Interactive dialogue control method and system for embodied intelligent robot

Abstract

The application discloses an interactive dialogue control method and system for embodied intelligent robots, and relates to the field of control of embodied intelligent robots, S1: a rule-based dialogue model is established; S2: fuzzy matching is performed on the rules; S3: control of an interactive dialogue process; S4: selection of a dialogue strategy, including an active dialogue strategy, an adaptive dialogue strategy and a multi-mode information dialogue strategy, the selection of the dialogue strategy being based on the rules of the dialogue model. The interactive dialogue control method and system for embodied intelligent robots can guide the process and reply of dialogue through pre-defined rules, have high explainability and controllability, and can control the dialogue process, and through combination of the rules, a complete dialogue process can be constructed, and based on the interactive dialogue content of a user, matching of the combined rules can be directly performed, so that a rule meeting the conditions can be directly found and corresponding actions are performed.

Description

Technical Field

[0001] This invention relates to the field of control of embodied intelligent robots, and particularly to an interactive dialogue control method and system for embodied intelligent robots. Background Technology

[0002] Embossed intelligent robots are a new type of robot that integrates knowledge from multiple disciplines such as artificial intelligence, robotics, and biology. They have the ability to perceive their bodies, make autonomous decisions, and act, and can interact with their environment and complete various tasks.

[0003] When embodied intelligent robots engage in interactive dialogue, the content of their dialogue can feel fragmented in different application scenarios, affecting the user's experience. Furthermore, the embodied intelligent robot cannot adopt different interactive dialogue methods based on the current user group, which can negatively impact its practicality.

[0004] Therefore, it is necessary to propose interactive dialogue control methods and systems for embodied intelligent robots to solve the above problems. Summary of the Invention

[0005] The main objective of this invention is to provide an interactive dialogue control method and system for embodied intelligent robots, which can effectively solve the problems in the background art.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: An interactive dialogue control method for embodied intelligent robots includes the following operational steps: S1: Establish a rule-based dialogue model, analyze the application scenarios of embodied intelligent robots, organize the user's needs and common problems based on the application scenarios, convert the needs and problems into rules, extract the converted rules, and finally store them in the rule base. S2: Perform fuzzy matching on the rules; S3: Control of interactive dialogue flow; S4: Selection of dialogue strategies, including proactive dialogue strategies, adaptive dialogue strategies, and multimodal information dialogue strategies. The selection of dialogue strategies is based on the rules of the dialogue model.

[0007] Preferably, in step S1, rule extraction is performed both automatically and manually, including collecting dialogue data, analyzing the dialogue data, and understanding users' common problems, needs, and expression methods based on text mining technology. When extracting rules manually, domain experts from various application scenarios manually extract rules from the collected dialogue data. When extracting rules automatically, similar dialogue data is automatically grouped based on clustering algorithms, and keywords from the groups are extracted as rules. The grouped rules are then organized, including the following steps: S101: Classify rules according to their function, application scenario, question type, and dialogue flow; S102: Optimize and organize the extracted rules, check the syntax and logic of the rules; simplify and summarize the rules; test and verify the rules; S103: Use visualization tools to display rules and create rule documents. Visualization tools include flowcharts, decision trees, tables, and lists.

[0008] Preferably, the process of judging and classifying the expression methods involves an embodied intelligent robot interacting with the user based on their dialogue expression habits. The expression methods are categorized into direct statements, questions, examples, metaphors, and analogies, and rules for expression methods are established based on these rules. When the user's dialogue is expressed in the form of direct narration, the embodied intelligent robot directly listens to the content of the user's dialogue and makes corresponding dialogue and action responses. When a user asks a question, the embodied intelligent robot interacts with the conversation based on the question asked. When a user's dialogue is expressed through examples, the embodied intelligent robot learns the dialogue content in conjunction with the current application scenario, makes judgments based on the examples given in the user's dialogue, and interacts with the user based on the judgment results. When user dialogue uses metaphors and analogies, the system learns the dialogue content in conjunction with the current application scenario, and extracts keywords based on the metaphors and analogies used in the user dialogue to understand the specific content of the user dialogue, and then interacts with the user accordingly.

[0009] Preferably, in step S1, the application scenario is determined based on the functions equipped by the embodied intelligent robot, and the target is determined based on the above-mentioned premise. The target includes children, parents, men, women, couples, children, and the elderly. When conducting interactive dialogue, the current application scenario is determined based on the dialogue data, and interactive dialogue is conducted with the corresponding user through the determined application scenario. Different interactive dialogue control methods are adopted when the embodied intelligent robot is in different application scenarios.

[0010] Preferably, in step S2, fuzzy matching includes: performing fuzzy matching based on edit distance, backing up the dialogue data, editing the distance of the backed-up content, setting an edit distance threshold for text data in a rule, and determining that the user's dialogue input matches the rule when the edit distance is less than the threshold. The embodied intelligent robot will then conduct corresponding interactive dialogue with the user. Editing includes inserting, deleting, and replacing text data. Otherwise, it does not match, and a new rule is generated and recorded. The fuzzy matching based on edit distance is used to handle spelling and grammatical errors in dialogue data; Fuzzy matching also includes: fuzzy matching based on text data similarity calculation, measuring the similarity between the user's dialogue input and the text data in the rules through cosine similarity, recording the content of the user's dialogue input and the text data in the rules as strings, representing them as vectors, calculating the cosine similarity between the two strings, and when the similarity is close, it is determined that the user's dialogue input matches the rules, and the embodied intelligent robot will conduct corresponding interactive dialogue with the user; The fuzzy matching based on text data similarity calculation is used to process semantically similar but not completely identical expressions in dialogue data; The specific formula for measuring cosine similarity is as follows: ; in The input vector for the user dialogue. This is a vector of text data in the rule. For vectors sum vector dot product, and Representing vectors respectively sum vector The model, This is the calculated cosine similarity. The value of cosine similarity ranges from -1 to 1. The closer the value is to 1, the stronger the similarity between the vectors. sum vector The greater the similarity, the more likely the user's dialogue input matches the rule; the closer the value is to -1, the stronger the vector. sum vector The more dissimilar the rules are, the more likely the user's dialogue input is to not match the rules, and a new rule will be generated and recorded.

[0011] Preferably, in step S3, the rules are combined to construct a complete interactive dialogue process. When the user has a dialogue with the embodied intelligent robot, the matching is performed based on the corresponding rules, and the corresponding operation is executed. Based on the application scenario of the embodied intelligent robot, the interactive dialogue process is divided into linear and branching. The linear process follows the order of the pre-combined rules, and the branching process selects different rules to conduct interactive dialogue based on the different conditions provided by the user during the interactive dialogue.

[0012] Preferably, the proactive dialogue strategy is a strategy in which the embodied intelligent robot proactively initiates dialogue and guides the user to engage in interactive dialogue, establishes proactive dialogue rules, and the embodied intelligent robot proactively provides relevant information to the user based on the user's behavior; The process of distinguishing between dangerous and normal user behavior includes the following steps: A: Data collection: Visual information data is provided through video monitoring. The visual information data uses target detection algorithms to detect the user's position and posture, and extract the human body's contour and joint position. Environmental information data is provided through sensor data. The environmental information data is extracted from the current environment using environmental sensors, including temperature and humidity sensors and hazardous gas sensors. By analyzing user dialogue data, we can identify any dangerous behavior data, analyze the dialogue data based on rules, and establish a database of dangerous dialogue texts. B: The judgment between dangerous and normal behaviors is based on the formula and the provided information data. The formula is as follows: ; in For behavioral frequency indicators, For behavioral intensity indicators, For environmental factor indicators, assign weights to each indicator. The total weight is 1. A risk threshold is set based on the application scenario and rules. When the value is less than or equal to the threshold, it is considered normal behavior; when... If the value exceeds the threshold, it is judged as dangerous behavior; C: The provided information and data are labeled, clearly indicating whether the information and data constitute normal or dangerous behavior. When a user engages in dangerous behavior, the embodied intelligent robot will proactively initiate a dialogue with the user.

[0013] Preferably, the adaptive dialogue strategy dynamically adjusts the dialogue strategy of the embodied intelligent robot based on user feedback and the progress of the dialogue.

[0014] Preferably, the multi-modal information dialogue strategy is based on the embodied intelligent robot utilizing information in multiple modes, including but not limited to voice, images, and gestures, to conduct corresponding interactive dialogues with the user.

[0015] An interactive dialogue control system for embodied intelligent robots, and a method for implementing interactive dialogue control of embodied intelligent robots, including: The rules module is used to formulate rules for the interactive dialogue control of the embodied intelligent robot, and to establish a dialogue model based on the rules module. It extracts the user's dialogue data, and judges the conditions and actions of the extracted dialogue data. The conditions are used to determine whether the user's dialogue data is within the working range of the embodied intelligent robot, and the actions are the response actions that the embodied intelligent robot will take when the working range is met. A rule base, which stores the rule content formulated for the embodied intelligent robot; The application scenario analysis and organization module is used to analyze the application scenarios used by the embodied intelligent robot, and to analyze and adjust the content and method of interactive dialogue based on the application scenarios obtained after analysis. The fuzzy matching module is used to respond to spelling errors, grammatical errors and expression errors generated by the user during interactive dialogue, and to conduct interactive dialogue based on the rules of high similarity of the content provided by the user. A dialogue strategy selection module, which uses different strategies to interact with the user based on information data collected by the embodied intelligent robot, the embodied intelligent robot including at least one video monitor and environmental sensors. The dialogue strategy selection module also includes a behavior judgment unit, which judges whether the user's behavior is normal or dangerous based on the collected information data, and conducts corresponding interactive dialogues with the user based on the behavior to reduce the user's dangerous behavior. The interactive dialogue flow control module is used to combine stored rules, classify the combinations into appropriate application scenarios, and sort the order of the combinations. When interacting with the user, the module interacts with the user based on the combinations, application scenarios, and order.

[0016] Compared with the prior art, the present invention provides an interactive dialogue control method and system for embodied intelligent robots, which has the following beneficial effects: This interactive dialogue control method and system for embodied intelligent robots establishes a rule-based dialogue model. When interacting with users, it can respond accordingly based on rules and extract text data through rules. It can also mine and understand users' questions and needs, and determine the current application scenario and goal to conduct corresponding interactive dialogues with users. It can select different interactive dialogue content based on children, parents, men, women, couples, and the elderly, making its application scope wide and well applicable to industrial manufacturing, logistics and warehousing, medical and health care, service industry, public safety, education, and family life.

[0017] This interactive dialogue control method and system for embodied intelligent robots can determine the user's dialogue goals during interactive dialogues and guide the user to provide the required information based on the goals, thereby better completing the interactive dialogue with the user.

[0018] This interactive dialogue control method and system for embodied intelligent robots can, under normal circumstances, judge the user's dangerous and normal behaviors, ensure the user's safety based on the collected information data, and label the calculated behavioral data. When labeled as dangerous behavior, it can actively engage in dialogue with the user to prevent the user from continuing to engage in dangerous behavior.

[0019] This interactive dialogue control method and system for embodied intelligent robots can guide the dialogue flow and responses through predefined rules, exhibiting high interpretability and controllability. It allows for control of the dialogue flow, and by combining rules, a complete dialogue process can be constructed. Based on the user's interactive dialogue content, the combined rules can be directly matched to find matching rules and execute corresponding actions. During interactive dialogue, users can clearly understand the embodied intelligent robot's decision-making process and response logic, ensuring the reliability and security of the interactive dialogue. It is also well-suited for specific fields, including medicine, law, and finance, guaranteeing the professional knowledge of the embodied intelligent robot during interactive dialogue and providing accurate and professional responses.

[0020] The interactive dialogue control method and system for embodied intelligent robots can effectively improve the naturalness of dialogue through fuzzy matching, allowing users to input less precise interactive dialogue to a certain extent, enabling the embodied intelligent robot to better understand the user's intentions. At the same time, it can increase the robustness of interactive dialogue, and can help the embodied intelligent robot to give corresponding answers to spelling errors and incomplete expressions in user interactive dialogue. Attached Figure Description

[0021] Figure 1 This is a flowchart of the overall method of the present invention; Figure 2 is the flowchart for judging the behavior of the present invention; Figure 3 is the system block diagram of the present invention. Detailed implementation manners

[0022] To make the technical means, creative features, achieved purposes and functions of the present invention easy to understand, the present invention will be further described below in conjunction with the specific implementation manners. Embodiment

[0023] As Figure 1 , Figure 2 shown, an interactive dialogue control method for an embodied intelligent robot, an interactive dialogue control method for an embodied intelligent robot, includes the following operating steps: S1: Establish a rule-based dialogue model, analyze the application scenarios of the embodied intelligent robot, and sort out the needs and common problems faced by the user based on the application scenarios, convert the needs and problems into rules, then extract the converted rules, and finally store them in the rule library; The extraction of rules is carried out automatically and manually, including collecting dialogue data, analyzing the dialogue data, and understanding the common problems, needs and expression ways of the user based on text mining technology, where the text mining technology includes: Convert the dialogue data into text data, extract the problems and needs therein, and select relevant data sources according to the specific problems and needs to ensure that the collected text data is targeted and effective. The relevant data sources are collected through previous user feedback, social media, online forums and questionnaires. Web crawlers, API interfaces and data scraping tools can also be used to collect text data from the selected data sources to ensure that the collected data has a standardized format and complete content; Text preprocessing, clean the collected text data, remove the noise in the text data, including HTML tags, special characters, punctuation marks, and correct the spelling mistakes and grammar mistakes in the text; Word segmentation processing, split the text data into individual independent words for subsequent analysis and processing; Stop word filtering, remove the stop words in the text, such as common meaningless words like "de", "shi", "zai".

[0024] Feature extraction, count the occurrence frequency of each word in the text data to understand the theme and keywords of the text; Use the Latent Dirichlet Allocation technology for topic modeling to extract potential topics and subjects from the text data; Sentiment analysis, perform sentiment analysis on the text data to understand the sentiment tendency expressed in the text, such as positive, negative or neutral.

[0025] Problem and requirement analysis, classifying problems in text data based on extracted features, and using machine learning algorithms to classify and predict problems in text data; Demand mining involves extracting user needs and expectations from text data based on the above content.

[0026] The expression methods are judged and classified. The embodied intelligent robot interacts with users based on their dialogue expression habits. The expression methods are divided into direct statement, question, example, metaphor and analogy. Based on this, the rules of expression methods are established. When the user's dialogue is expressed in the form of direct narration, the embodied intelligent robot directly listens to the content of the user's dialogue and makes corresponding dialogue and action responses. When a user asks a question, the embodied intelligent robot interacts with the conversation based on the question asked. When the user's dialogue is expressed through examples, the embodied intelligent robot learns the dialogue content in combination with the current application scenario, makes judgments based on the examples given by the user, and interacts with the user according to the judgment results. If the example is judged to be incorrect, the embodied intelligent robot interacts with the correct example. This can be used to teach users through the embodied intelligent robot. When users express themselves through metaphors and analogies, the system learns the content of the dialogue in conjunction with the current application scenario, and extracts keywords based on the metaphors and analogies used in the user's dialogue to understand the specific content of the user's dialogue, and then interacts with the user accordingly. When rules are extracted manually, domain experts from various application scenarios manually extract rules from the collected dialogue data. When rules are extracted automatically, similar dialogue data is automatically grouped based on clustering algorithms, and keywords in the groups are extracted as rules. The grouped rules are then organized, including the following steps: S101: Classify rules according to their function, application scenario, question type, and dialogue flow. When the content of one category is in the same domain as the content of another category, integrate them and classify them into the same group. S102: Optimize and organize the extracted rules, check the syntax and logic of the rules; simplify and summarize the rules; test and verify the rules; Remove duplicate rules: Check if there are duplicate rules in the rules; if so, keep one and delete the others. Simplify rules: For complex rules, simplify them by merging multiple similar conditions into a single, more concise condition, and by combining multiple actions into a single, more general action. Optimize rule order: Based on the importance and frequency of use of the rules, optimize the order of the rules, placing the most frequently used rules first, so that the embodied intelligent robot can respond to the rules at the beginning more quickly; S103: Use visualization tools to display rules and create rule documents. Visualization tools include flowcharts, decision trees, tables, and lists. Create a document for each rule, recording its purpose, applicable scenarios, and specific content. Use clear language to describe the rule in the document, avoiding vague and ambiguous expressions.

[0027] Application scenarios are determined based on the functions equipped by the embodied intelligent robot. These scenarios include industrial manufacturing, logistics and warehousing, medical and health care, service industry, public safety, education, and family life. Based on the above, targets are determined, including children, parents, men, women, couples, the elderly, and patients. During interactive dialogue, the current application scenario is determined based on the dialogue data, and interactive dialogue is conducted with the corresponding users according to the determined application scenario. Different interactive dialogue control methods are adopted when the embodied intelligent robot is in different application scenarios.

[0028] By establishing a rule-based dialogue model, when interacting with users, it can respond accordingly based on rules, extract text data through rules, and understand users' questions and needs. It can also determine the current application scenario and goal to conduct appropriate interactive dialogues with users. It can select different interactive dialogue content based on children, parents, men, women, couples, and the elderly, making its application scope wide and well applicable to industrial manufacturing, logistics and warehousing, medical and health care, service industry, public safety, education, and family life.

[0029] S2: Perform fuzzy matching on the rules. Fuzzy matching includes: performing fuzzy matching based on edit distance, backing up the dialogue data, editing the distance of the backed-up content, setting an edit distance threshold for the text data in a rule, and determining that the user's dialogue input matches the rule when the edit distance is less than the threshold. The embodied intelligent robot will then engage in corresponding interactive dialogue with the user. Editing includes inserting, deleting, and replacing text data. Otherwise, it will not match, and a new rule will be generated and recorded. Fuzzy matching based on edit distance is used to handle spelling and grammatical errors in dialogue data; Fuzzy matching also includes: fuzzy matching based on text data similarity calculation, measuring the similarity between the user's dialogue input and the text data in the rules through cosine similarity, recording the content of the user's dialogue input and the text data in the rules as strings, representing them as vectors, calculating the cosine similarity between the two strings, and when the similarity is close, it is determined that the user's dialogue input matches the rules, and the embodied intelligent robot will conduct corresponding interactive dialogue with the user; Fuzzy matching based on text data similarity calculation is used to handle semantically similar but not completely related expressions in dialogue data; The specific formula for measuring cosine similarity is as follows: ; in The input vector for the user dialogue. This is a vector of text data in the rule. For vectors sum vector dot product, and Representing vectors respectively sum vector The model, This is the calculated cosine similarity. The value of cosine similarity ranges from -1 to 1. The closer the value is to 1, the stronger the similarity between the vectors. sum vector The greater the similarity, the more likely the user's dialogue input matches the rule; the closer the value is to -1, the stronger the vector. sum vector The more dissimilar the rules are, the more likely the user's dialogue input is to not match the rules, and a new rule will be generated and recorded.

[0030] Fuzzy matching can effectively improve the naturalness of dialogue, allowing users to input less precise information in their interactions. This enables embodied intelligent robots to better understand the user's intentions and increases the robustness of interactive dialogue. It can also help embodied intelligent robots provide appropriate responses to spelling errors and incomplete expressions in user interactions.

[0031] S3: Control of interactive dialogue process, combining rules to construct a complete interactive dialogue process. When a user has a dialogue with the embodied intelligent robot, the system matches the combinations based on the corresponding rules and executes the corresponding operations. Based on the application scenario of the embodied intelligent robot, the interactive dialogue process is divided into linear and branching. The linear process follows the order of the pre-combined rules, while the branching process selects different rules to conduct interactive dialogue based on the different conditions provided by the user during the interactive dialogue. Before controlling the interactive dialogue process, the user's dialogue goals are determined. These goals include providing information, solving problems, completing tasks, and social activities. Based on this, the topic, scope, and process of the interactive dialogue are determined. Based on the established dialogue goals, explain to the user the purpose and process of the dialogue, and pose guiding questions to lead the user into the dialogue topic. Based on the user's interactive dialogue content, provide multiple rules, and proceed with the subsequent process after the user confirms them. Based on the goals and rules of the dialogue, design corresponding questions to gradually guide users to provide the required information. Use open-ended, closed-ended, and multiple-choice questions to obtain different types of information, provide feedback on the user's answers, confirm whether the understanding is correct, and ask follow-up questions as needed. It also includes dialogue state management, which locates the current interactive dialogue state, including the initial state, question answering state, and task execution state. Users can freely switch between states and make corresponding decisions based on the state. When in the question answering state, the system can select an appropriate answer to reply to the user's question. When in the task execution state, the system can perform corresponding interactive dialogue operations according to the progress of the task.

[0032] The dialogue flow and responses can be guided by predefined rules, offering high interpretability and controllability. The dialogue process can be controlled, and complete dialogue flows can be constructed by combining rules. Based on the user's interactive dialogue content, the combined rules can be directly matched to find matching rules and execute corresponding actions. During interactive dialogue, users can clearly understand the embodied intelligent robot's decision-making process and response logic, ensuring the reliability and security of the interactive dialogue. It is also well-suited for specific fields, including medicine, law, and finance, guaranteeing the professional knowledge of the embodied intelligent robot during interactive dialogue and providing accurate and professional responses.

[0033] S4: Selection of dialogue strategy, including proactive dialogue strategy, adaptive dialogue strategy and multimodal information dialogue strategy. The selection of dialogue strategy is based on the rules of the dialogue model. The proactive dialogue strategy is a strategy in which the embodied intelligent robot proactively initiates dialogue and guides users to engage in interactive dialogue. It establishes proactive dialogue rules, and the embodied intelligent robot proactively provides relevant information to users based on their behavior. The process of distinguishing between dangerous and normal user behavior includes the following steps: A: Data collection, providing visual information data through video monitoring, and using target detection algorithms to detect the user's position and posture, extracting the human body's contour and joint positions; Environmental information data is provided through sensor data. The environmental information data is extracted from the current environment by environmental sensors, including temperature and humidity sensors and hazardous gas sensors. By analyzing user dialogue data, we can identify any dangerous behavior data, analyze the dialogue data based on rules, and establish a database of dangerous dialogue texts. B: The judgment between dangerous and normal behaviors is based on the formula and the provided information data. The formula is as follows: ; in For behavioral frequency indicators, For behavioral intensity indicators, For environmental factor indicators, assign weights to each indicator. The total weight is 1. A risk threshold is set based on the application scenario and rules. When the value is less than or equal to the threshold, it is considered normal behavior; when... If the value exceeds the threshold, it is judged as dangerous behavior; C: The provided information and data are labeled, clearly indicating whether the information and data are normal or dangerous behaviors. When the user is engaging in dangerous behavior, the embodied intelligent robot will proactively initiate a dialogue with the user. The adaptive dialogue strategy dynamically adjusts the dialogue strategy of the embodied intelligent robot based on user feedback and the progress of the dialogue. When the user is not satisfied with the answer of the embodied intelligent robot, the embodied intelligent robot adjusts the content of subsequent interactive dialogue based on the corresponding content of the user feedback in order to meet the user's needs. The multimodal information dialogue strategy is based on embodied intelligent robots using multiple modes of information, including but not limited to voice, images and gestures, to conduct corresponding interactive dialogues with users.

[0034] Example 2: like Figure 3 As shown, the interactive dialogue control system for embodied intelligent robots includes: The rules module is used to formulate rules for the interactive dialogue control of the embodied intelligent robot, and to build a dialogue model based on the rules module. It extracts the user's dialogue data, and judges the conditions and actions of the extracted dialogue data. The conditions are used to determine whether the user's dialogue data is within the working range of the embodied intelligent robot, and the actions are the response actions that the embodied intelligent robot will take when the working range is met. The rules module includes a rule-setting unit, a reward unit, and a priority setting unit; The rule-making unit allows users to create rules for the embodied intelligent robot based on their own needs. After the rules are created, the embodied intelligent robot compares and judges the rules based on its own functions. If the rules are judged to be in line with the functions, the rule creation is completed. If the rules are judged to be in line with the functions, the embodied intelligent robot will interact and guide the user to create rules that are in line with the functions. The reward unit, based on the guidance of the embodied intelligent robot and the interactive dialogue with the user, will conduct praise-type interactive dialogue when the user performs specific behaviors, in order to improve the user's experience and enthusiasm for using the embodied intelligent robot. The specific behaviors include, but are not limited to, when the embodied intelligent robot guides the user to make corresponding answers, when the embodied intelligent robot teaches children, and when the user performs certain motor actions. The priority setting unit sets the priority of rules based on the content, application scenario, and user goal. When multiple rules are applicable at the same time, the embodied intelligent robot automatically determines the priority of the rules. The content, application scenario, and user goal of the rules can be combined with each other. For example, based on a single application scenario, the priority is determined according to the content of the rules and the user goal.

[0035] The rule base is used to store the rules defined for embodied intelligent robots.

[0036] The application scenario analysis and organization module is used to analyze the application scenarios used by the embodied intelligent robot, and to analyze and adjust the content and method of interactive dialogue based on the application scenarios obtained after analysis. The application scenario analysis and organization module includes a requirements analysis unit, a behavior analysis unit, and an organization unit; The demand analysis unit, based on the current application scenario obtained from the analysis, delves into and mines user needs. User needs include, but are not limited to, searching, answering questions, and communicating. It predicts and sorts the rules for subsequent interactive dialogues based on the analyzed current user needs. When the rule is food search, it predicts the content of the user's next interactive dialogue and provides the user with food information, food location, and food recommendations based on the network, thereby improving the fluency of the user's dialogue with the embodied intelligent robot. The behavior analysis unit analyzes the user's interactive dialogue behavior based on the rules set by the user and the content of the interaction dialogue with the embodied intelligent robot. It cleans the raw data of the interactive dialogue behavior to remove noise and erroneous data, and performs data standardization to make data from different sources comparable. Based on the application scenario, it identifies different behavior patterns from the processed data, including information sharing behavior patterns, social interaction behavior patterns, and transportation behavior sharing patterns. It transforms the identified behavior patterns into quantifiable feature forms and selects rules similar to user behavior based on the feature forms. The sorting unit is used to integrate the information obtained from the demand analysis unit and the behavior analysis unit, combine the user's needs and user's habitual behaviors, and select the most appropriate rules to interact and communicate with the user based on the integrated information.

[0037] The fuzzy matching module is used during interactive dialogue to respond to spelling errors, grammatical errors, and expression errors made by users, and to conduct interactive dialogue based on rules that match the content provided by the user with high similarity. The fuzzy matching module includes a data preprocessing unit, a matching rule setting unit, and a result output and sorting unit; The data preprocessing unit is used to clean the data collected by the embodied intelligent robot and standardize the data. It can convert the data into text data and audio data in the same format. The matching rule setting unit sets matching rules according to specific application scenarios, which is used to set the leniency of key rule data matching, including allowing partial keyword matching, synonym matching, and similar behavior matching; The result output and sorting unit outputs the matching results after the algorithm completes the matching. The results are sorted according to their similarity, with the most similar results placed at the top. The user is then prompted to interact with the system based on the top results. If the user is not satisfied with the interaction, they can choose a subsequent result to continue the interaction.

[0038] The dialogue strategy selection module uses different strategies to interact with the user based on the information data collected by the embodied intelligent robot. The embodied intelligent robot includes at least one video monitor and environmental sensors. The dialogue strategy selection module also includes a behavior judgment unit, which judges whether the user's behavior is normal or dangerous based on the collected information data, and conducts corresponding interactive dialogues with the user based on the behavior to reduce the user's dangerous behavior. The interactive dialogue flow control module is used to combine stored rules, classify the combinations into appropriate application scenarios, and sort the order of the combinations. When interacting with users, the module interacts with users based on the combinations, application scenarios, and order.

[0039] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. An interactive dialogue control method for embodied intelligent robots, characterized in that: The following steps are included: S1: Establish a rule-based dialogue model, analyze the application scenarios of embodied intelligent robots, and organize the user's needs and common problems based on these scenarios. These needs and problems are then converted into rules, which are extracted and stored in a rule base. Rule extraction is performed both automatically and manually, including collecting and analyzing dialogue data. Text mining techniques are used to understand users' common problems, needs, and expression methods. For manual rule extraction, domain experts from various application scenarios manually extract rules from the collected dialogue data. For automatic rule extraction, clustering algorithms are used to automatically group similar dialogue data, extracting keywords from the groups as rules. The grouped rules are then organized, including the following steps: S101: Classify rules according to their function, application scenario, question type, and dialogue flow; S102: Optimize and organize the extracted rules, check the syntax and logic of the rules; simplify and summarize the rules; test and verify the rules; S103: Use visualization tools to display rules and create rule documents. Visualization tools include flowcharts, decision trees, tables, and lists. S2: Perform fuzzy matching on the rules; S3: Control of interactive dialogue process, combining rules to construct a complete interactive dialogue process. When a user has a dialogue with the embodied intelligent robot, the system matches the combinations based on the corresponding rules and executes the corresponding operations. Based on the application scenario of the embodied intelligent robot, the interactive dialogue process is divided into linear and branching. The linear process follows the order of the pre-combined rules, while the branching process selects different rules to conduct interactive dialogue based on the different conditions provided by the user during the interactive dialogue. S4: Selection of dialogue strategies, including proactive dialogue strategies, adaptive dialogue strategies, and multimodal information dialogue strategies. The selection of dialogue strategies is based on the rules of the dialogue model.

2. The interactive dialogue control method for an embodied intelligent robot according to claim 1, characterized in that: The expression methods are judged and classified, and the embodied intelligent robot interacts with the user based on the user's habit of expressing themselves in dialogue. The expression methods are divided into direct statement, question, example, metaphor and analogy, and rules for expression methods are established based on this. When the user's dialogue is expressed in the form of direct narration, the embodied intelligent robot directly listens to the content of the user's dialogue and makes corresponding dialogue and action responses. When a user asks a question, the embodied intelligent robot interacts with the conversation based on the question asked. When a user's dialogue is expressed through examples, the embodied intelligent robot learns the dialogue content in conjunction with the current application scenario, makes judgments based on the examples given in the user's dialogue, and interacts with the user based on the judgment results. When user dialogue uses metaphors and analogies, the system learns the dialogue content in conjunction with the current application scenario, and extracts keywords based on the metaphors and analogies used in the user dialogue to understand the specific content of the user dialogue, and then interacts with the user accordingly.

3. The interactive dialogue control method for an embodied intelligent robot according to claim 1, characterized in that: In step S1, the application scenario is determined based on the functions equipped by the embodied intelligent robot, and the target is determined based on the determined application scenario. The target includes children, parents, men, women, couples, children and the elderly. When conducting interactive dialogue, the current application scenario is determined based on the dialogue data, and interactive dialogue is conducted with the corresponding user through the determined application scenario. Different interactive dialogue control methods are adopted when the embodied intelligent robot is in different application scenarios.

4. The interactive dialogue control method for an embodied intelligent robot according to claim 1, characterized in that: In step S2, fuzzy matching includes: performing fuzzy matching based on edit distance, backing up the dialogue data, editing the distance of the backed-up content, setting an edit distance threshold for text data in a rule, and determining that the user's dialogue input matches the rule when the edit distance is less than the threshold. The embodied intelligent robot will then conduct corresponding interactive dialogue with the user. Editing includes inserting, deleting, and replacing text data. Otherwise, it does not match, and a new rule is generated and recorded. The fuzzy matching based on edit distance is used to handle spelling and grammatical errors in dialogue data; Fuzzy matching also includes: fuzzy matching based on text data similarity calculation, measuring the similarity between the user's dialogue input and the text data in the rules through cosine similarity, recording the content of the user's dialogue input and the text data in the rules as strings, representing them as vectors, calculating the cosine similarity between the two strings, and when the similarity is close, it is determined that the user's dialogue input matches the rules, and the embodied intelligent robot will conduct corresponding interactive dialogue with the user; The fuzzy matching based on text data similarity calculation is used to process semantically similar but not completely identical expressions in dialogue data; The specific formula for measuring cosine similarity is as follows: ； in The input vector for the user dialogue. This is a vector of text data in the rule. For vectors sum vector dot product, and Representing vectors respectively sum vector The model, This is the calculated cosine similarity, with values ​​ranging from -1 to 1. A value closer to 1 indicates a stronger similarity between the vectors. sum vector The more similar the values, the better the match between the user's dialogue input and the rule. Values ​​closer to -1 indicate a stronger vector. sum vector The more dissimilar the rules are, the more likely the user's dialogue input does not match the rules, and a new rule is generated and recorded.

5. The interactive dialogue control method for an embodied intelligent robot according to claim 1, characterized in that: The proactive dialogue strategy is a strategy in which the embodied intelligent robot proactively initiates dialogue and guides the user to engage in interactive dialogue. It establishes proactive dialogue rules, and the embodied intelligent robot proactively provides relevant information to the user based on the user's behavior. The process of distinguishing between dangerous and normal user behavior includes the following steps: A: Data collection: Visual information data is provided through video monitoring. The visual information data uses target detection algorithms to detect the user's position and posture, and extract the human body's contour and joint position. Environmental information data is provided through sensor data. The environmental information data is extracted from the current environment using environmental sensors, including temperature and humidity sensors and hazardous gas sensors. By analyzing user dialogue data, we can identify any dangerous behavior data, analyze the dialogue data based on rules, and establish a database of dangerous dialogue texts. B: The judgment between dangerous and normal behaviors is based on the formula and the provided information data. The formula is as follows: ； in For behavioral frequency indicators, For behavioral intensity indicators, For environmental factor indicators, assign weights to each indicator. The total weight is 1. A risk threshold is set based on the application scenario and rules. When the value is less than or equal to the threshold, it is considered normal behavior; when... If the value exceeds the threshold, it is judged as dangerous behavior; C: The provided information and data are labeled, clearly indicating whether the information and data constitute normal or dangerous behavior. When a user engages in dangerous behavior, the embodied intelligent robot will proactively initiate a dialogue with the user.

6. The interactive dialogue control method for an embodied intelligent robot according to claim 1, characterized in that: The adaptive dialogue strategy dynamically adjusts the dialogue strategy of the embodied intelligent robot based on user feedback and the progress of the dialogue.

7. The interactive dialogue control method for an embodied intelligent robot according to claim 1, characterized in that: The multi-modal information dialogue strategy is based on the embodied intelligent robot using multiple modes of information, including but not limited to voice, images and gestures, to conduct corresponding interactive dialogues with the user.

8. An interactive dialogue control system for an embodied intelligent robot, used to implement the interactive dialogue control method for an embodied intelligent robot as described in any one of claims 1-7, characterized in that: include: The rules module is used to formulate rules for the interactive dialogue control of the embodied intelligent robot, and to establish a dialogue model based on the rules module. It extracts the user's dialogue data, and judges the conditions and actions of the extracted dialogue data. The conditions are used to determine whether the user's dialogue data is within the working range of the embodied intelligent robot, and the actions are the response actions that the embodied intelligent robot will take when the working range is met. A rule base, which stores the rule content formulated for the embodied intelligent robot; The application scenario analysis and organization module is used to analyze the application scenarios used by the embodied intelligent robot, and to analyze and adjust the content and method of interactive dialogue based on the application scenarios obtained after analysis. The fuzzy matching module is used to respond to spelling errors, grammatical errors and expression errors generated by the user during interactive dialogue, and to conduct interactive dialogue based on the rules of high similarity of the content provided by the user. A dialogue strategy selection module, which uses different strategies to interact with the user based on information data collected by the embodied intelligent robot, the embodied intelligent robot including at least one video monitor and environmental sensors. The dialogue strategy selection module also includes a behavior judgment unit, which judges whether the user's behavior is normal or dangerous based on the collected information data, and conducts corresponding interactive dialogues with the user based on the behavior to reduce the user's dangerous behavior. The interactive dialogue flow control module is used to combine stored rules, classify the combinations into appropriate application scenarios, and sort the order of the combinations. When interacting with the user, the module interacts with the user based on the combinations, application scenarios, and order.

Images