Household equipment control method and device, computer equipment and storage medium

Abstract

The invention relates to a home equipment control method and device, computer equipment and a storage medium. The method comprises the steps that entities corresponding to different keywords in user voice data and the spatial position relation between the entities can be known by inquiring a spatial knowledge graph, the defect that existing intention recognition lacks understanding of the spatial entity relation is overcome, and therefore the spoken language and spatial instruction control requirement can be met; the problems that existing intention recognition lacks description of physical space entity relations, consequently, spatial semantic understanding is lacked, and spoken language and spatial instruction control requirements are difficult to meet are solved.

Description

Technical Field

[0001] This application relates to the field of home appliance control, and more particularly to a home appliance control method, apparatus, computer device, and storage medium. Background Technology

[0002] Smart home is a residential platform that integrates facilities related to home life using comprehensive wiring technology, network communication technology, security technology, automatic control technology, and audio-visual technology. It builds an efficient management system for residential facilities and daily household affairs, improves home security, convenience, comfort, and aesthetics, and achieves an environmentally friendly and energy-saving living environment.

[0003] In the field of smart homes, current methods rely on multimodal information to recognize user intent and then accurately control home appliances. However, existing intent recognition methods lack descriptions of physical spatial entity relationships, resulting in a lack of spatial semantic understanding. Therefore, they are unable to meet the control needs of spoken and spatial commands, such as the TV background wall being too dark. Summary of the Invention

[0004] This application provides a home appliance control method, device, computer equipment, and storage medium to address the problem that existing intent recognition lacks description of physical spatial entity relationships, resulting in a lack of spatial semantic understanding and difficulty in meeting the needs of conversational and spatial command control.

[0005] In a first aspect, this application provides a method for controlling home appliances, the method comprising: When user voice data is acquired, it is parsed to obtain multiple keywords and control targets; Query multiple entities corresponding to the keywords in the spatial knowledge graph as target entities. The entity types in the spatial knowledge graph include user entities, spatial entities, and device entities. The semantic relationships between different entities in the spatial knowledge graph include spatial location relationships and functional control relationships. Based on the current dynamic parameters of the target entity in the spatial knowledge graph and the control target, determine the target control command for the target home appliance; The target control command is sent to the target home appliance, wherein the target control command is used to control the target home appliance to achieve the control target.

[0006] Optionally, when acquiring user voice data, the method further includes: Multimodal data is acquired through an edge gateway, wherein the multimodal data includes user status data, environmental status data, and device status data; The dynamic parameters of each entity in the spatial knowledge graph are updated using the multimodal acquisition data.

[0007] Optionally, determining the target control command for the target home appliance based on the current dynamic parameters of the target entity in the spatial knowledge graph and the control target includes: Based on the control objective, the device entity in the target entity is identified as the target home appliance; Obtain the current dynamic parameters of the target entity from the spatial knowledge graph; By combining the current dynamic parameters of the target entity and the control target using a fuzzy semantic parsing model, target control instructions are generated.

[0008] Optionally, the step of generating target control instructions by combining the current dynamic parameters of the target entity and the control target using a fuzzy semantic parsing model includes: By using a fuzzy semantic parsing model to combine the current dynamic parameters of the target entity with the reference control parameters and / or historical preference parameters corresponding to the control target, a target control command is generated.

[0009] Optionally, the step of generating target control instructions by combining the current dynamic parameters of the target entity and the reference control parameters and / or historical preference parameters corresponding to the control target using a fuzzy semantic parsing model includes: The user's voice data is processed by voiceprint recognition to determine the target user corresponding to the user's voice data; Obtain the historical preference parameters of the target user for the control objective; By using a fuzzy semantic parsing model to combine the current dynamic parameters of the target entity, the reference control parameters corresponding to the control target, and / or the historical preference parameters corresponding to the target user, a target control command is generated.

[0010] Optionally, after issuing the target control command to the target home appliance, the method further includes: Upon receiving user feedback data, the fuzzy semantic parsing model and / or the historical preference parameters corresponding to the control target are updated based on the user feedback data to obtain the updated fuzzy semantic parsing model and / or the updated historical preference parameters.

[0011] Optionally, after updating the fuzzy semantic parsing model and / or the historical preference parameters corresponding to the control target based on the user feedback data to obtain the updated fuzzy semantic parsing model and / or the updated historical preference parameters, the method further includes: Based on the updated fuzzy semantic parsing model and / or the updated historical preference parameters, the steps of generating target control instructions and issuing the target control instructions to the target home device are re-executed by combining the current dynamic parameters of the target entity and the reference control parameters and / or historical preference parameters corresponding to the control target using the fuzzy semantic parsing model.

[0012] Secondly, this application provides a home appliance control device, the device comprising: The parsing module is used to parse and process the user's voice data when it is acquired, so as to obtain multiple keywords and control targets; The query module is used to query multiple entities corresponding to the keywords in the spatial knowledge graph as target entities. The entity types in the spatial knowledge graph include user entities, spatial entities, and device entities. The semantic relationships between different entities in the spatial knowledge graph include spatial location relationships and functional control relationships. The processing module is used to determine the target control command of the target home device based on the current dynamic parameters of the target entity in the spatial knowledge graph and the control target; The control module is used to send the target control command to the target home appliance, wherein the target control command is used to control the target home appliance to achieve the control target.

[0013] Thirdly, this application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described home appliance control method.

[0014] Fourthly, this application also provides a computer storage medium storing computer-executable instructions for executing the above-described home appliance control method.

[0015] Compared with the prior art, the technical solution provided in this application has the following advantages: The method provided in this application, upon acquiring user voice data, parses and processes the user voice data to obtain multiple keywords and control targets; queries multiple entities corresponding to the keywords in a spatial knowledge graph as target entities, wherein the entity types in the spatial knowledge graph include user entities, spatial entities, and device entities, and the semantic relationships between different entities in the spatial knowledge graph include spatial location relationships and functional control relationships; based on the current dynamic parameters of the target entity in the spatial knowledge graph and the control targets, a target control instruction for the target home device is determined; the target control instruction is sent to the target home device, wherein the target control instruction is used to control the target home device to achieve the control targets.

[0016] Based on the above method, by querying the spatial knowledge graph, we can know the entities corresponding to different keywords in the user's voice data and the spatial positional relationships between entities. This makes up for the lack of understanding of spatial entity relationships in existing intent recognition, thereby realizing the needs of spoken language and spatial command control. It solves the problem that existing intent recognition lacks description of physical spatial entity relationships, resulting in a lack of spatial semantic understanding and difficulty in meeting the needs of spoken language and spatial command control. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0019] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0020] Figure 1 An application environment diagram of a home appliance control method provided in this application embodiment; Figure 2 A flowchart illustrating a home appliance control method provided in an embodiment of this application; Figure 3 A structural block diagram of a home appliance control device provided in an embodiment of this application; Figure 4 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] The following disclosure provides numerous different embodiments or examples for implementing various structures of the invention. To simplify the disclosure, specific examples of components and arrangements are described below. These are merely examples and are not intended to limit the scope of the invention. Furthermore, reference numerals and / or letters may be repeated in different examples. Such repetition is for simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or arrangements discussed.

[0023] Figure 1 This is a diagram illustrating the application environment of a home appliance control method in one embodiment. (Refer to...) Figure 1 This home appliance control method is applied to a home appliance control system. The control system includes home appliances 110 and a home appliance control device 120. The home appliances 110 and the control device 120 are connected via a network. Specifically, the home appliance 110 may be a smart lamp, smart refrigerator, smart TV, smart curtains, smart door lock, smart speaker, smart robot, etc. The control device 120 can be integrated into the home appliance 110, or it can be implemented using a separate control panel, or it can be implemented using a separate server or a server cluster consisting of multiple servers.

[0024] In one embodiment, Figure 2 This is a flowchart illustrating a home appliance control method in one embodiment, with reference to... Figure 2 This invention provides a method for controlling home appliances. This embodiment primarily applies this method to the aforementioned... Figure 1 Taking the home appliance control device 120 as an example, the home appliance control method specifically includes the following steps: Step S210: When user voice data is acquired, the user voice data is parsed and processed to obtain multiple keywords and control targets.

[0025] Specifically, user voice data can be voice data acquired when voice acquisition conditions are triggered. Voice acquisition conditions include the user speaking a voice trigger keyword, the user pressing a voice acquisition button, or the user sending voice data to the home appliance control device through a bound terminal.

[0026] The user voice data is parsed and processed. First, the acquired user voice data is formatted to meet the requirements of the speech recognition model. If the voice data is in a common audio format such as MP3 or WAV, an open-source audio processing library (such as FFmpeg) is used to convert it to a specific format supported by the model. Then, mature speech recognition technologies, such as end-to-end speech recognition models based on deep learning, such as DeepSpeech and Wav2Vec 2.0, are used to input the converted voice data into the speech recognition model, which converts the voice signal into corresponding text information. Next, keywords are extracted from the obtained text information. A method based on TF-IDF (Term Frequency-Inverse Document Frequency) algorithm can be used to count the term frequency and inverse document frequency of each word in the text, and words with high TF-IDF values ​​are selected as preliminary keyword candidates. At the same time, a stop word list is used to remove some meaningless function words and interjections from the text. Part-of-speech tagging technology can also be used to prioritize the retention of nouns, verbs, and other parts of speech with actual meaning as keywords.

[0027] Determining the control objective requires analysis based on keywords and control scenario knowledge. A control rule base is established, containing control objective information corresponding to different keyword combinations. For example, when the keywords include "turn on" and "television," the control objective can be determined as "turn on the television device" according to the rule base. If no perfectly matching rule exists in the rule base, a semantic understanding model (such as BERT) is used to perform semantic analysis on the text information. The output of the semantic understanding model is compared with predefined control objective categories to determine the most likely control objective. Finally, the extracted keywords and determined control objectives are organized and stored for use in subsequent control operations.

[0028] Step S220: Query multiple entities corresponding to the keywords in the spatial knowledge graph as target entities. The entity types in the spatial knowledge graph include user entities, spatial entities, and device entities. The semantic relationships between different entities in the spatial knowledge graph include spatial location relationships and functional control relationships.

[0029] Specifically, the spatial knowledge graph includes user entities, spatial entities, device entities, state entities, and semantic relationships between different entities, including spatial location relationships and functional control relationships. Different user entities indicate different user objects, such as user A and user B. The static parameters of user entities include name, role, voiceprint features, and historical preference parameters, while the dynamic parameters include user location data and user activity status. Different spatial entities indicate different spatial areas, such as bedroom, living room, kitchen, study, bedside area, and desk area. Different spatial entities have hierarchical relationships; for example, the bedside area belongs to the bedroom. Different device entities indicate different home appliances, such as smart lamps, smart air conditioners, smart curtains, air purifiers, smart TVs, and smart speakers. The static parameters of device entities include device identifier, device model, and communication interface, while the dynamic parameters include operating mode, on / off status, operating parameters, and operating status. Different state entities indicate different control scenarios or user feedback states, such as comfort level, night mode, reading scenario, and sleep scenario, serving as semantic abstractions of user intent and environmental state.

[0030] The physical location relationships between different entities include terms such as located in, part of, adjacent to, opposite to, and behind. Examples: bedside lamp → located in → bedroom, meaning the bedside lamp is located in the bedroom; desk → adjacent to → bed, meaning the desk is adjacent to the bed. Functional control relationships between different entities include control and being controlled. Examples: air conditioner → controls → indoor temperature, meaning the air conditioner controls the indoor temperature; bedside lamp → is Controlled By → user A, meaning the bedside lamp is controlled by user A. These relationships not only describe the physical layout of devices and spaces but also support the precise parsing of spatially directional expressions in natural language commands such as "the bedside light is too bright."

[0031] Because spatial knowledge graphs contain the physical location relationships between different home appliances, when spatial descriptions exist in user voice data, the target entities related to the user's voice data can be accurately identified by querying the spatial knowledge graph to meet the control requirements of spatial commands. For example, if the user's voice data is "The light on the TV background wall is too dim," then based on the spatial location relationship between the TV and the background wall area lights in the spatial knowledge graph, the device entity related to the user's voice data can be determined to be the TV background wall area lights. Even if the user's voice data does not explicitly specify the TV background wall area lights as the home appliance to be controlled, the user's control intention can be accurately identified as controlling the brightness of the TV background wall area lights.

[0032] Spatial knowledge graphs can use graph databases as the underlying storage engine, supporting complex path queries and multiple inferences; they can index high-frequency query paths (such as "find all lighting devices in a certain space") to accelerate the process; they can support incremental update mechanisms, synchronizing only changed entities or attributes to reduce system overhead; and they can provide standard SPARQL or Cypher query interfaces for seamless integration with NLU modules and control engines.

[0033] Before using spatial knowledge graphs, spatial structures and device distributions are created manually or semi-automatically based on home layout diagrams or user input; static spatial relationships (such as near, locatedIn) are populated through rule engines or manual annotation; device control interfaces, sensor data sources, and graph entities are mapped; IoT data acquisition modules are deployed to achieve automatic injection of environment and device status; and mechanisms such as user feedback, scene learning, and preference transfer are supported to achieve adaptive evolution of spatial knowledge graphs.

[0034] Step S230: Determine the target control command for the target home appliance based on the current dynamic parameters of the target entity in the spatial knowledge graph and the control target.

[0035] Specifically, based on the control objective and the current dynamic parameters of the target entity associated with the control objective, the target home appliances and target control instructions for achieving the control objective are determined. The target control instructions are used to control the target home appliances to achieve the control objective.

[0036] Step S240: Send the target control command to the target home appliance, wherein the target control command is used to control the target home appliance to achieve the control target.

[0037] Specifically, the target home appliance includes at least one home appliance to be controlled, and the target control instruction is used to control the target home appliance. Therefore, when the target home appliance includes only one home appliance, the target control instruction is used to control that home appliance; when the target home appliance includes multiple home appliances, the target control instruction is a linkage instruction, used to control the multiple home appliances included in the target home appliance, and the control objective is achieved by controlling multiple home appliances simultaneously.

[0038] Based on the above method, by querying the spatial knowledge graph, we can know the entities corresponding to different keywords in the user's voice data and the spatial positional relationships between entities. This makes up for the lack of understanding of spatial entity relationships in existing intent recognition, thereby realizing the needs of spoken language and spatial command control. It solves the problem that existing intent recognition lacks description of physical spatial entity relationships, resulting in a lack of spatial semantic understanding and difficulty in meeting the needs of spoken language and spatial command control.

[0039] In one embodiment, when user voice data is acquired, the method further includes: Multimodal data is acquired through an edge gateway, wherein the multimodal data includes user status data, environmental status data, and device status data; The dynamic parameters of each entity in the spatial knowledge graph are updated using the multimodal acquisition data.

[0040] Specifically, edge gateways are key network devices deployed at the edge of the Internet of Things (IoT). They connect edge nodes such as terminal devices and sensors with the cloud platform and have core functions such as data transmission, protocol conversion, edge computing, and device management. They are the core hub for achieving collaboration between the edge and the cloud.

[0041] Multimodal acquisition data is obtained from multiple detection devices within the environment via an edge gateway. These devices include UWB devices, Wi-Fi fingerprint positioning devices, cameras, temperature sensors, humidity sensors, thermal imaging sensors, positioning detectors, and brightness sensors. The multimodal acquisition data specifically includes user status data, environmental status data, and device status data. User status data includes user location data and user activity status. User location data indicates the user's precise location indoors, specifically obtained in real-time through UWB positioning, Wi-Fi fingerprinting, or microphone array triangulation technology, and then bound to the user's location data in a spatial knowledge graph. User activity status indicates the user's activity state, such as exercising, sleeping, cooking, reading, or watching a movie. Environmental status data includes indoor ambient temperature, indoor ambient humidity, indoor ambient brightness, and indoor air quality. Device status data specifically includes the operating mode, operating parameters, and operating status of home appliances, such as the on / off status and display brightness of lights, and the on / off status, playback volume, playback channel, and screen brightness of smart TVs.

[0042] When user voice data is acquired, multimodal data acquisition is triggered to avoid wasting communication resources and saving communication costs caused by real-time acquisition of multimodal data. The multimodal data acquired at this time is used to update the dynamic parameters of each entity in the spatial knowledge graph, that is, the multimodal data is written into the dynamic parameters of the corresponding entity, so that the latest dynamic parameters can be used to output target control commands corresponding to the control target.

[0043] In one embodiment, determining the target control command for the target home device based on the current dynamic parameters of the target entity in the spatial knowledge graph and the control target includes: Based on the control objective, the device entity in the target entity is identified as the target home appliance; Obtain the current dynamic parameters of the target entity from the spatial knowledge graph; By combining the current dynamic parameters of the target entity and the control target using a fuzzy semantic parsing model, target control instructions are generated.

[0044] Specifically, since the target entity is an entity related to the control target, and the target entity may include at least one of the following: device entity, user entity, space entity, and state entity, if there is a device entity in the target entity, then the device entity is identified as the target home device.

[0045] Once the target entity is identified, its current dynamic parameters are retrieved from the spatial knowledge graph. These current dynamic parameters refer to the most up-to-date dynamic parameters. For user entities, these include user location data and user activity status. For spatial entities, they include the spatial area indicated by the user location data and environmental status data. For device entities, they include the operating mode, operating parameters, and operating status of the home appliances. Operating parameters include operating speed, operating frequency, operating load, and operating energy consumption.

[0046] The current dynamic parameters and control objectives of the target entity are input into the fuzzy semantic parsing model. The fuzzy semantic parsing model analyzes the current dynamic parameters and control objectives of the target entity and outputs the target control instructions for the target home device to achieve the control objectives.

[0047] Through the above process, precise and efficient control of target home appliances can be achieved. Because it is based on the current dynamic parameters of the target entity, factors such as the user's location and activity status, the spatial area, and the specific operating mode, parameters, and status of the home appliances are all taken into consideration. This ensures that the target control commands output by the fuzzy semantic parsing model closely match every factor that might affect the control result. For example, when a user is in the bedroom and issues fuzzy voice data to adjust the bedroom air conditioner temperature, the system will determine the location as a bedroom based on the user's location data, obtain the current operating mode, temperature setting, and other operating parameters and status of the bedroom air conditioner, and thus output the most accurate temperature adjustment target control command, avoiding control errors or deviations.

[0048] In terms of efficiency, thanks to the rapid acquisition of the target entity's current dynamic parameters from the spatial knowledge graph and the powerful analytical capabilities of the fuzzy semantic parsing model, the entire process can be completed in a short time. Without complex manual intervention or multiple adjustments, the system can automatically and quickly output target control commands, allowing target home appliances to respond swiftly and meet the user's actual needs. Moreover, this technology can adapt to different types of home appliances and complex, ever-changing real-world scenarios. Whether controlling simple devices such as smart door locks and lighting systems, or controlling complex scenarios involving multiple devices working together, this technology can accurately generate appropriate control commands, effectively improving the overall performance and user experience of the home automation system.

[0049] In one embodiment, generating target control instructions by combining the current dynamic parameters of the target entity and the control target using a fuzzy semantic parsing model includes: By using a fuzzy semantic parsing model to combine the current dynamic parameters of the target entity with the reference control parameters and / or historical preference parameters corresponding to the control target, a target control command is generated.

[0050] Specifically, the reference control parameters corresponding to the control objective refer to the standard parameters of each home appliance used to achieve the control objective, while the historical preference parameters corresponding to the control objective refer to the historical parameters of each home appliance used to achieve the control objective in the past. In other words, the fuzzy semantic parsing model can generate target control instructions for the target home appliance used to achieve the control objective based solely on the current dynamic parameters of the target entity and the reference control parameters corresponding to the control objective.

[0051] The fuzzy semantic parsing model can also generate target control instructions for the target home appliance to achieve the control objective based solely on the target entity's current dynamic parameters and the historical preference parameters corresponding to the control objective. Furthermore, the fuzzy semantic parsing model can generate target control instructions based on the target entity's current dynamic parameters, the reference control parameters corresponding to the control objective, and the historical preference parameters.

[0052] For example, if a user's voice data is "The bedside light is too bright," then the semantic parsing process is initiated. After noise reduction and recognition, the keywords extracted by natural language understanding technology include "light," the spatial clue "bedside," and the vague description "too bright." Combined with the user's current space (bedroom) as context, the system queries all lighting devices in the "bedroom" in the spatial knowledge graph and uses spatial relationships such as "near (bedside lamp, bed)" to accurately locate the target device as "bedside lamp." Subsequently, the fuzzy semantic parsing model is invoked to calculate the target brightness should be adjusted to 30% by taking into account factors such as the current ambient illuminance (200 lux), the current brightness of the device (80%), the reference control parameters corresponding to the nighttime usage scenario (nighttime light intensity is recommended not to exceed 50 lux), and the user's historical preferences (usually set to 40% brightness). A gradual dimming command (target control command) is generated. The gradual dimming command is used to adjust the brightness to 30% within a gradual time (e.g., 2 seconds). The target control command is sent to the bedside lamp (target home device) for execution through the IoT interface. At the same time, the system writes the new state back to the spatial knowledge graph, thus completing a closed-loop control.

[0053] From a technical perspective, generating target control commands based on different parameter combinations can significantly improve the accuracy and personalization of home appliance control. Generating target control commands solely based on reference control parameters and the current dynamic parameters of the target entity ensures that home appliances operate according to standard specifications, achieving the basic functions of the control target and guaranteeing the stability and consistency of the home appliance control system. For example, when adjusting indoor temperature, reference control parameters enable the air conditioner to accurately reach the set standard temperature, creating a comfortable indoor environment for the user. Generating target control commands solely based on historical preference parameters and the current dynamic parameters of the target entity fully considers the user's past usage habits, providing services more tailored to individual needs. For instance, if a user habitually dims the bedroom lights when sleeping at night, historical preference parameters can automatically adjust the lights to a suitable brightness, improving the user experience. When combining the current dynamic parameters of the target entity, the corresponding reference control parameters, and historical preference parameters, the fuzzy semantic parsing model comprehensively considers real-time conditions, standard requirements, and user habits, generating more intelligent and accurate target control commands. For example, in intelligent security systems, the working status of home security devices can be flexibly adjusted based on dynamic parameters of the current environment (such as personnel activity, time, etc.), reference control parameters (security standard settings), and historical preference parameters (users' habitual settings for security modes) to achieve more efficient security protection.

[0054] Furthermore, the fuzzy semantic parsing model can be continuously optimized and learned through machine learning algorithms, dynamically adjusting reference control parameters and historical preference parameters based on user feedback and new data to adapt to the needs of different users and different scenarios.

[0055] In one embodiment, generating target control instructions by combining the current dynamic parameters of the target entity and the reference control parameters and / or historical preference parameters corresponding to the control target using a fuzzy semantic parsing model includes: The user's voice data is processed by voiceprint recognition to determine the target user corresponding to the user's voice data; Obtain the historical preference parameters of the target user for the control objective; By using a fuzzy semantic parsing model to combine the current dynamic parameters of the target entity, the reference control parameters corresponding to the control target, and / or the historical preference parameters corresponding to the target user, a target control command is generated.

[0056] Specifically, voiceprint recognition processing is performed on user voice data to identify the target user corresponding to the voice data, such as a child, male homeowner, female homeowner, or elderly person. Different target users may have different historical preference parameters for the same control target; for example, a child's preference for lighting brightness in a reading environment differs from that of an elderly person. Therefore, the historical preference parameters of the target user for the control target are obtained, and a fuzzy semantic parsing model is used to generate target control commands based on the current dynamic parameters of the target entity, the reference control parameters corresponding to the control target, and / or the historical preference parameters.

[0057] By incorporating historical preference parameters of the target user, control commands are more tailored to the user's actual needs. For example, in smart home scenarios, parameters such as light brightness and temperature set for different target users can accurately match their personalized preferences. Compared to generic control commands, this greatly improves the accuracy of control and avoids the uncomfortable experience caused by a one-size-fits-all approach.

[0058] The fuzzy semantic parsing model takes into account the current dynamic parameters and reference control parameters of the target entity. The control commands can adapt to changes in the environment in a timely manner. Taking air conditioning control as an example, when dynamic parameters such as indoor temperature and humidity change, the model will combine historical preferences and reference control parameters to dynamically adjust the air conditioning operation mode and temperature setting to ensure that the indoor environment is always kept within the range of comfort for the target user.

[0059] The control command generation process based on historical preference parameters reduces manual intervention by the user. The system can automatically generate appropriate control commands based on existing historical data and real-time environmental information, responding quickly and executing control operations. For example, when a user enters a room, the system can automatically adjust the status of devices such as lights, air conditioning, and curtains within a short time based on the user's historical preferences and the current environmental conditions, providing a convenient and efficient user experience. Moreover, this control method based on historical data and fuzzy semantic parsing models can be continuously learned and optimized to further improve the efficiency and quality of control command generation, better meeting the control needs of different target users in various scenarios.

[0060] In one embodiment, after issuing the target control command to the target home appliance, the method further includes: Upon receiving user feedback data, the fuzzy semantic parsing model and / or the historical preference parameters corresponding to the control target are updated based on the user feedback data to obtain the updated fuzzy semantic parsing model and / or the updated historical preference parameters.

[0061] Specifically, user feedback data can be user facial expressions or subsequent voice data. This feedback data is used to update the fuzzy semantic parsing model and / or historical preference parameters. The updated fuzzy semantic parsing model can more accurately identify and process the ambiguous semantics of user input, significantly improving the accuracy and efficiency of semantic understanding and reducing misunderstandings and erroneous responses caused by semantic ambiguity. For example, in the scenario of intelligent voice assistants, it can more accurately understand diverse and non-standard user commands, such as "I want relaxing music." The updated model can provide music recommendations that better match the user's needs based on historical preferences and semantic understanding. At the same time, the updated historical preference parameters enable the system to more accurately grasp the user's personalized needs, provide services that better suit user habits, and enhance the interaction stickiness and satisfaction between the user and the system.

[0062] Based on the updated fuzzy semantic parsing model and historical preference parameters, the system's functionality and application scenarios can be further expanded. This technology can be combined with big data analytics to collect and analyze vast amounts of user feedback data and interaction information, uncovering broader user demand patterns and trends, providing strong support for product optimization and innovation. It can also be integrated with reinforcement learning algorithms in machine learning, allowing the system to continuously learn and optimize through ongoing user interaction, achieving adaptive fuzzy semantic parsing and personalized services. Furthermore, this technology can be extended to multilingual environments, breaking down language barriers by parsing and processing fuzzy semantics in different languages, enabling cross-language intelligent interaction, and providing a more convenient and efficient service experience for global users.

[0063] In one embodiment, after updating the fuzzy semantic parsing model and / or the historical preference parameters corresponding to the control target based on the user feedback data, and obtaining the updated fuzzy semantic parsing model and / or the updated historical preference parameters, the method further includes: Based on the updated fuzzy semantic parsing model and / or the updated historical preference parameters, the steps of generating target control instructions and issuing the target control instructions to the target home device are re-executed by combining the current dynamic parameters of the target entity and the reference control parameters and / or historical preference parameters corresponding to the control target using the fuzzy semantic parsing model.

[0064] Specifically, re-executing the process with updated models and parameters significantly improves the accuracy and adaptability of target control command generation. The updated fuzzy semantic parsing model can more accurately understand the fuzzy semantics of user input, reducing command errors caused by semantic misunderstanding. Combined with updated historical preference parameters, it can better align with users' personalized habits, making the control of target home devices more in line with user expectations and improving the user experience. For example, for controlling smart air conditioners, it can accurately adjust parameters such as temperature and fan speed based on users' historical usage preferences in different seasons and time periods.

[0065] Furthermore, deep integration with cloud servers enables real-time data updates and sharing, continuously optimizing fuzzy semantic parsing models and historical preference parameters, thereby improving the performance and stability of the entire home appliance control system. Simultaneously, machine learning algorithms can be introduced to continuously learn and analyze user behavior, further uncovering potential user needs and providing more personalized and intelligent services.

[0066] Figure 2 This is a flowchart illustrating a home appliance control method in one embodiment. It should be understood that, although... Figure 2 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 2 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0067] In one embodiment, such as Figure 3 As shown, a home appliance control device is provided, comprising: The parsing module 310 is used to parse and process the user voice data when it is acquired, so as to obtain multiple keywords and control targets; The query module 320 is used to query multiple entities corresponding to the keywords in the spatial knowledge graph as target entities. The entity types in the spatial knowledge graph include user entities, spatial entities and device entities. The semantic relationships between different entities in the spatial knowledge graph include spatial location relationships and functional control relationships. Processing module 330 is used to determine the target control command of the target home device based on the current dynamic parameters of the target entity in the spatial knowledge graph and the control target; The control module 340 is used to send the target control command to the target home appliance, wherein the target control command is used to control the target home appliance to achieve the control target.

[0068] In one embodiment, the parsing module 310 is further configured to: Multimodal data is acquired through an edge gateway, wherein the multimodal data includes user status data, environmental status data, and device status data; The dynamic parameters of each entity in the spatial knowledge graph are updated using the multimodal acquisition data.

[0069] In one embodiment, the processing module 330 is further configured to: Based on the control objective, the device entity in the target entity is identified as the target home appliance; Obtain the current dynamic parameters of the target entity from the spatial knowledge graph; By combining the current dynamic parameters of the target entity and the control target using a fuzzy semantic parsing model, target control instructions are generated.

[0070] In one embodiment, the processing module 330 is further configured to: By using a fuzzy semantic parsing model to combine the current dynamic parameters of the target entity with the reference control parameters and / or historical preference parameters corresponding to the control target, a target control command is generated.

[0071] In one embodiment, the processing module 330 is further configured to: The user's voice data is processed by voiceprint recognition to determine the target user corresponding to the user's voice data; Obtain the historical preference parameters of the target user for the control objective; By using a fuzzy semantic parsing model to combine the current dynamic parameters of the target entity, the reference control parameters corresponding to the control target, and / or the historical preference parameters corresponding to the target user, a target control command is generated.

[0072] In one embodiment, the processing module 330 is further configured to: Upon receiving user feedback data, the fuzzy semantic parsing model and / or the historical preference parameters corresponding to the control target are updated based on the user feedback data to obtain the updated fuzzy semantic parsing model and / or the updated historical preference parameters.

[0073] In one embodiment, the processing module 330 is further configured to: Based on the updated fuzzy semantic parsing model and / or the updated historical preference parameters, the steps of generating target control instructions and issuing the target control instructions to the target home device are re-executed by combining the current dynamic parameters of the target entity and the reference control parameters and / or historical preference parameters corresponding to the control target using the fuzzy semantic parsing model.

[0074] It should be noted that the examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the content disclosed in the above embodiments. It should also be noted that the above modules, as part of a device, can operate in environments such as... Figure 1 The hardware environment shown can be implemented either through software or through hardware.

[0075] like Figure 4 As shown, this application provides a computer device including a processor 711, a communication interface 712, a memory 713, and a communication bus 714. The processor 711, the communication interface 712, and the memory 713 communicate with each other through the communication bus 714. The memory 713 is used to store computer programs. When the processor 711 executes the program stored in the memory 713, it implements the home device control method provided in any of the aforementioned method embodiments.

[0076] The memory and processor in the aforementioned electronic devices communicate with each other via a communication bus and a communication interface. The communication bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into an address bus, a data bus, a control bus, etc.

[0077] The memory may include random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0078] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0079] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0080] According to another aspect of the embodiments of this application, a computer program product or computer program is also provided, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps of any of the above embodiments.

[0081] In one embodiment, the home appliance control device provided in this application can be implemented as a computer program, which can be implemented in, for example... Figure 4 The computer device shown is running the program. The computer device's memory can store the various program modules that make up the home appliance control device, for example, Figure 3 The parsing module 310, query module 320, processing module 330, and control module 340 are shown. The computer program comprised of these modules causes the processor to execute the home appliance control methods of the various embodiments of this application described in this specification.

[0082] Figure 4 The computer device shown can be used as follows Figure 3 The parsing module 310 in the home appliance control device shown performs parsing processing on the acquired user voice data to obtain multiple keywords and control targets. The computer device can use the query module 320 to query the entities corresponding to the multiple keywords in the spatial knowledge graph as target entities. The entity types in the spatial knowledge graph include user entities, spatial entities, and device entities, and the semantic relationships between different entities in the spatial knowledge graph include spatial location relationships and functional control relationships. The computer device can use the processing module 330 to determine the target control command for the target home appliance based on the current dynamic parameters of the target entity in the spatial knowledge graph and the control target. The computer device can use the control module 340 to send the target control command to the target home appliance, wherein the target control command is used to control the target home appliance to achieve the control target.

[0083] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the home appliance control method provided in any of the foregoing method embodiments.

[0084] Optionally, in embodiments of this application, the computer-readable medium is configured to store program code for the processor to perform the following steps: When user voice data is acquired, it is parsed to obtain multiple keywords and control targets; Query multiple entities corresponding to the keywords in the spatial knowledge graph as target entities. The entity types in the spatial knowledge graph include user entities, spatial entities, and device entities. The semantic relationships between different entities in the spatial knowledge graph include spatial location relationships and functional control relationships. Based on the current dynamic parameters of the target entity in the spatial knowledge graph and the control target, determine the target control command for the target home appliance; The target control command is sent to the target home appliance, wherein the target control command is used to control the target home appliance to achieve the control target.

[0085] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments, and will not be repeated here.

[0086] It is understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, microcode, or a combination thereof. For hardware implementation, the processing unit can be implemented in one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers, microprocessors, other electronic units for performing the functions described herein, or combinations thereof.

[0087] For software implementation, the techniques described herein can be implemented by units that perform the functions described herein. The software code can be stored in memory and executed by a processor. The memory can be implemented in the processor or external to the processor.

[0088] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0089] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0090] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0091] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0092] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0093] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented using software plus a general-purpose hardware platform, or of course, using hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the related technology, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a USB flash drive, external hard drive, ROM, RAM, magnetic disk, or optical disk, or other media capable of storing program code, including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0094] It should be understood that the terminology used herein is for the purpose of describing particular exemplary embodiments only and is not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “described” as used herein may also mean including the plural forms. The terms “comprising,” “including,” “containing,” and “having” are inclusive and therefore indicate the presence of the stated features, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, elements, components, and / or combinations thereof. The method steps, processes, and operations described herein are not construed as requiring them to be performed in a particular order described or illustrated unless the order of performance is explicitly indicated. It should also be understood that alternatives or substitutions may be used.

[0095] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for controlling home appliances, characterized in that, The method includes: When user voice data is acquired, it is parsed to obtain multiple keywords and control targets; Query multiple entities corresponding to the keywords in the spatial knowledge graph as target entities. The entity types in the spatial knowledge graph include user entities, spatial entities, and device entities. The semantic relationships between different entities in the spatial knowledge graph include spatial location relationships and functional control relationships. Based on the current dynamic parameters of the target entity in the spatial knowledge graph and the control target, determine the target control command for the target home appliance; The target control command is sent to the target home appliance, wherein the target control command is used to control the target home appliance to achieve the control target.

2. The method according to claim 1, characterized in that, When acquiring user voice data, the method further includes: Multimodal data is acquired through an edge gateway, wherein the multimodal data includes user status data, environmental status data, and device status data. The dynamic parameters of each entity in the spatial knowledge graph are updated using the multimodal acquisition data.

3. The method according to claim 2, characterized in that, The step of determining the target control command for the target home appliance based on the current dynamic parameters of the target entity in the spatial knowledge graph and the control target includes: Based on the control objective, the device entity in the target entity is identified as the target home appliance; Obtain the current dynamic parameters of the target entity from the spatial knowledge graph; By using a fuzzy semantic parsing model to combine the current dynamic parameters of the target entity and the control target, a target control command is generated.

4. The method according to claim 3, characterized in that, The step of generating target control instructions by combining the current dynamic parameters of the target entity and the control target using a fuzzy semantic parsing model includes: By using a fuzzy semantic parsing model to combine the current dynamic parameters of the target entity with the reference control parameters and / or historical preference parameters corresponding to the control target, a target control command is generated.

5. The method according to claim 4, characterized in that, The step of generating target control instructions by combining the current dynamic parameters of the target entity and the reference control parameters and / or historical preference parameters corresponding to the control target using a fuzzy semantic parsing model includes: The user's voice data is processed by voiceprint recognition to determine the target user corresponding to the user's voice data; Obtain the historical preference parameters of the target user for the control objective; By using a fuzzy semantic parsing model to combine the current dynamic parameters of the target entity, the reference control parameters corresponding to the control target, and / or the historical preference parameters corresponding to the target user, a target control command is generated.

6. The method according to claim 4, characterized in that, After issuing the target control command to the target home appliance, the method further includes: Upon receiving user feedback data, the fuzzy semantic parsing model and / or the historical preference parameters corresponding to the control target are updated based on the user feedback data to obtain the updated fuzzy semantic parsing model and / or the updated historical preference parameters.

7. The method according to claim 6, characterized in that, After updating the fuzzy semantic parsing model and / or the historical preference parameters corresponding to the control target based on the user feedback data, and obtaining the updated fuzzy semantic parsing model and / or the updated historical preference parameters, the method further includes: Based on the updated fuzzy semantic parsing model and / or the updated historical preference parameters, the steps of generating target control instructions and issuing the target control instructions to the target home device are re-executed by combining the current dynamic parameters of the target entity and the reference control parameters and / or historical preference parameters corresponding to the control target using the fuzzy semantic parsing model.

8. A home appliance control device, characterized in that, The device includes: The parsing module is used to parse and process the user's voice data when it is acquired, so as to obtain multiple keywords and control targets; The query module is used to query multiple entities corresponding to the keywords in the spatial knowledge graph as target entities. The entity types in the spatial knowledge graph include user entities, spatial entities, and device entities. The semantic relationships between different entities in the spatial knowledge graph include spatial location relationships and functional control relationships. The processing module is used to determine the target control command of the target home device based on the current dynamic parameters of the target entity in the spatial knowledge graph and the control target; The control module is used to send the target control command to the target home appliance, wherein the target control command is used to control the target home appliance to achieve the control target.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

Images