "Good afternoon, children! How are you so happy?" At the same time, make a gesture of squatting, Keep the same height as the little girl. The little girl responded: "Mom bought me a new dress today." The robot will respond: "I will dance for you" and play dance music.

It can be seen that the interaction mode of the robot is also very different due to the different user comprehensive information embodied by the user. In the prior art, the active interaction content of the robot for each user is the same, after the user is identified. It will say: "Hello, what can I help you?" Unlike the present invention, it is possible to output personalized active interactive content according to different characteristics carried by different users.

The above interaction rules and interaction resources can be set by themselves. The robot will find corresponding responses in the interaction rules according to the user comprehensive information, and generate active interaction content in combination with the interaction resources to perform active interaction. It can be seen that the comprehensive information of users of different users is different. The robot generates different active interactive content according to different genders, different expressions and different ages of different users, so that the active interactive content is personalized.

As shown in FIG. 2, the present invention also provides an embodiment of a method for human-computer interaction, including:

S0 sets an application scenario of the robot that matches the environment information according to the environment information of the current robot, and the application scenario includes an interaction rule of the robot interaction and an interaction resource used in the interaction.

S201: filtering, in the interaction rule, a plurality of candidate rule nodes that match the user comprehensive information;

S202: Generate active interactive content according to the interaction resources corresponding to the plurality of candidate rule nodes; the active interaction content includes voice interaction content, action interaction content, and multimedia interaction content.

The interaction resource includes: voice content, action content, or multimedia content; the interaction rule includes multiple rule nodes, and each rule node embodies a mapping relationship between different user comprehensive information and different interaction resources.

The interaction rule includes: whether the robot responds to certain specific detection results when interacting, for example, whether to query the user name, whether to detect the system time, etc.; and when the detection result of the response is selected, the specific content of the corresponding feedback is selected. For example, different names for different genders, different actions for different users, and so on.

The interaction resources include: resources required under corresponding interaction rules, such as all voice text content, all optional action interaction content, all music and video content, and the like.

The current user's personal information includes: gender, age, expression, face angle, face space location, face appearance times, user name, current detected face number, and voice information; the current robot system Environmental information includes: time, location, temperature, weather, network connection status, system language.

Specifically, in this embodiment, the system performs various identification and detection by human body recognition, face recognition, environment system detection, etc., and comprehensively outputs a set of user information descriptions for the user to describe the specific information of the current interactive user. The user's comprehensive information form is expressed as follows:

Head;key1:value1;key2:value2;key3:value3;...;

Where head is fixed, the string identifier of the string identifies the specific content of the string to distinguish it from other types of strings. The key is a feature in the user comprehensive information, that is, a user feature keyword and an environmental feature keyword, and each key represents a feature used to describe the current user, and the features may include: face number, face name, gender, Age, time, expression, face angle, face position, face size, weather, location, temperature, type of exercise, network connection, etc. The value value is a specific parameter value corresponding to the current key value, that is, the user characteristic parameter value and the environmental characteristic parameter value described in the present invention. The <key:value> pair can modify the number and content of the user description features according to different human body recognition, face recognition, system detection, motion detection and other detection and recognition tool output changes.

Simple user comprehensive information is as follows:

Example 1:

Rvn;ultrasound;event;payload;type:passby;

Where rvn;ultrasound;event;payload; as a string header, indicating that the string is a description of the user information containing the ultrasonic sensor. The description of the article is relatively simple and only indicates that the robot has sensed that someone has passed through it through the ultrasonic sensor.

Example 2:

Rvf;vison;event;face;payload;type:face;stop:yes;name:avatar;gender:masculine;age:53;time:13.04;emotion:none;roll:-13.61209;pitch:23.196611;yaw:23.330135 ;faceID:1;number:1;sequence:1;px:646;py:189;pw:352;ph:352;

Where rvf;vison;event;face;payload; as a string header, indicating that the string is a comprehensive information description of the user containing the visual sensor information. The key, value of each pair represents an information feature of the user. The specific can be interpreted as: the user's face information is continuous; user name: avatar; gender: male; age: 53 years old; record generation time: 13:04; user face expression: no; face angle roll value :-13.61209 degrees; face angle pitch value: 23.196611 degrees; face angle yaw value: 23.330135 degrees; face record number: 1; the number of faces in the current picture; 1; the user's face is in the total person The face is the first one; the face position X value: 646px; the face position Y value: 189px; the face width: 352px; the face length: 352px;

Different users, different environments, different levels of familiarity, and different detection and identification tools generate different user description results. The comprehensive description of these users is a personalized user description. Through the analysis of these different descriptions, the system generates active interactive content corresponding to the rules and resources of the current scene.

The method generates a set of active interactive content corresponding to the input user comprehensive information. The active interactive content of the feedback includes the following three types: voice interactive content, action interactive content, and multimedia interactive content. The voice interactive content is an active voice prompt played by the robot; the action interactive content is a set of sports content of the active part of the head, the limbs, etc.; the multimedia interactive content includes pictures, music, video, application, etc., and passes through the chest of the robot. The display platform is played. The content of the multimedia can be played simultaneously with the voice prompt, or can be played after the voice prompt ends, to meet the needs of different scenes.

The rule node includes a Node node, and the interaction rule stores the voice, action, and multimedia interaction content corresponding to each recognition result in a tree-shaped data structure. The interaction rule tree includes a plurality of Node nodes, each of which includes a plurality of preset feature keywords and corresponding preset feature parameter values, and also includes multiple interaction resources such as voice, action, and multimedia. .

The Key value in the Node node describes the necessary conditions for the set of statements, actions, and multimedia to be selected. First, when there is a comprehensive information of a user, each group of Node nodes will match the currently integrated user information. If the current user comprehensive information is satisfied with the necessary conditions of the Node node, the Node node will become a candidate. Node, wait for later selection. If the current user comprehensive information and the necessary conditions of the Node node are not completely satisfied, the Node node will not become a candidate rule node.

As shown in FIG. 3, the present invention further provides an embodiment of a method for human-computer interaction, including:

S10: when the robot detects a user who needs active interaction, acquires user comprehensive information, and assigns a corresponding user feature keyword and a user feature parameter value to each user feature in the current user's personal information, and gives the current robot Each environmental feature in the environmental information of the system is assigned a corresponding environmental feature keyword and an environmental feature parameter value.

S2011 determines whether each rule node meets a preset condition;

If S2012 is, the rule node that satisfies the condition is used as the candidate rule node;

S2021 analyzes a priority value of each of the candidate rule nodes, and sorts the candidate rule nodes according to the priority value; and randomly selects one candidate rule node randomly or weighted for multiple candidate rule nodes of the same priority value. Participate in sorting;

S2022 combines the interaction resources corresponding to the sorted candidate rule nodes in sequence to generate the active interaction content.

Specifically, in this embodiment, the preset condition is determining whether all preset feature keywords and corresponding preset feature parameter values in each rule node are related to partial feature keywords and corresponding features in the user comprehensive information. The parameter values are the same. The feature keyword includes a user feature keyword and an environment feature keyword, and the feature parameter value includes a user feature parameter value and an environment feature parameter value. It is assumed that the feature keyword is a key value, the feature parameter value is a value corresponding to the key value, and the rule node is a Node node, that is, a specific step of filtering out a plurality of candidate rule nodes that match the user comprehensive information according to a preset condition: If the Value value of all the features in the Node node is equal to the Value value of the same feature in the user comprehensive information, the Node node is used as the candidate rule node. If the value of the feature in the Node node is All, it means that the feature is considered to be satisfied for the Value result of the corresponding feature in all user information. The features in the user's comprehensive information are usually more than the feature values required by the Node node. For the extra feature values, the system does not judge and filter based on the results.

While matching the candidate Node nodes, for a Node node that becomes a candidate node, a complete voice prompt will be combined according to the value of its priority value Priority. The method decomposes a complete speech interaction content into different sentence segments, and each sentence segment is a segment of a complete voice prompt. The priority value Priority in each Node node indicates the position of the statement segment in the complete statement.

Simply, we can divide a complete statement into multiple statement segments, such as a title segment, a time segment, a content segment, and a problem segment. The invention does not limit the number of segments of the statement, and the user can segment the segment according to the completeness of the statement. There is a free combination between each paragraph and the next paragraph. Therefore, the final combination of the complete statement will become very flexible. For multiple candidate nodes in the same location, the method selects a Node node that satisfies the condition by random selection.

For example, the voice interaction content to be generated includes a calling segment, a greeting segment, and a content segment. The calling segment has two Node nodes with the same priority value, namely “old man” and “old man”; the greeting segment has three Node nodes with the same priority value. "Hello!" "Good morning!" "Good morning!" The content segment has a Node node, which is "You are in good health." For the title segment and the greeting segment, you can randomly select one of the content to participate in the voice message. Sort. It can be seen that there are six combinations of voice message content. Under the same conditions, the content of the robot's voice interaction will constantly change, and it will not be very rigid and affect the user experience.

In some Node nodes, a plurality of Item options (ie, content executed by the robot) may be set in advance, and for a plurality of Item options existing in the Node node, a final result that satisfies the condition is selected according to the difference of the Key values thereof. If the same Key value result has multiple corresponding Item options, the method selects a result as a final result by sequential selection or random selection, and outputs the result.

Since the content of the interaction depends on the combination of the candidate rule nodes and the selection of the Item option in each candidate rule node, in the same scenario, the candidate rule nodes of the same priority value only randomly select one participation order, for each candidate rule. A node, whose multiple item options corresponding to the key value can be randomly, weighted randomly, or selected in a certain order, the combination is very much. In the interaction process, the content of the interaction is different for the same scenario, and the fixed mode interaction is not very rigid.

For example, when greeting, the action corresponding to the key value may be three item options: handshake, wave, and salute. In this case, one of the items may be selected for output. When you say hello, the interaction of the robot is not too rigid because the action is too single. Similarly, the greeting language can be set to multiple, assigned to different Item options. When an item is randomly selected, the language of the output will be different, and the content will be diversified when interacting.

The action interaction content and the multimedia interaction content exist by being appended to each specific voice sentence. By defining additional actions and multimedia interactions after each specific statement, when the active feedback content combination is completed, the corresponding actions and multimedia interaction content are also generated. The action and multimedia interaction content is determined by the additional content of the last statement component. Therefore, different content, different lengths of statements can also make different actions and prompts on multimedia interactive content. Flexible use of voice scripts and resource scripts, as the user's comprehensive information input, the feedback information generated by this method can also be changed accordingly.

Through the face recognition method, combined with the personal information in the Internet, some personal information of the user, such as basic information such as name, age, and gender, can be obtained. This embodiment will illustrate an interactive system of the present invention by way of an example.

For example, when the robot recognizes that there is a crying boy in front, the collected user comprehensive information may include: male, 8 years old, name, facial expression crying, the current time is 10 am, the current weather is 2 degrees Celsius, the current location is In the hospital, the air quality is good, and after the value is assigned, a string is formed for the system to retrieve. The robot finds the candidate rule nodes that satisfy the matching condition in the rule node according to the user comprehensive information, sorts the rule nodes according to the priority values, and sequentially combines the interaction resources corresponding to the candidate rule nodes. After the robot recognizes the little boy, it will greet the little boy actively. When you say hello, there may be a lot of Item values corresponding to the key value in the Node node (the voice interaction content is “Hello Kids!”, “Hello. "Little, handsome boy!" "hi, children", the action interaction content is wave, handshake, swear, salute) the robot can only choose one of the interactive content in the voice interactive content or action interaction content, therefore, The active interactions of each output are constantly changing and not too rigid.

As shown in FIG. 4, the present invention provides an embodiment of an interactive robot, including:

The information obtaining module 2 is configured to acquire current user information and environment information;

The user integrated information module 3 is electrically connected to the information acquiring module 2, and is configured to generate user comprehensive information according to the current user information and environment information;

The processing module 4 is electrically connected to the user integrated information module 3, and configured to generate active interactive content according to the user comprehensive information and the application scenario;

The interaction module 5 is configured to perform active human-computer interaction according to the active interaction content.

Preferably, the interactive robot further includes: a scene setting module 1 configured to set an application scenario of the robot that matches the environment information according to the environment information of the current robot, where the application scenario includes an interaction rule of the robot interaction and The interactive resources used in the interaction.

Preferably, the scene setting module 1 is further configured to store a preset interaction rule of the robot, and an interaction resource required when the robot interacts under the interaction rule, and use the interaction rule and the application resource as An application scenario; the interaction rule includes multiple rule nodes, and each rule node includes multiple preset feature keywords and corresponding preset feature parameter values.

Preferably, the user integrated information module 3 is further configured to assign a corresponding user feature keyword and a user feature parameter value to each user feature in the current user information, and assign a corresponding to each environment feature in the environment information. The environment feature keyword and the environment feature parameter value; the user integrated information module 3 is further configured to combine the user feature keyword and the corresponding user feature parameter value, the environment feature keyword, and the corresponding environment feature parameter value into a string And assigning the string to the corresponding string identifier; using the string with the string identifier as the current user comprehensive information.

Preferably, the processing module 4 specifically includes: a matching sub-module 41, configured to filter, in the application scenario, a plurality of candidate rule nodes that match the user comprehensive information; and an active interaction content generation sub-module 42 according to The interaction resources corresponding to the plurality of candidate rule nodes generate active interaction content; the active interaction content includes voice interaction content, action interaction content, and multimedia interaction content.

Preferably, the matching sub-module 41 is further configured to determine, by one by one, whether all preset feature keywords and corresponding preset feature parameter values in each rule node are corresponding to some feature keywords in the user comprehensive information. The feature parameter values are the same; the feature keyword includes a user feature keyword and an environment feature keyword, and the feature parameter value includes a user feature parameter value and an environment feature parameter value; if yes, the rule node that satisfies the condition is used as a candidate rule node.

Preferably, the active interaction content generation sub-module 42 is further configured to analyze respective priority values of the plurality of candidate rule nodes, and sort the candidate rule nodes according to the priority value; for the same priority value The candidate rule nodes select one of the candidate rule nodes to participate in the sorting; and the interactive resources corresponding to the sorted candidate rule nodes are sequentially combined to generate the active interactive content.

Specifically, the information obtaining module 2 has two major functions, one is to identify the current user, and obtain the user resource; for example, when it is detected that there is a user within the preset range, the user is initially identified, and the user mainly recognizes the user's face. feature. Through the face recognition technology, the current user's expression can be identified, and combined with the Internet big data, some basic data information of the user is obtained. The second is to obtain environmental information such as the current location, time, weather, etc. of the current robot according to the current robot system.

The processing module 4 may be composed of a processor of a robot, and the interaction module 5 includes a voice control system, a display system, a drive system, and the like used in the robot interaction process. For the process of the robot interaction, reference may be made to the foregoing method embodiment, and details are not described herein again.

It should be noted that the above embodiments can be freely combined as needed. The above description is only a preferred embodiment of the present invention, and it should be noted that those skilled in the art can also make several improvements and retouchings without departing from the principles of the present invention. It should be considered as the scope of protection of the present invention.

Claims

A method for human-computer interaction, comprising the steps of:

S1, when the robot detects a user who needs active interaction, acquiring user comprehensive information, where the user comprehensive information includes personal information of the user, and environment information of the current robot system;

S2 generates active interactive content that matches the user comprehensive information;

S3 actively interacts with the user according to the active interaction content.
The method of human-computer interaction according to claim 1, wherein the step S1 comprises the steps of:

S0 sets an application scenario of the robot that matches the environment information according to the environment information of the current robot, and the application scenario includes an interaction rule of the robot interaction and an interaction resource used in the interaction.
The method of human-computer interaction according to claim 2, wherein the step S2 specifically includes:

S20 acquires, according to the interaction rule, an interaction resource that matches the user comprehensive information, so as to generate active interaction content.
A method of human-computer interaction according to claim 3, wherein:

The interaction resource includes: voice content, action content, or multimedia content; the interaction rule includes multiple rule nodes, and each rule node embodies a mapping relationship between different user comprehensive information and different interaction resources.
A method for human-computer interaction according to any one of claims 1 to 4, characterized in that:

The personal information of the user includes: gender, age, expression, face angle, face space location, number of face appearances, user name, number of faces currently detected, and voice information;

The environmental information of the current robot system includes: time, location, temperature, weather, network connection status, system language.
An interactive robot characterized by comprising:

An information obtaining module, configured to acquire user comprehensive information when the robot detects a user who needs active interaction, where the user comprehensive information includes personal information of the user, and environment information of the current robot system;

The processing module is electrically connected to the information acquiring module, and configured to generate active interactive content that matches the user comprehensive information;

The interaction module is configured to actively interact with the user according to the active interaction content.
The interactive robot according to claim 6, further comprising:

The scenario setting module is further configured to set an application scenario of the robot that matches the environment information according to the environment information of the current robot, where the application scenario includes an interaction rule of the robot interaction and an interaction resource used in the interaction.
An interactive robot according to claim 7, wherein:

The processing module is further configured to acquire, according to the interaction rule, an interaction resource that matches the user comprehensive information, thereby generating active interaction content.
An interactive robot according to claim 8 wherein:

The interaction resource includes: voice content, action content, or multimedia content; the interaction rule includes multiple rule nodes, and each rule node embodies a mapping relationship between different user comprehensive information and different interaction resources.
An interactive robot according to any one of claims 6-9, characterized in that:

The personal information of the current user includes: gender, age, expression, face angle, face space location, number of face appearances, user name, number of faces currently detected, and voice information;

The environmental information of the current robot system includes: time, location, temperature, weather, network connection status, system language.