Equipment control methods, devices, electronic equipment, media and products
By querying user instance-level semantic mapping rules in a three-layer semantic graph, and dynamically generating manufacturer API adaptation instructions, the problem of low control accuracy of cross-brand devices is solved, and unified control and efficient adaptation of cross-ecosystem devices are achieved.
Patent Information
- Application Number
- CN202611278988.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-21
- Publication Date
- 2026-11-17
AI Technical Summary
Existing technologies cannot achieve unified natural language control across brands and product categories, resulting in reduced device control accuracy, especially in multi-device scenarios where ambiguous commands or erroneous executions are likely to occur.
By querying the current user instance-level semantic mapping rules that match the device to be controlled in the three-layer semantic graph, target control commands adapted to the manufacturer's API are dynamically generated, enabling unified control of devices across the ecosystem.
It improves the accuracy and adaptability of equipment control, and can automatically generate instance-level semantic mapping rules that match specific equipment, ensuring that control commands are executed correctly.
Smart Images

Figure CN123123052A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of smart home control technology, specifically to a device control method, apparatus, electronic device, readable storage medium, and computer program product. Background Technology
[0002] Currently, smart homes have evolved from single-device remote control to whole-house smart scenarios. Users expect to control all devices in their homes across brands and categories through a unified natural language interaction interface. However, control protocols, parameter formats, and semantic expressions vary between different manufacturers' devices, resulting in fragmented ecosystems and making it difficult to achieve a unified natural language control interface. Existing technologies typically establish a unified control layer on top of each manufacturer's cloud service, distributing user commands to the corresponding manufacturer's cloud for execution. Because the control parameter formats of different manufacturers' devices vary, independent parameter mapping configurations need to be maintained for each device. Furthermore, when a user owns multiple devices of the same model or multiple similar devices in the same location, existing technologies cannot determine the specific device the user intends to target, easily leading to command ambiguity or misexecution, and reducing the accuracy of device control in multi-device scenarios. Summary of the Invention
[0003] In view of the above problems, this application provides a device control method, apparatus, electronic device, readable storage medium, and computer program product, which can solve the problem of reduced device control accuracy caused by the inability to distinguish specific user instances in multi-device scenarios.
[0004] In a first aspect, this application provides a device control method, comprising: Receive device control commands from the target user for the device to be controlled; Based on the device control command and the user identifier of the target user, a query is performed in the three-layer semantic graph to obtain the current user instance-level semantic mapping rule that matches the device to be controlled; Generate target control instructions that are compatible with the vendor API based on the device control instructions and the current user instance-level semantic mapping rules; The device to be controlled is controlled accordingly based on the target control command.
[0005] In the above technical solution, the method first receives device control commands to obtain the control requirements of the target user, so as to provide input for subsequent semantic queries; then, it queries the semantic graph to determine the current user instance-level semantic mapping rule that matches the device to be controlled, so as to accurately match the specific device of the target user; next, it performs parameter adaptation according to the current user instance-level semantic mapping rule to generate target control commands in the format of the manufacturer's API, so as to be compatible with the API interfaces of different manufacturers; finally, it issues the target control commands to execute control, and performs corresponding control on the device to be controlled, thereby achieving the effect of unified cross-ecosystem device control based on standardized capability description files.
[0006] In some embodiments, before receiving device control commands from the target user for the device to be controlled, the method further includes: Receive multiple target capability description files submitted by manufacturers in accordance with field definition specifications, each corresponding to a different target device model; Based on the multiple target capability description files, generate multiple model-level standardized capability description data that correspond one-to-one with multiple target device models; A model-level capability description file library is constructed based on the multiple target device models and the multiple model-level standardized capability description data.
[0007] In the above technical solution, the method can standardize the capabilities of devices from different manufacturers and build a model-level capability description file library, so that a single capability description can cover all devices of the same model, thereby helping to improve the access efficiency of cross-ecosystem devices.
[0008] In some implementations, generating multiple model-level standardized capability description data corresponding one-to-one with multiple target device models based on the multiple target capability description files includes: Obtain multiple file information corresponding one-to-one with the multiple target capability description files; wherein, the file information includes file extension, root node tag or file header information; Based on the multiple file information, obtain multiple file formats that correspond one-to-one with the multiple target capability description files; wherein, the file format is JSON format, XML format, or YAML format; Based on the multiple file formats, the corresponding format parsers are called to parse the multiple target capability description files to obtain multiple standardized intermediate data models that correspond one-to-one with the multiple target device models. Based on the multiple standardized intermediate data models, multiple model-level standardized capability description data are generated, each corresponding to a different target equipment model.
[0009] In the above technical solution, the method can automatically identify and uniformly convert device description files of various formats by recognizing the file format and calling the corresponding format parser, thereby being compatible with heterogeneous data sources and eliminating the need for manual adaptation.
[0010] In some embodiments, before receiving device control commands from the target user for the device to be controlled, the method further includes: Receive a user-level device list file reported by the manufacturer's application; wherein, the user-level device list file includes a user identifier, a manufacturer identifier, a device identifier, a device model corresponding to the device identifier, and a custom device name corresponding to the device identifier; Using the user identifier and manufacturer identifier in the user-level device inventory file as a combined index, update the user-level device inventory file to the user-level device ledger corresponding to the user identifier; The triggering times for the manufacturer's application to report the user-level device list file include device network configuration completion, device unbinding, device information change, firmware upgrade, and regular full synchronization.
[0011] In the above technical solution, the method can establish and maintain a user equipment ledger based on the real-time reporting of equipment change information from the manufacturer's application terminal, thereby ensuring that the system always has the latest and most complete equipment information of the user.
[0012] In some implementations, after updating the user-level device inventory file to the user-level device ledger corresponding to the user identifier using the user identifier and vendor identifier as a combined index, the method further includes: Upon receiving a device authorization instruction, user instance information is obtained according to the device authorization instruction; wherein, the user instance information includes a user identifier and a device identifier to be authorized; Based on the user instance information, obtain the model of the device to be authorized corresponding to the device to be authorized identifier from the user-level device ledger that matches the user identifier; Based on the model number of the device to be authorized, a search is performed in the model-level capability description file library based on the device model field to obtain an initial static capability template that matches the model number of the device to be authorized; wherein, the static capability template includes model-level standardized capability description data; Retrieve the current firmware version corresponding to the device to be authorized from the user-level device ledger corresponding to the user identifier; The current firmware version is matched with the minimum firmware version requirement in the initial static capability template to obtain the target static capability template. Based on the target static capability template and the user instance information, generate a target user instance-level semantic mapping rule corresponding to the device identifier to be authorized; The target user instance-level semantic mapping rules are written into the mapping layer of the three-layer semantic graph; The three-layer semantic graph includes an intent layer, a capability layer, and a mapping layer. The intent layer stores standardized user intent identifier definitions and semantic templates. The capability layer stores capability nodes of various manufacturers' devices. The mapping layer stores the target user instance-level semantic mapping rules.
[0013] In the above technical solution, the method can automatically generate exclusive user instance-level semantic mapping rules for authorized devices, thereby enabling the mapping rules to accurately match the devices and thus improving the accuracy of control adaptation.
[0014] In some implementations, the step of querying the three-layer semantic graph based on the device control command and the user identifier of the target user to obtain the current user instance-level semantic mapping rule matching the device to be controlled includes: The device control commands are semantically parsed to obtain standardized control intentions; wherein, the standardized control intentions include user intention identifiers, device type, location identifiers, and standardized control parameters; Obtain user authorization records and user-level device ledgers based on the user identifier of the target user; Based on the location identifier, the user authorization record, and the user-level device ledger, obtain the device identifier to be controlled for the device to be controlled. When the authorized link status corresponding to the user identifier is active, the query is performed in the mapping layer of the three-layer semantic graph using the identifier of the device to be controlled, the user intent identifier, and the user identifier as a joint index to obtain the current user instance-level semantic mapping rule corresponding to the identifier of the device to be controlled.
[0015] In the above technical solution, the method can translate user instructions into structured information that the system can understand, thereby accurately and efficiently finding the corresponding control rules from the graph.
[0016] In some implementations, generating target control instructions adapted to the vendor API based on the device control instructions and the current user instance-level semantic mapping rules includes: Based on the current user instance-level semantic mapping rules, the standardized control parameters are converted to obtain the target control parameters adapted to the vendor API. Target control commands are generated based on the target control parameters.
[0017] In the above technical solution, the method can translate standardized control parameters into a format that can be recognized by the manufacturer's API and generate control commands, thereby being compatible with different manufacturer interfaces and ensuring that the control commands are correctly executed by the device.
[0018] In some embodiments, after controlling the device to be controlled according to the target control command, the method further includes: Receive the execution result fed back by the device to be controlled in response to the target control command; The execution result is stored in the target execution result set corresponding to the manufacturer identifier of the device to be controlled; Calculate the execution success rate of the target execution result set; Determine whether the execution success rate is lower than a preset success rate threshold; If so, then based on the vendor identifier and the target execution result set, obtain the user instance-level semantic mapping rule to be optimized and the execution status log for cases where the execution result is a failure. Based on the execution log, generate alternative mapping rule schemes; The validity of the alternative mapping rule schemes is verified, and the verification results are obtained. When the verification result indicates that the alternative mapping rule scheme is valid, the alternative mapping rule scheme is determined as the optimized rule, and the user instance-level semantic mapping rule to be optimized in the three-layer semantic graph is updated to the optimized rule.
[0019] In the above technical solution, the method can automatically discover inaccurate mapping rules and generate alternative solutions based on the feedback of the device execution results. After verifying their effectiveness, the rules in the graph are updated, so that the mapping rules become more and more accurate with use, thereby continuously improving the success rate of device control.
[0020] In some embodiments, the method further includes: Receive the new capability description file corresponding to the device model to be updated submitted by the manufacturer's application; Obtain the legacy capability description file corresponding to the device model to be updated; The new version of the capability description file and the old version of the capability description file are compared to obtain a change list; The old capability description file is incrementally updated according to the change list to obtain the updated capability description file.
[0021] In the above technical solution, the method can compare the old and new capability description files and find the changes, thereby achieving the effect of upgrading the equipment capability by only updating the changed parts, thus reducing the update cost and improving the efficiency of file library maintenance.
[0022] Secondly, this application provides a device control apparatus, comprising: The receiving unit is used to receive device control commands from the target user for the device to be controlled. The query unit is used to query the three-layer semantic graph based on the device control command and the user identifier of the target user to obtain the current user instance-level semantic mapping rule that matches the device to be controlled. The generation unit is used to generate target control instructions adapted to the vendor API based on the device control instructions and the current user instance-level semantic mapping rules; The control unit is used to control the device to be controlled according to the target control command.
[0023] In the above technical solution, the device first receives device control commands to obtain the control requirements of the target user, providing input for subsequent semantic queries. Then, it queries the semantic graph to determine the current user instance-level semantic mapping rule that matches the device to be controlled, so as to accurately match the specific device of the target user. Next, it performs parameter adaptation according to the current user instance-level semantic mapping rule to generate target control commands in the format of the manufacturer's API, so as to be compatible with the API interfaces of different manufacturers. Finally, it issues the target control commands to execute control, thereby controlling the device to be controlled, thereby improving the effect of unified control of cross-ecosystem devices.
[0024] Thirdly, this application provides an electronic device including a memory and a processor, the memory storing a computer program and the processor running the computer program to cause the electronic device to perform the device control method described in any one of the first aspects.
[0025] Fourthly, this application provides a readable storage medium storing a computer program, which, when executed by a processor, performs the device control method described in any one of the first aspects.
[0026] Fifthly, this application provides a computer program product comprising a computer program that, when executed by a processor, performs the device control method described in any one of the first aspects.
[0027] The beneficial effects of this application include: By abstracting equipment capabilities into model-level capability description files and combining them with real-time reporting and maintenance of user-level equipment ledgers, the control link can achieve unified control of cross-ecosystem equipment without relying on equipment-side modifications. By automatically associating model-level capability templates with user device instances, instance-level semantic mapping rules that match specific devices can be automatically generated, thereby avoiding the need to repeatedly write mappings for devices of the same model. By reporting the device list when triggered by device network configuration, unbinding, information changes, firmware upgrades and regular synchronization, the user's device ledger can be updated in a timely manner as the devices change, and the real-time perception of device changes can be achieved. By parsing user commands into standardized intents and querying and adapting them in an instance-level semantic graph, it is possible to distinguish multiple devices of the same model from the same user and achieve dynamic adaptation that accurately corresponds to specific device instances. By identifying and unifying the capability description files for heterogeneous formats, new vendors can access the platform without having to develop proprietary protocol adaptation layers for each vendor. By comparing the old and new capability description files for incremental updates, it is possible to avoid a full reconstruction of user data or mappings when capabilities change. Through closed-loop feedback of execution results, semantic mapping rules can be continuously corrected, thereby optimizing the adaptation rules with the actual control effect. Attached Figure Description
[0028] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0029] Figure 1 This is a flowchart illustrating the device control method in some embodiments of this application; Figure 2 This is a schematic diagram of the structure of the device control device in some embodiments of this application; Figure 3 This is a schematic diagram of the structure of an electronic device in some embodiments of this application. Detailed Implementation
[0030] The embodiments of the technical solution of this application will now be described in detail with reference to the accompanying drawings. These embodiments are only used to more clearly illustrate the technical solution of this application and are therefore merely examples, and should not be used to limit the scope of protection of this application.
[0031] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.
[0032] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more (including two), similarly, "multiple sets" refers to two or more sets (including two sets), and "multiple pieces" refers to two or more pieces (including two pieces), unless otherwise explicitly specified.
[0033] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.
[0034] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.
[0035] In existing technologies, devices from different manufacturers access a unified cloud control platform through their respective cloud SDKs. This allows the platform to translate user control commands based on pre-configured adaptation rules and distribute them to the devices for execution. However, in this approach, the adaptation rules require manual pre-configuration, necessitating the manual writing of adaptation code when adding new devices from different manufacturers. This prevents dynamic adaptation and results in poor scalability for unified control across different ecosystems.
[0036] To address the aforementioned technical issues, this application provides a device control method. This method aims to dynamically generate target control instructions adapted to the manufacturer's API by querying the current user instance-level semantic mapping rules that match the device to be controlled in a three-layer semantic graph, thereby achieving dynamic and unified control of devices from different manufacturers.
[0037] like Figure 1As shown, some embodiments of this application provide a device control method, which includes: S100: Receive device control commands from the target user for the device to be controlled.
[0038] In this embodiment, the device control command refers to the command issued by the target user to control the device, including voice commands and / or text commands; the voice command is collected by the voice acquisition and recognition module, and the voice is converted into text by calling the voice recognition interface before being input.
[0039] S200: Based on the device control command and the user identifier of the target user, query the three-layer semantic graph to obtain the current user instance-level semantic mapping rule that matches the device to be controlled.
[0040] In this embodiment, the three-layer semantic graph refers to a data structure consisting of an intent layer, a capability layer, and a mapping layer. The intent layer stores standardized user intent identifier definitions and semantic templates, the capability layer stores device capability nodes from various manufacturers, and the capability nodes are instantiated from capability description files. The mapping layer stores user instance-level semantic mapping rules, and the indexes are user identifier, device identifier, and standardized intent.
[0041] In this embodiment, the user instance-level semantic mapping rule refers to the rule generated for a specific user's specific device, which is used to map standardized control intentions to vendor API call information.
[0042] S300 generates target control instructions adapted to the vendor API based on the device control instructions and the current user instance-level semantic mapping rules.
[0043] In this embodiment, API (Application Programming Interface) is the application programming interface.
[0044] In this embodiment, vendor API refers to the application programming interface provided by each vendor for controlling its devices.
[0045] In this embodiment, the target control instruction refers to the instruction generated in accordance with the manufacturer's API call specification and that can be directly executed by the manufacturer's API interface.
[0046] S400: Perform corresponding control on the device to be controlled according to the target control command.
[0047] In this embodiment, the unified API call interface refers to a unified interface used to issue control commands to APIs of various manufacturers.
[0048] In this embodiment, control refers to issuing instructions to the device to be controlled so that it can perform corresponding operations.
[0049] In the above embodiments, the method can achieve compatibility with multiple brands of devices by accurately matching the user's specific device and adapting to the API interfaces of different manufacturers, so as to enable unified control of cross-ecosystem devices, thereby effectively improving the convenience and universality of cross-ecosystem device control.
[0050] In some embodiments, prior to step S100, the method may further include: S510: Receive multiple target capability description files submitted by the manufacturer in accordance with the field definition specifications, each corresponding to a different target device model.
[0051] In this embodiment, the capability description file refers to the file submitted by the manufacturer in accordance with the field definition specification, which is used to declare the capability information of a certain model of its equipment, and is model-level static data.
[0052] In this embodiment, the field definition specification refers to the JSON Schema field definition standard followed by the capability description file. Here, JSON Schema is a specification for defining JSON data formats, and in this application, it is used as a format validation template for the capability description file.
[0053] For example, the vendor access process includes: Manufacturers obtain standardized capability description documents and JSON Schema from the platform provider; Manufacturers, based on the actual control capabilities of their various equipment models, compile capability description documents in accordance with specifications, declaring model identifiers, capability lists, and API parameter formats; Manufacturers upload capability description files through the developer console.
[0054] In this embodiment, the data base adopts a two-level data structure that separates model-level templates and user-level instances: the capability description file is model-level data, which is submitted by the manufacturer during access and is used to describe the capability list and API parameter format of a certain model of device. It is updated less frequently, and one capability description file can serve multiple devices of that model.
[0055] In this embodiment, the field definition specifications for the capability description file include: The root node vendor (vendor information, required) includes the vendor's unique identifier id, vendor name name, and API version number api_version; The root node device_model (device model information, required) includes the device model (association key), device type device_type, and minimum firmware version requirement min_firmware_version; The root node, capabilities (device capability list, required), includes a unique capability identifier (capability_id), a corresponding standardized intent identifier (standard_intent), and API call information (api_info). api_info includes the API action name (action) and a parameter list (params). params includes the parameter name (name), parameter type (type), value range (min / max), enumerated values (enum_values), and parameter unit (unit).
[0056] S520. Generate multiple model-level standardized capability description data that correspond one-to-one with multiple target equipment models based on multiple target capability description files.
[0057] In this embodiment, the model-level standardized capability description data refers to the data that corresponds one-to-one with the equipment model and describes the capabilities of the equipment model in a standardized format.
[0058] To ensure compatibility with heterogeneous data sources, step S520 may include: S521. Obtain multiple file information corresponding one-to-one with multiple target capability description files; wherein, the file information includes file extension, root node label or file header information.
[0059] In this embodiment, heterogeneous data sources refer to device description information in different formats and from different sources.
[0060] In this embodiment, file information refers to information used to identify the capability description file format, including file extension, root node label, or file header information.
[0061] In this embodiment, the prior art has a variety of device description formats, such as the Matter protocol (Matter is a general connectivity standard in the smart home field), the Device Description File (the device description file format defined by the Matter Smart Home Alliance), the object model definitions formulated by various national or regional standards organizations, and the proprietary formats of various manufacturers. If only a single format is supported, it will be difficult to be compatible with device information from different sources.
[0062] S522. Obtain multiple file formats that correspond one-to-one with multiple target capability description files based on multiple file information; wherein, the file format is JSON format, XML format or YAML format.
[0063] In this embodiment, the file format refers to the format type used by the capability description file.
[0064] In this embodiment, the method can automatically identify the corresponding file format type based on the file extension, root node label, or file header information of the target capability description file.
[0065] S523. Based on multiple file formats, call the corresponding format parser to parse the multiple target capability description files to obtain multiple standardized intermediate data models that correspond one-to-one with the multiple target device models.
[0066] In this embodiment, "calling" refers to executing the corresponding parsing program.
[0067] In this embodiment, the format parser refers to a parsing program that corresponds to the file format and is used to parse the original file into a standardized intermediate data model.
[0068] In this embodiment, the standardized intermediate data model refers to the intermediate data model obtained after parsing the capability description file, which has the same semantic structure as JSONSchema.
[0069] In this embodiment, the method can call a format parser corresponding to the file format to parse the target capability description file into a standardized intermediate data model within the system; wherein, the standardized intermediate data model has the same semantic structure as the JSON Schema defined in this application.
[0070] In some embodiments, in addition to JSON format, capability description files can also use XML or YAML format as alternative data carriers. When using XML format, the system can parse it into a standardized intermediate data model using the corresponding XML parser, and can use XSD (XML Schema Definition) to perform stronger constraint validation on the file; when using YAML format, the system can parse it using the corresponding YAML parser. Regardless of whether the original format of the capability description file is JSON, XML, or YAML, after conversion by their respective format parsers, they are all uniformly converted into the standardized intermediate data model and follow the same subsequent processing path, thereby ensuring the consistency of the processing flow under heterogeneous formats.
[0071] S524. Based on multiple standardized intermediate data models, generate multiple model-level standardized capability description data that correspond one-to-one with multiple target equipment models.
[0072] In this embodiment, before generating model-level standardized capability description data, the target capability description file is sequentially checked for format, schema, semantics, and integrity.
[0073] In this embodiment, the model-level standardized capability description data generated based on the standardized intermediate data model and the model-level standardized capability description data generated based on the capability description file in JSON format of this application participate in subsequent processing along the same path.
[0074] S530. Construct a model-level capability description file library based on multiple target equipment models and multiple model-level standardized capability description data.
[0075] In this embodiment, the model-level capability description file library refers to a database used to store standardized model-level capability description data that has passed verification.
[0076] In this embodiment, the method can store the verified capability description file into the model-level capability description file library; after the capability description file is stored in the library, the system returns a successful access notification to the manufacturer.
[0077] In this embodiment, for existing users who have already reported their device lists, the method automatically triggers a matching process between the newly connected capability description files and the user device ledger.
[0078] In this embodiment, the method can generate instance-level semantic mapping rules by automatically associating model-level capability templates with user-level device instances, thereby achieving dynamic adaptation to devices from different manufacturers.
[0079] In the above embodiments, the method can standardize the capabilities of devices from different manufacturers and build a model-level capability description file library, so that one capability description can cover all devices of the same model, thereby helping to improve the access efficiency of cross-ecosystem devices.
[0080] In some embodiments, prior to step S100, the method may further include: S610: Receive user-level device list file reported by the manufacturer's application terminal; wherein, the user-level device list file includes user identifier, manufacturer identifier, device identifier, device model corresponding to the device identifier, and custom device name corresponding to the device identifier.
[0081] In this embodiment, the vendor application refers to the client application provided by the vendor.
[0082] In this embodiment, the user-level device manifest file refers to the file reported by the manufacturer's application that describes a list of device instances specifically owned by the user.
[0083] For example, after a user completes device configuration and binding through the manufacturer's application, the manufacturer's application obtains a list of all devices currently bound by the user in the configuration success callback, assembles JSON data according to the device list file format, and sends the user-level device list file to this system through the data reporting interface.
[0084] In this embodiment, the user-level device list file is dynamic data used to describe specific device instances owned by a user. The user-level device list file is generated and reported in real time by the manufacturer's application after the user's network configuration is completed. It is updated frequently and reflects changes in user devices in real time.
[0085] In this embodiment, the field definitions of the user-level device manifest file include: user_id (unique user identifier, required), vendor_id (vendor identifier, required), reported_at (reported timestamp, required), and devices (list of device instances, required). Each device instance in devices includes device_id (unique device identifier, required), model (device model, required, is the association key, corresponding to the model in the capability description file), name (user-defined name), firmware_version (current firmware version), and location (device location).
[0086] S620. Using the user identifier and manufacturer identifier in the user-level equipment inventory file as a combined index, update the user-level equipment inventory file to the user-level equipment ledger corresponding to the user identifier.
[0087] In this embodiment, the user-level device ledger refers to the device ledger stored using user identifier and manufacturer identifier as a combined index.
[0088] In this embodiment, a composite index refers to an index composed of multiple fields used to locate data.
[0089] In this embodiment, the triggering times for the manufacturer's application to report the user-level device list file include device network configuration completion, device unbinding, device information change, firmware upgrade, and periodic full synchronization.
[0090] In this embodiment, the user-level device ledger is updated using the user identifier and the manufacturer identifier as a combined index, supporting incremental updates and consistency checks.
[0091] In the above embodiments, the method can establish and maintain a user equipment ledger based on real-time reporting of device change information from the manufacturer's application, thereby ensuring that the system always has the latest and most complete device information for the user.
[0092] In some embodiments, after step S620, the method may further include: S710. Upon receiving a device authorization instruction, obtain user instance information according to the device authorization instruction; wherein, the user instance information includes the user identifier and the identifier of the device to be authorized.
[0093] In this embodiment, the device authorization instruction refers to the instruction initiated by the user to authorize control of a certain device.
[0094] In this embodiment, user instance information refers to information related to the user device instance, including the user identifier and the identifier of the device to be authorized.
[0095] For example, when a user initiates device authorization control on the unified control interface, the system retrieves all devices that the user has reported from the user-level device ledger and displays them, requesting the user to authorize them all at once.
[0096] S720. Based on the user instance information, retrieve the device model to be authorized corresponding to the device identifier to be authorized from the user-level device ledger that matches the user identifier.
[0097] In this embodiment, the device model to be authorized refers to the device model corresponding to the device to be authorized.
[0098] S730. Based on the model of the device to be authorized, perform a search in the model-level capability description file library based on the device model field to obtain an initial static capability template that matches the model of the device to be authorized; wherein, the static capability template includes model-level standardized capability description data.
[0099] In this embodiment, the initial static capability template refers to the static capability template that matches the model of the device to be authorized.
[0100] In this embodiment, the static capability template refers to a capability template that includes model-level standardized capability description data.
[0101] In this embodiment, the device model field is the association key between the capability description file and the user-level device list, and the corresponding capability description file is matched according to the device model in the device ledger.
[0102] S740: Obtain the current firmware version corresponding to the device to be authorized from the user-level device ledger corresponding to the user identifier.
[0103] In this embodiment, the current firmware version refers to the firmware version currently running on the device.
[0104] In this embodiment, the current firmware version is obtained from the firmware version information of the device instance stored in the user-level device ledger, and the same device model can correspond to multiple firmware versions.
[0105] S750: Match the current firmware version with the minimum firmware version requirement in the initial static capability template to obtain the target static capability template.
[0106] In this embodiment, matching refers to the process of comparing two pieces of data to determine their correspondence.
[0107] In this embodiment, the minimum firmware version requirement refers to the minimum firmware version declared in the capability description file that the device needs to support the corresponding capability.
[0108] In this embodiment, the target static capability template refers to the static capability template obtained after matching the current firmware version.
[0109] In this embodiment, different firmware versions of the same device model can correspond to different capability description files, and the capability description file declares the minimum firmware version requirement; the system can match the corresponding capability description file according to the current firmware version reported by the device list, so that devices with different firmware versions can obtain control capabilities that match their versions.
[0110] S760: Based on the target static capability template and user instance information, generate target user instance-level semantic mapping rules corresponding to the identifier of the device to be authorized.
[0111] In this embodiment, the target user instance-level semantic mapping rule refers to the user instance-level semantic mapping rule generated for the device to be authorized.
[0112] In this embodiment, a user's specific device instance is associated with a capability description file, and specific semantic mapping rules for the user and the device are automatically generated without the need for manual adaptation code writing.
[0113] In this embodiment, the two levels of data are associated through the device model field, and the two are decoupled, so that updating one does not affect the other. After the two levels of data are associated through the device model field, user instance-level semantic mapping rules are generated and written into the mapping layer of the three-layer semantic graph.
[0114] S770. Write the target user instance-level semantic mapping rules into the mapping layer of the three-layer semantic graph; wherein, the three-layer semantic graph includes an intent layer, a capability layer and a mapping layer. The intent layer is used to store standardized user intent identifier definitions and semantic templates, the capability layer is used to store the capability nodes of various manufacturers' devices, and the mapping layer is used to store the target user instance-level semantic mapping rules.
[0115] In this embodiment, the intent layer refers to the layer in the three-layer semantic graph used to store standardized user intent identifier definitions and semantic templates.
[0116] In this embodiment, the capability layer refers to the layer in the three-layer semantic graph used to store the capability nodes of each manufacturer's equipment.
[0117] In this embodiment, the mapping layer refers to the layer in the three-layer semantic graph used to store user instance-level semantic mapping rules.
[0118] In this embodiment, after the user instance-level semantic mapping rules are written into the mapping layer of the three-layer semantic graph, the device can participate in unified control.
[0119] In the above embodiments, the method can automatically generate exclusive user instance-level semantic mapping rules for authorized devices, thereby enabling the mapping rules to accurately match the devices and thus improving the accuracy of control adaptation.
[0120] In some embodiments, step S200 may include: S210. Perform semantic parsing on the device control commands to obtain standardized control intents; wherein, the standardized control intents include user intent identifiers, device type, location identifiers, and standardized control parameters.
[0121] In this embodiment, parsing refers to the process of converting instructions into structured information.
[0122] In this embodiment, standardized control intent refers to the standardized intent information obtained after semantic parsing of device control commands, including user intent identifier, device type, location identifier, and standardized control parameters.
[0123] In this embodiment, the user intent identifier refers to the intent identifier defined in the standardized intent library.
[0124] In this embodiment, the location identifier refers to the identifier of the device's location.
[0125] In this embodiment, the standardized control parameters refer to the control parameters carried in the standardized control intent.
[0126] In this embodiment, the large language model refers to the semantic parsing model used for intent recognition and slot extraction.
[0127] In this embodiment, the standardized intent library refers to the set of definitions for standardized intent identifiers and their typical parameters.
[0128] In this embodiment, semantic parsing is based on a large language model to perform intent recognition and slot extraction, and to parse device control commands into standardized control intents.
[0129] In this embodiment, the standardized intent library includes: TogglePower (on / off control, parameter status(boolean)); SetTemperature (temperature setting, parameter temperature(number)); SetFanSpeed (fan speed adjustment, parameter speed(enum / integer)); SetMode (mode switching, parameter mode(enum)); SetHumidity (humidity setting, parameter humidity(number)); StartCleaning (start cleaning); AdjustVolume (volume adjustment, parameter volume(integer)); SetBrightness (brightness adjustment, parameter brightness(integer)); SetColor (color setting, parameter color(string / enum)); OpenCurtain (open curtains); CloseCurtain (close curtains); LockDoor (lock door); UnlockDoor (unlock door); QueryStatus (status query); SetSleepMode (sleep mode setting, parameter enabled(boolean)).
[0130] S220. Obtain user authorization records and user-level device ledgers based on the target user's user identifier.
[0131] In this embodiment, the user authorization record refers to the information that records the user's authorization of the device.
[0132] In this embodiment, user authorization records are stored in an authorization record table, which is used to manage the authorization link status between users, devices and manufacturers.
[0133] S230. Obtain the device identifier to be controlled based on the location identifier, user authorization record, and user-level device ledger.
[0134] For example, this method can determine the target device based on the location identifier in the device control command, combined with the user authorization record and the user-level device ledger; Among them, the device instances in the user-level device ledger record the location of the devices.
[0135] S240. When the authorized link status corresponding to the user identifier is in an effective state, use the device identifier to be controlled, the user intent identifier, and the user identifier as a joint index to perform a query in the mapping layer of the three-layer semantic graph to obtain the current user instance-level semantic mapping rule corresponding to the device identifier to be controlled.
[0136] In this embodiment, the authorization link status refers to the status of the authorization link between the user, device and manufacturer, including the effective status.
[0137] For example, the method can verify whether the authorization link between the user, device and manufacturer is in effect before querying; wherein, the user instance-level semantic mapping rules in the mapping layer are refined from the manufacturer level to the user instance level, which can resolve the conflict of multiple devices having the same name.
[0138] In the above embodiments, the method can translate user instructions into structured information that the system can understand, thereby accurately and efficiently finding the corresponding control rules from the graph.
[0139] In some embodiments, step S300 may include: S310. Based on the current user instance-level semantic mapping rules, perform format conversion on the standardized control parameters to obtain the target control parameters adapted to the vendor API.
[0140] In this embodiment, format conversion refers to converting data from one format to another.
[0141] In this embodiment, the target control parameter refers to the control parameter that is obtained after format conversion and is adapted to the manufacturer's API.
[0142] For example, the method can convert standardized control parameters into a format required by the target vendor based on the current user instance-level semantic mapping rules; wherein the parameter definition in the capability description file includes parameter name, parameter type, numerical range, enumerated values and parameter unit.
[0143] S320. Generate target control commands based on target control parameters.
[0144] For example, this method can assemble and generate target control instructions by carrying target control parameters according to the API action names declared in the capability description file.
[0145] In the above embodiments, the method can translate standardized control parameters into a format that the manufacturer's API can recognize and generate control commands, thereby ensuring compatibility with different manufacturer interfaces and ensuring that the control commands are correctly executed by the device.
[0146] In some embodiments, after step S400, the method may further include: S810: Receive the execution result of the control command from the device to be controlled in response to the target.
[0147] In this embodiment, the execution result refers to the execution status of the device in response to the target control command, including success or failure status and response time.
[0148] In this embodiment, the method can record the execution result of each control command and organize and store it in the target execution result set according to the manufacturer's identifier. The execution result includes a success or failure status and a response time.
[0149] S820. Store the execution results in the target execution result set corresponding to the manufacturer identifier of the device to be controlled.
[0150] In this embodiment, the target execution result set refers to the set of execution results organized by vendor identifier.
[0151] S830. Calculate the execution success rate of the target execution result set.
[0152] In this embodiment, the execution success rate refers to the proportion of the target execution results that are successfully executed.
[0153] For example, this method can statistically analyze the execution success rate of each standardized intent of a corresponding vendor by vendor identifier.
[0154] S840. Determine whether the execution success rate is lower than the preset success rate threshold. If yes, proceed to step S850; otherwise, end the process.
[0155] In this embodiment, the preset success rate threshold refers to a pre-set success rate threshold used to determine whether optimization is triggered.
[0156] In this embodiment, when the success rate of a manufacturer's execution of a certain standardized intent is lower than a preset success rate threshold, the mapping rule optimization process is triggered.
[0157] S850: Based on the vendor identifier and the target execution result set, obtain the semantic mapping rules for the user instance to be optimized and the execution logs for those with failed execution results.
[0158] In this embodiment, the user instance-level semantic mapping rule to be optimized refers to the user instance-level semantic mapping rule that needs to be optimized.
[0159] In this embodiment, the execution status log refers to the log of execution results that are failures.
[0160] S860. Generate alternative mapping rule schemes based on the execution status log.
[0161] In this embodiment, the alternative mapping rule scheme refers to the scheme generated based on the execution log and used to replace the rule to be optimized.
[0162] S870. Verify the validity of the alternative mapping rule schemes and obtain the verification results.
[0163] In this embodiment, the method can verify the effectiveness of alternative mapping rule schemes through A / B testing.
[0164] S880. When the verification result shows that the alternative mapping rule scheme is valid, the alternative mapping rule scheme is determined as the optimized rule, and the user instance-level semantic mapping rule to be optimized in the three-layer semantic graph is updated to the optimized rule.
[0165] In this embodiment, the optimized rule refers to the rule that is determined and updated to the three-layer semantic graph after being verified as valid.
[0166] In this embodiment, after the optimized rules are updated to the mapping layer of the three-layer semantic graph, the mapping rules are continuously optimized according to the actual control effect.
[0167] In the above embodiments, the method can automatically identify inaccurate mapping rules and generate alternative solutions based on the feedback of the device execution results, and update the rules in the graph after verifying their effectiveness, so that the mapping rules become more and more accurate with use, thereby continuously improving the success rate of device control.
[0168] In some embodiments, the method may further include: S910: Receives the new capability description file corresponding to the device model to be updated, submitted by the manufacturer's application.
[0169] In this embodiment, the new capability description file refers to the updated capability description file submitted by the manufacturer's application for the device model to be updated.
[0170] For example, the new capability description file is an updated capability description file submitted by the manufacturer's application for the device model to be updated, including the capability description file of the new model device or the capability description file of the existing model after adding capabilities.
[0171] S920. Obtain the legacy capability description file corresponding to the device model to be updated.
[0172] In this embodiment, the old version of the capability description file refers to the capability description file currently in effect for the device model to be updated.
[0173] S930. Compare the new version of the capability description file with the old version of the capability description file to obtain a change list.
[0174] In this embodiment, the change list refers to the list of changes obtained after comparing the old and new capability description documents, including three categories: additions, deletions, and changes.
[0175] In this embodiment, the change list includes three types of change items: additions, deletions, and modifications.
[0176] S940. Based on the change list, incrementally update the old version of the capability description file to obtain the updated capability description file.
[0177] In this embodiment, incremental update refers to an update method that only updates the changed parts to complete the capability description file upgrade.
[0178] In this embodiment, canary release refers to a release method in which a portion of the data is activated first, and the entire dataset is activated only after verification.
[0179] For example, during incremental updates, newly added capabilities are marked as invalid, deleted capabilities are marked as invalid, and parameter changes are updated. The system supports canary releases, with full implementation taking effect after verification.
[0180] In the above embodiments, the method can compare the old and new capability description files and find the changes, thereby achieving the effect of upgrading the device capability by only updating the changed parts, thereby reducing the update cost and improving the efficiency of file library maintenance.
[0181] To facilitate understanding, examples of this method in several typical application scenarios are given below.
[0182] (1) In the scenario of vendor access The first manufacturer's developers write a capability description file for its first model of air conditioning equipment according to the specifications, upload it to the system through the developer console, and after the system verifies it, it is stored in the capability description file library and automatically triggers the matching of existing users. Users whose reported equipment list includes this model immediately have the ability to control the equipment.
[0183] The access process for other vendors is the same.
[0184] (2) In user network configuration and automatic discovery scenarios After a user purchases air conditioning equipment from the first manufacturer and completes network configuration through that manufacturer's mini-program, the mini-program automatically retrieves a complete list of the user's devices in the successful configuration callback, generates a JSON device list in a standard format, and reports it to the system. The system stores the device ledger and automatically generates mapping rules by matching the capability description file with the `model` field.
[0185] Based on this, users can open the unified control interface and click "Authorize" to control the devices.
[0186] (3) In cross-brand control scenarios The user's living room has an air conditioner from a first-party manufacturer, the bedroom has an air conditioner from a second-party manufacturer, and the study has an air purifier from a third-party manufacturer. The user issues voice commands such as "turn on the living room air conditioner," "set the bedroom air conditioner to 24 degrees," and "turn on the air purifier in the study." The system recognizes these as standardized intents from different manufacturers, queries the corresponding user instance-level mapping rules, executes them accordingly, and finally provides unified feedback on the control results.
[0187] Based on this, users can control the entire house through a single entry point without switching applications.
[0188] (4) In the scenario of precise firmware version matching The same device model has two firmware versions, V1.0 and V2.0, with V2.0 adding a sleep mode capability. Manufacturers submit capability description files for both versions, declaring different minimum firmware version requirements. The system matches the corresponding capability description file based on the firmware version reported in the device list.
[0189] As a result, V1.0 users receive a message indicating that the device does not support the "turn on sleep mode" command, while V2.0 users can respond normally.
[0190] (5) In heterogeneous format compatibility scenarios If a user purchases a smart device that supports a certain standard protocol, but the system detects that there is no capability description file in JSON format for this solution, but a description file in the standard format exists, this method can automatically identify and parse the heterogeneous format, thereby converting it into a standardized intermediate data model and generating mapping rules, so that the user can control the device using natural language commands.
[0191] like Figure 2 As shown, some embodiments of this application provide a schematic diagram of the structure of a device control apparatus. It should be understood that this apparatus is related to... Figure 1 The method executed in the middle corresponds to the steps involved in the aforementioned method. The specific functions and effects of the device can be found in the description above. To avoid repetition, detailed descriptions are omitted here.
[0192] The control device for this equipment includes: The receiving unit 1010 is used to receive device control commands from the target user for the device to be controlled; The query unit 1020 is used to query the three-layer semantic graph based on the device control command and the user identifier of the target user to obtain the current user instance-level semantic mapping rule that matches the device to be controlled. The generation unit 1030 is used to generate target control instructions that are compatible with the vendor API based on the device control instructions and the current user instance-level semantic mapping rules. The control unit 1040 is used to perform corresponding control on the device to be controlled according to the target control command.
[0193] like Figure 3As shown, this application provides an electronic device 1100, which includes a processor 1101 and a memory 1102. The processor 1101 and the memory 1102 are interconnected and communicate with each other through a communication bus 1103 and / or other forms of connection mechanism (not shown). The memory 1102 stores a computer program that can be executed by the processor 1101. When the computer program is run, the processor 1101 executes the computer program to perform the method in any of the aforementioned optional implementations.
[0194] This application provides a readable storage medium storing a computer program, which, when executed by a processor, performs the method in any of the aforementioned optional implementations.
[0195] The readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.
[0196] This application provides a computer program product, which includes a computer program that, when run by a processor, executes the method in any of the aforementioned optional implementations.
[0197] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and not to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and they should all be covered within the scope of the claims and specification of this application. In particular, as long as there is no conflict, the various technical features mentioned in the embodiments can be combined in any way. This application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.
Claims
1. A device control method, characterized in that, include: Receive device control commands from the target user for the device to be controlled; Based on the device control command and the user identifier of the target user, a query is performed in the three-layer semantic graph to obtain the current user instance-level semantic mapping rule that matches the device to be controlled; Generate target control instructions that are compatible with the vendor API based on the device control instructions and the current user instance-level semantic mapping rules; The device to be controlled is controlled accordingly based on the target control command.
2. The equipment control method according to claim 1, characterized in that, Before receiving the device control command from the target user for the device to be controlled, the method further includes: Receive multiple target capability description files submitted by manufacturers in accordance with field definition specifications, each corresponding to a different target device model; Based on the multiple target capability description files, generate multiple model-level standardized capability description data that correspond one-to-one with multiple target device models; A model-level capability description file library is constructed based on the multiple target device models and the multiple model-level standardized capability description data.
3. The equipment control method according to claim 2, characterized in that, The step of generating multiple model-level standardized capability description data corresponding one-to-one with multiple target device models based on the multiple target capability description files includes: Obtain multiple file information corresponding one-to-one with the multiple target capability description files; wherein, the file information includes file extension, root node tag or file header information; Based on the multiple file information, obtain multiple file formats that correspond one-to-one with the multiple target capability description files; wherein, the file format is JSON format, XML format, or YAML format; Based on the multiple file formats, the corresponding format parsers are called to parse the multiple target capability description files to obtain multiple standardized intermediate data models that correspond one-to-one with the multiple target device models. Based on the multiple standardized intermediate data models, multiple model-level standardized capability description data are generated, each corresponding to a different target equipment model.
4. The equipment control method according to claim 1, characterized in that, Before receiving the device control command from the target user for the device to be controlled, the method further includes: Receive a user-level device list file reported by the manufacturer's application; wherein, the user-level device list file includes a user identifier, a manufacturer identifier, a device identifier, a device model corresponding to the device identifier, and a custom device name corresponding to the device identifier; Using the user identifier and manufacturer identifier in the user-level device inventory file as a combined index, update the user-level device inventory file to the user-level device ledger corresponding to the user identifier; The triggering times for the manufacturer's application to report the user-level device list file include device network configuration completion, device unbinding, device information change, firmware upgrade, and regular full synchronization.
5. The equipment control method according to claim 4, characterized in that, After updating the user-level device inventory file to the user-level device ledger corresponding to the user identifier using the user identifier and vendor identifier as a combined index, the method further includes: Upon receiving a device authorization instruction, user instance information is obtained according to the device authorization instruction; wherein, the user instance information includes a user identifier and a device identifier to be authorized; Based on the user instance information, obtain the model of the device to be authorized corresponding to the device to be authorized identifier from the user-level device ledger that matches the user identifier; Based on the model number of the device to be authorized, a search is performed in the model-level capability description file library based on the device model field to obtain an initial static capability template that matches the model number of the device to be authorized; wherein, the static capability template includes model-level standardized capability description data; Retrieve the current firmware version corresponding to the device to be authorized from the user-level device ledger corresponding to the user identifier; The current firmware version is matched with the minimum firmware version requirement in the initial static capability template to obtain the target static capability template. Based on the target static capability template and the user instance information, generate a target user instance-level semantic mapping rule corresponding to the device identifier to be authorized; The target user instance-level semantic mapping rules are written into the mapping layer of the three-layer semantic graph; The three-layer semantic graph includes an intent layer, a capability layer, and a mapping layer. The intent layer stores standardized user intent identifier definitions and semantic templates. The capability layer stores capability nodes of various manufacturers' devices. The mapping layer stores the target user instance-level semantic mapping rules.
6. The equipment control method according to claim 1, characterized in that, The step of querying the three-layer semantic graph based on the device control command and the user identifier of the target user to obtain the current user instance-level semantic mapping rule matching the device to be controlled includes: The device control commands are semantically parsed to obtain standardized control intents; wherein, the standardized control intents include user intent identifiers, device type, location identifiers, and standardized control parameters; Obtain user authorization records and user-level device ledgers based on the user identifier of the target user; Based on the location identifier, the user authorization record, and the user-level device ledger, obtain the device identifier to be controlled for the device to be controlled. When the authorized link status corresponding to the user identifier is in an active state, the query is performed in the mapping layer of the three-layer semantic graph using the identifier of the device to be controlled, the user intent identifier, and the user identifier as a joint index to obtain the current user instance-level semantic mapping rule corresponding to the identifier of the device to be controlled.
7. The equipment control method according to claim 1, characterized in that, The step of generating target control instructions adapted to the vendor API based on the device control instructions and the current user instance-level semantic mapping rules includes: Based on the current user instance-level semantic mapping rules, the standardized control parameters are converted to obtain the target control parameters adapted to the vendor API. Target control commands are generated based on the target control parameters.
8. The equipment control method according to claim 1, characterized in that, After performing corresponding control on the device to be controlled according to the target control command, the method further includes: Receive the execution result fed back by the device to be controlled in response to the target control command; The execution result is stored in the target execution result set corresponding to the manufacturer identifier of the device to be controlled; Calculate the execution success rate of the target execution result set; Determine whether the execution success rate is lower than a preset success rate threshold; If so, then based on the vendor identifier and the target execution result set, obtain the user instance-level semantic mapping rule to be optimized and the execution status log for cases where the execution result is a failure. Based on the execution log, generate alternative mapping rule schemes; The validity of the alternative mapping rule schemes is verified, and the verification results are obtained. When the verification result indicates that the alternative mapping rule scheme is valid, the alternative mapping rule scheme is determined as the optimized rule, and the user instance-level semantic mapping rule to be optimized in the three-layer semantic graph is updated to the optimized rule.
9. The equipment control method according to claim 1, characterized in that, The method further includes: Receive the new capability description file corresponding to the device model to be updated submitted by the manufacturer's application; Obtain the legacy capability description file corresponding to the device model to be updated; The new version of the capability description file and the old version of the capability description file are compared to obtain a change list; The old capability description file is incrementally updated according to the change list to obtain the updated capability description file.
10. A device control apparatus, characterized in that, The equipment control device includes: The receiving unit is used to receive device control commands from the target user for the device to be controlled. The query unit is used to query the three-layer semantic graph based on the device control command and the user identifier of the target user to obtain the current user instance-level semantic mapping rule that matches the device to be controlled. The generation unit is used to generate target control instructions adapted to the vendor API based on the device control instructions and the current user instance-level semantic mapping rules; The control unit is used to control the device to be controlled according to the target control command.
11. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor running the computer program to cause the electronic device to perform the device control method according to any one of claims 1 to 9.
12. A readable storage medium, characterized in that, The readable storage medium stores a computer program, which, when executed by a processor, performs the device control method according to any one of claims 1 to 9.
13. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, performs the device control method according to any one of claims 1 to 9.
