Internet of Things equipment control method, device and equipment based on large language model

Abstract

The invention provides an Internet of Things equipment control method, device and equipment based on a large language model, and relates to the technical field of artificial intelligence, in particular to the fields of natural language processing, big data, Internet of Things, deep learning, digital twinning and large language models. The Internet of Things equipment control method based on the large language model provided by the invention comprises the following steps: acquiring an equipment service demand for the Internet of Things; based on a pre-trained large language model, performing semantic retrieval on a semantic network of the Internet of Things by adopting the equipment service demand, and determining target Internet of Things equipment meeting the equipment service demand and control description data of the target Internet of Things equipment; and controlling the target Internet of Things equipment to provide service based on the equipment service demand and the control description data of the target Internet of Things equipment. According to the invention, the intelligent degree of Internet of Things equipment control can be improved, the flexible control requirement of the Internet of Things equipment can be met, and the Internet of Things equipment use experience of a user can be provided.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and in particular to the fields of natural language processing, big data, the Internet of Things, deep learning, digital twins, and large language models. Background Art

[0002] In the process of developing and using IoT applications, IoT capabilities such as device control and device query of IoT devices are often needed.

[0003] Related technologies generally require customized coding of IoT devices and their various device functions in IoT applications in advance. When the IoT devices are needed to provide services, the corresponding IoT devices are manually searched according to the customized codes, and the corresponding interfaces of the IoT devices are manually called to control the IoT devices to provide services.

[0004] The relevant technology controls IoT devices in a highly customized manner. The degree of intelligence in the control of IoT devices is low. When a large number of IoT devices are connected to IoT applications, it is often difficult to meet the flexible control needs of IoT devices, and users' experience of using IoT devices is poor. Summary of the invention

[0005] The present invention provides a method, apparatus and device for controlling an Internet of Things device based on a large language model.

[0006] According to one aspect of the present disclosure, a method for controlling an IoT device based on a large language model is provided, the method comprising:

[0007] Obtain device service requirements for the Internet of Things;

[0008] Based on the pre-trained large language model, the device service requirements are used to perform semantic retrieval on the semantic network of the Internet of Things to determine the target Internet of Things device that meets the device service requirements and the control description data of the target Internet of Things device;

[0009] Based on the device service requirement and the control description data of the target IoT device, the target IoT device is controlled to provide a service.

[0010] According to another aspect of the present disclosure, there is provided an IoT device control device based on a large language model, the device comprising:

[0011] Demand acquisition module, used to obtain device service requirements for the Internet of Things;

[0012] A semantic detection module, which is used to perform semantic retrieval on the semantic network of the Internet of Things based on the pre-trained large language model and the device service requirements, and determine the target Internet of Things device that meets the device service requirements and the control description data of the target Internet of Things device;

[0013] The device control module is used to control the target IoT device to provide services based on the device service requirements and the control description data of the target IoT device.

[0014] According to another aspect of the present disclosure, an electronic device is provided, the electronic device comprising:

[0015] at least one processor; and

[0016] a memory communicatively connected to the at least one processor; wherein,

[0017] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the Internet of Things device control method based on a large language model described in any embodiment of the present disclosure.

[0018] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable a computer to execute the Internet of Things device control method based on a large language model described in any embodiment of the present disclosure.

[0019] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, the method for controlling an Internet of Things device based on a large language model as described in any embodiment of the present disclosure is implemented.

[0020] According to the technology disclosed in the present invention, the intelligence level of IoT device control can be improved, the flexible control requirements of IoT devices can be met, and it is beneficial to provide users with IoT device usage experience.

[0021] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0023] Figure 1 is a flow chart of an Internet of Things device control method based on a large language model provided according to an embodiment of the present disclosure;

[0024] Figure 2Ais a flowchart of another method for controlling an IoT device based on a large language model according to an embodiment of the present disclosure;

[0025] Figure 2B It is a structural diagram of a semantic network of the Internet of Things provided according to an embodiment of the present disclosure;

[0026] Figure 3 is a flowchart of another method for controlling an IoT device based on a large language model according to an embodiment of the present disclosure;

[0027] Figure 4A It is a flowchart of the construction process and use process of the semantic network of the Internet of Things provided by the embodiment of the present disclosure;

[0028] Figure 4B It is an application example of an Internet of Things device control method based on a large language model provided according to an embodiment of the present disclosure;

[0029] Figure 5 It is a structural schematic diagram of an Internet of Things device control device based on a large language model provided according to an embodiment of the present disclosure;

[0030] Figure 6 It is a block diagram of an electronic device used to implement the Internet of Things device control method based on a large language model according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0031] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0032] Figure 1 This is a flowchart of a method for controlling an IoT device based on a large language model according to an embodiment of the present disclosure. The embodiment of the present disclosure is applicable to the case of controlling an IoT device. The method can be executed by an IoT device control device based on a large language model, which can be implemented in software and / or hardware and configured in an electronic device. Figure 1 As shown, the IoT device control method based on the large language model of this embodiment may include:

[0033] S101, obtaining device service requirements for the Internet of Things.

[0034] S102, based on the pre-trained large language model, using the device service demand to perform semantic retrieval on the semantic network of the Internet of Things, to determine a target Internet of Things device that meets the device service demand and control description data of the target Internet of Things device.

[0035] S103: Based on the device service requirement and the control description data of the target IoT device, control the target IoT device to provide a service.

[0036] The device service demand is proposed for the Internet of Things and is used to request the Internet of Things devices in the Internet of Things to provide services. The device service demand is a natural language. For example, the device service demand is "Pour me a glass of iced Americano" or "Room 410, Zone 3 is too hot". The data format of the device service demand is not limited here, and the device service demand can be voice or text. In the case where the device service demand is voice, a voice-to-text conversion algorithm can be used to convert the device service demand from voice to text.

[0037] The pre-trained large language model can understand the meaning of device service requirements. Among them, the large language model (LLM) refers to a deep learning model trained with a large amount of text data, which can generate natural language text or understand the meaning of language text. The large language model can handle a variety of natural language tasks, such as text classification, question and answer, and dialogue. Through training, the large language model captures knowledge from a large amount of labeled and unlabeled data and stores the knowledge in a large number of parameters, and its model parameters can reach tens of billions or hundreds of billions.

[0038] In an optional embodiment, the semantic network is obtained by adding semantic information to device description data of the Internet of Things devices in the basic network using device technical data of the Internet of Things devices.

[0039] The basic network of the Internet of Things includes the Internet of Things devices in the Internet of Things. The basic network is used to describe the device association relationship between the Internet of Things devices in the Internet of Things. The Internet of Things devices in the basic network are associated with device description data.

[0040] The device description data is used to determine the device description dimension of the IoT device. Optionally, the device description data is metadata used to describe the IoT device.

[0041] However, the device description data associated with the IoT devices in the basic network does not have corresponding semantic information. Therefore, the basic network of IoT devices cannot be understood by the large language model. The disclosed embodiment uses the device technical data of IoT devices to add semantic information to the device description data of IoT devices in the basic network to build a semantic network of the IoT, so that the semantic network of the IoT can be understood by the large language model. This provides technical support for using the large language model to control IoT devices.

[0042] Among them, equipment technical information refers to the technical data and information required in the development, use and maintenance of IoT devices. Equipment technical information Equipment technical information can describe IoT devices in an all-round and multi-angle manner. For example, equipment technical information can be equipment development documents, equipment operation manuals, and public promotional materials related to IoT devices. Semantic information of device description data can be extracted from the equipment technical information of IoT devices.

[0043] Both the device service requirements and the semantic network of the Internet of Things can be understood by the large language model. The large language model uses the device service requirements to perform semantic retrieval on the semantic network of the Internet of Things to determine the target Internet of Things devices that meet the device service requirements and the control description data of the target Internet of Things devices.

[0044] Among them, the target IoT device refers to an IoT device that can meet the device service requirements. Exemplarily, when the device service requirement is "Room 410 in Zone 3 is too hot", the target IoT device is an air conditioner located in Zone 3 that can control Room 410. It is understandable that the scale of the Internet of Things is very large, and the number and variety of IoT devices are large. The semantic network of the Internet of Things may include air conditioners responsible for temperature control in different rooms at multiple different locations, and may also include other types of IoT devices such as coffee machines. The large language model determines the target IoT device that can meet the device service requirements by performing semantic retrieval on the semantic network of the Internet of Things. After the target IoT device is determined, the control description data of the target IoT device is obtained from the semantic network of the Internet of Things.

[0045] The control description data is used to describe the device control method of the IoT device, and the device service requirement is used to describe the device control requirement of the IoT device.

[0046] Based on the device service requirements and the control description data of the target IoT device, the target IoT device can be controlled to provide the service described by the device service requirements. Optionally, based on a pre-trained large language model, the device service requirements and the control description data of the target IoT device are used to control the target IoT device to provide the service.

[0047] The technical solution disclosed in the present invention constructs a semantic network of the Internet of Things that can be understood by a large language model. The large language model is used to control the Internet of Things devices. With the help of the language understanding ability of the large language model, the device service requirements are understood through the large language model, and semantic retrieval is performed on the semantic network of the Internet of Things to determine the target Internet of Things devices that meet the device service requirements and the control description data of the target Internet of Things devices. Based on the device service requirements and the control description data of the target Internet of Things devices, the target Internet of Things devices are controlled to provide services. Users only need to express their device service requirements, and the target Internet of Things devices that can meet the device service requirements can provide services, and the control process of the target Internet of Things devices no longer requires human participation. Compared with the related art, determining the target Internet of Things devices that meet the device service requirements and controlling the target Internet of Things devices both require human participation. The Internet of Things device control process provided by the embodiment of the present invention is simpler and more intelligent, which can meet the flexible control requirements of Internet of Things devices and is conducive to providing users with an experience in using Internet of Things devices.

[0048] In an optional embodiment, based on the device service requirements and the control description data of the target IoT device, the target IoT device is controlled to provide services, including: parsing the control description data of the target IoT device to obtain a device call interface and an interface call command of the target IoT device; determining interface call parameters of the device call interface according to the device service requirements; and based on the interface call parameters and the interface call command, calling the device call interface of the target IoT device to control the target IoT device to provide services.

[0049] Optionally, the control description data includes: a device call interface (Interface) and an interface call command (Command). The control description data of the target IoT device is parsed, and the device call interface and the device call command are extracted from the control description data.

[0050] Among them, the device call interface refers to the interface used to control the target IoT device; the interface call command refers to the command used to call the control interface of the target IoT device. Through the device call interface, the interface call command can be sent to the target IoT device, thereby realizing remote control and monitoring of the target IoT device.

[0051] Device service requirements are used to request IoT devices in the IoT to provide services. Based on the device service requirements, the device control requirements of the IoT devices can be determined, and then the interface call parameters of the device call interface can be determined. Among them, the interface call parameters are the parameters that need to be passed to the interface when calling the device call interface. Taking the case where the target IoT device is a coffee machine as an example, the interface call parameters can be the type of coffee and the temperature of the coffee, or the time when the coffee is made, etc.

[0052] Configure appropriate interface call parameters for the interface call command, send the configured interface call command to the target IoT device through the device call interface, and control the target IoT device to provide services according to the device service requirements.

[0053] Optionally, the large language model is used as a proxy for the semantic network of the Internet of Things, and the semantic network of the Internet of Things is connected to the communication of the Internet of Things devices. The large language model configures appropriate interface call parameters for the interface call command based on the device service requirements, and then sends the configured interface call command to the target Internet of Things device through the semantic network of the Internet of Things, controlling the target Internet of Things device to provide services according to the device service requirements.

[0054] The above technical solution provides a practical target IoT device control method, which can configure appropriate interface call parameters for the interface call command according to the device service requirements, and then send the configured interface call command to the target IoT device to control the target IoT device to provide services according to the device service requirements, thereby improving the intelligence level of IoT device control, meeting the flexible control requirements of IoT devices, and being conducive to providing users with IoT device usage experience.

[0055] Figure 2A It is a flowchart of another method for controlling an IoT device based on a large language model according to an embodiment of the present disclosure; this embodiment is an optional solution proposed on the basis of the above embodiment. The above embodiment discloses that a semantic network is obtained by adding semantic information to the device description data of the IoT device in the basic network using the device technical data of the IoT device. This embodiment provides a solution for determining the semantic information of the device description data.

[0056] See also Figure 2A , the Internet of Things device control method based on the large language model provided in this embodiment includes:

[0057] S201, constructing a semantic description paradigm for the pre-trained large language model based on preset description rules, the device description data and a first speech template.

[0058] The device description data is used to determine the device description dimension of the IoT device. Optionally, the device description data is metadata used to describe the IoT device. Optionally, the device description data includes function description data, location description data, telemetry description data (Telemetry), property description data (Property), and control description data, etc. Exemplarily, the property description data can be an overview and brand of the IoT device, etc.

[0059] The preset description rule is used to constrain the expression of the semantic information of the device description data. The preset description rule is not limited here and is determined according to actual business needs. Optionally, the preset description rule is to express the semantic information of the device description data in the form of SPO triples.

[0060] Taking the case where the IoT device is an air conditioner as an example, the semantic information of the function description data of the air conditioner is air conditioner-function-adjust temperature; the semantic information of the location description data of the air conditioner is air conditioner-location-zone 3; air conditioner-temperature control area-410, 411 and 412.

[0061] The first speech template can be understood as a semantic description sample of the device description data. The first speech template is intended to provide the large language model with correct grammatical, semantic and contextual information so that the large language model can learn and generate text that conforms to these templates. The first speech template can be manually designed or generated by a large language model trained with a large amount of data. The first speech template is not limited here, and the specific one is determined according to the actual application scenario.

[0062] Based on the preset description rules, the device description data and the first speech template, a semantic description paradigm is constructed for the pre-trained large language model, wherein the semantic description paradigm is obtained by instantiating the first speech template using the preset description rules and the device description data.

[0063] S202: Input the semantic description paradigm and the device technical information of the IoT device into the large language model.

[0064] The technical data of IoT devices is the corpus needed for the big language model to generate the semantic information of device description data. The semantic description paradigm is used as the input of the big language model to prompt or guide the big language model to generate the semantic information of device description data with preset description rules.

[0065] The semantic description paradigm and the technical data of the IoT devices are input into the big language model, and the text understanding ability of the big language model is used to extract the semantic information of the device description data from the technical data of the IoT devices. The text generation ability of the big language model is used to generate the semantic information of the device description data according to the preset description rules specified by the semantic description paradigm.

[0066] S203, outputting semantic information of the device description data through the large language model; wherein the semantic information of the device description data is a natural language and complies with a preset description rule.

[0067] The semantic information of the device description data output by the large language model has a standardized expression method and conforms to the preset description rules. The semantic information of the device description data output by the large language model is a natural language that supports semantic retrieval.

[0068] Optionally, the semantic information of the device description data output by the large language model is added to the basic network of the Internet of Things to obtain the semantic network of the Internet of Things.

[0069] The disclosed technical solution provides a practical method for constructing a semantic network of the Internet of Things, which uses the text understanding ability of a large language model to extract semantic information of device description data from device technical data of the Internet of Things devices, uses the text generation ability of the large language model to generate semantic information of the device description data according to preset description rules specified by the semantic description paradigm, adds the semantic information of the device description data output by the large language model to the basic network of the Internet of Things, and obtains the semantic network of the Internet of Things, so that the semantic network of the Internet of Things can be understood by the large language model, provides technical support for using the large language model to control Internet of Things devices, is conducive to improving the intelligent level of control of Internet of Things devices, and is conducive to providing users with Internet of Things device usage experience.

[0070] In an optional embodiment, the basic network of the Internet of Things is constructed in the following manner: determining the device association relationship between the Internet of Things devices in the Internet of Things and the Internet of Things devices; defining the ontology structure of the basic network based on the device association relationship between the Internet of Things devices; instantiating the ontology structure based on the device description data of the Internet of Things devices to obtain the basic network of the Internet of Things; wherein the device description data at least includes control description data.

[0071] Among them, IoT devices refer to electronic devices connected to the IoT. IoT devices can communicate and transmit data through the Internet, supporting intelligent control and remote management. IoT devices may include wireless sensors, smart home devices, and smart wearable devices. The device association relationship between IoT devices is determined based on the data flow between IoT devices. Optionally, the data flow between IoT devices includes: connection, monitoring, and control.

[0072] Optionally, the basic network of the Internet of Things is represented in the form of a knowledge graph. Based on the device association relationships between the IoT devices, the ontology structure of the basic network is defined. The ontology structure of the basic network describes the hierarchical structure, attributes, and constraints between the IoT devices and the device association relationships.

[0073] The ontology structure is instantiated based on the device description data of the IoT device to obtain the basic network of the IoT. The IoT devices in the basic network are associated with the device description data. Optionally, the device description data at least includes control description data to facilitate controlling the IoT device to provide services based on the control description data.

[0074] Figure 2BSchematic diagram of a semantic network structure of the Internet of Things provided according to an embodiment of the present disclosure. Figure 2B As shown, the IoT devices include AHU fresh air unit, AHU1A, variable air conditioner VAV2-4.DPR, variable air conditioner VAV2-3, control valve VAV2-4.DPR, temperature sensor VAV2-4.ZN-T, flow sensor VAV2-4.SUPFLOW, flow controller VAV2-4.SUPFLSP and control valve controller VAV2-4.DPRPOS. The level of the above IoT devices in the basic network is determined according to the device management relationship between the IoT devices. Figure 2B The “control valve point set point”, “temperature sensor point”, “flow sensor point” and “flow control set point”, “air conditioning control area VAV2-3.Zone” and “room 410, room 411 and room 412” are semantic information added to the basic network of the Internet of Things. Optionally, the basic network of the Internet of Things is constructed in the process of digital twinning the Internet of Things devices in the Internet of Things. After the basic network of the Internet of Things is constructed, the semantic network of the Internet of Things is obtained by adding semantic information that meets the preset description rules to the device description data of the Internet of Things devices in the basic network. In this way, the various dimensional information of the Internet of Things devices in the physical world can be accurately described through standard methods, standard digital twinning can be realized, and information loss in the digital twinning process can be effectively avoided. The above technical scheme provides a practical method for constructing the basic network of the Internet of Things, and provides technical support for constructing the semantic network of the Internet of Things.

[0075] Figure 3 It is a flowchart of another method for controlling an Internet of Things device based on a large language model provided according to an embodiment of the present disclosure; this embodiment is an optional solution proposed on the basis of the above-mentioned embodiment.

[0076] See also Figure 3 , the Internet of Things device control method based on the large language model provided in this embodiment includes:

[0077] S301, obtaining device service requirements for the Internet of Things.

[0078] S302, constructing a device retrieval paradigm for a pre-trained large language model according to the device service requirements and the second speech template.

[0079] The second speech template can be understood as a description sample retrieved by the target IoT device, which is used to instruct the large language model to understand the semantic information of the device description data in the semantic network, and determine the target IoT device that can meet the device service requirements from the semantic network of the IoT. The second speech template can be manually designed or generated by a large language model obtained through training with a large amount of data. The second speech template is not limited here, and the specific one is determined according to the actual application scenario.

[0080] Build a device retrieval paradigm based on the device service requirements and the second speech template for the pre-trained large language model.

[0081] S303: Based on the large language model, the device search paradigm is adopted to perform semantic search on the semantic network of the Internet of Things to determine a target Internet of Things device that meets the device service requirement.

[0082] The large language model uses the device retrieval paradigm to perform semantic retrieval on the semantic network of the Internet of Things. Optionally, the semantic information of the device service requirements in the device retrieval paradigm is matched with the semantic information of the device description data in the semantic network, and the IoT devices that match the semantics of the device service requirements are determined as the target IoT devices that meet the device service requirements.

[0083] S304: Acquire control description data of a target IoT device from a semantic network of the IoT based on the large language model.

[0084] The IoT devices in the semantic network of the IoT are associated with device description data, and the control description data of the target IoT device is obtained from the semantic network of the IoT based on the big language model.

[0085] S305: Based on the device service requirement and the control description data of the target IoT device, control the target IoT device to provide a service.

[0086] The disclosed technical solution uses a large language model for IoT device control. With the language understanding ability of the large language model, the device retrieval paradigm constructed based on the device service requirements and the second speech template is adopted to perform semantic retrieval on the semantic network of the IoT, determine the target IoT device that meets the device service requirements, and obtain the control description data of the target IoT device. Then, based on the device service requirements and the control description data of the target IoT device, the target IoT device is controlled to provide services, so that the control process of the IoT device no longer requires human participation, simplifies the control process of the IoT device, improves the intelligence level of the IoT device control, can meet the flexible control requirements of the IoT device, and is conducive to providing users with IoT device usage experience.

[0087] In an optional embodiment, the device retrieval paradigm is adopted based on the large language model to perform semantic retrieval on the semantic network of the Internet of Things to determine the target Internet of Things device that meets the device service demand, including: based on the large language model, the device retrieval paradigm is adopted to perform semantic retrieval on the semantic network of the Internet of Things to obtain candidate Internet of Things devices; if there are at least two candidate Internet of Things devices, a supplementary description of the requirements associated with the device service demand is obtained; based on the large language model, the supplementary description of the requirements is adopted to select the target Internet of Things device that meets the device service demand from the candidate Internet of Things devices.

[0088] The device retrieval paradigm is input into the large language model, and the large language model uses the device retrieval paradigm to perform semantic retrieval on the semantic network of the Internet of Things to obtain candidate Internet of Things devices. Among them, the candidate Internet of Things devices meet the device service requirements.

[0089] If there are at least two candidate IoT devices, a supplementary description of the requirements associated with the device service requirements is obtained. The supplementary description is a further supplement to the device service requirements, and is used to further screen the candidate IoT devices and determine the target IoT device from the candidate IoT devices.

[0090] Optionally, when there are at least two candidate IoT devices, the device description data of each candidate IoT device is obtained, and based on the device description data of the candidate IoT devices and the supplementary description of the requirements, the target IoT device is determined from the candidate IoT devices. Exemplarily, the candidate IoT devices are coffee machine A and coffee machine B. The semantic information of the location description data of coffee machine A is located on the 3rd floor, and the semantic information of the location description data of the coffee machine is located on the 5th floor. The supplementary description of the requirements is "I am on the 3rd floor". Based on this, the coffee machine can be determined as the target IoT device.

[0091] The above technical solution, when at least two candidate IoT devices are retrieved from the semantic network of the IoT, screens the candidate IoT devices according to the supplementary description of the requirements associated with the device service requirements, selects the target IoT device that meets the supplementary description of the requirements, and controls the target IoT device to provide services. This ensures the accuracy of IoT device control and is conducive to providing users with an IoT device usage experience.

[0092] In an optional embodiment, based on a pre-trained large language model, the device service demand is used to perform semantic retrieval on the semantic network of the Internet of Things to determine the target Internet of Things device that meets the device service demand, including: based on the large language model, using the device retrieval paradigm to determine the semantic information of the device service demand; based on the large language model, determining the semantic similarity between the semantic information of the device description data in the semantic network and the semantic information of the device service demand; based on the large language model, determining the target Internet of Things device that meets the device service demand according to the semantic similarity.

[0093] The large language model has text understanding capabilities and can understand the semantic information of device service requirements in the device retrieval paradigm. The large language model can also understand the semantic information of device description data in the semantic network of the Internet of Things. The large language model determines the semantic similarity between the semantic information of device description data and the semantic information of device service requirements in the semantic network.

[0094] The higher the semantic similarity, the closer the meanings expressed by the semantic information of the device description data and the semantic information of the device service requirements are. Optionally, the relative size relationship between the semantic similarity and the similarity threshold is determined, and the target IoT device that meets the device service requirements is determined based on the relative size relationship between the two. Optionally, the IoT device whose semantic similarity is greater than the similarity threshold is determined as the target IoT device. Among them, the similarity threshold is used to measure the similarity between the meanings expressed by the semantic information of the device description data and the semantic information of the device service requirements. The similarity threshold is not limited here and is determined based on actual conditions.

[0095] The above technical solution provides a practical semantic retrieval method, which determines the semantic similarity between the semantic information of device description data and the semantic information of device service requirements in the semantic network based on the large language model, and determines the target IoT device that meets the device service requirements based on the semantic similarity. The above embodiment uses the text understanding ability of the large language model to realize the retrieval of IoT devices, ensures the accuracy of IoT device control, and is conducive to providing users with IoT device usage experience.

[0096] Figure 4A is a flowchart of the construction process and use process of the semantic network of the Internet of Things provided by the embodiment of the present disclosure; see Figure 4A , Figure 4A It includes the process of building the semantic network of the Internet of Things and the process of using the semantic network of the Internet of Things.

[0097] Optionally, the semantic network of the Internet of Things is obtained by adding semantic information to device description data of the Internet of Things devices in the basic network using device technical data of the Internet of Things devices.

[0098] The basic network of the Internet of Things is constructed in the process of digital twinning the IoT devices in the Internet of Things. After the basic network of the Internet of Things is built, semantic information that meets the preset description rules is added to the device description data of the IoT devices in the basic network to obtain the semantic network of the Internet of Things.

[0099] like Figure 4A As shown in Figure 1, the construction process of the semantic network of the Internet of Things mainly includes: the construction process of the basic network of the Internet of Things and the semantic annotation process of the device description data. The construction of the basic network of the Internet of Things and the semantic annotation of the device description data can be achieved by editing the twin ontology.

[0100] Optionally, the twin ontology is edited using DTDL (Digital Twins Definition Language).

[0101] like Figure 4A The "visual editing of DTDL ontology structure" means using DTDL language to define the ontology structure of the basic network of the Internet of Things in a visual editing manner to describe the hierarchical structure, attributes and constraints between the IoT devices and the device association relationships. Then, the ontology structure is instantiated based on the device description data of the IoT device to obtain the basic network of the IoT. The device description data of the IoT device can be the metadata of the IoT device.

[0102] like Figure 4A "DTDL semantic annotation" means annotating semantic information of device description data of IoT devices in the basic network using DTDL language according to the annotation suggestions of device description data. The annotation suggestions of device description data include semantic information of device description data.

[0103] Optionally, the annotation of device description data is recommended by first performing knowledge retrieval on the pre-built IoT knowledge graph library, and then using the semantic service of the knowledge graph to perform semantic disambiguation on the retrieved knowledge entities. It can be known that the large language model has a strong language understanding ability and supports semantic disambiguation and intent extraction. Optionally, the large language model is used to perform semantic disambiguation and semantic extraction on the knowledge entities retrieved from the knowledge graph library.

[0104] Optionally, the big language model is directly used to process the device technical data of the IoT device, and the big language model outputs annotation suggestions for the device description data.

[0105] After the semantic information is labeled, the semantic model of the Internet of Things can be obtained. At this point, the construction of the semantic network of the Internet of Things is completed.

[0106] Optionally, the semantic model of the Internet of Things is stored in a semantic model library of the Internet of Things in preparation for subsequently combining the semantic model of the Internet of Things with the large language model to control the Internet of Things devices.

[0107] The semantic network of the Internet of Things includes the semantic information of the device description data, so that the semantic network of the Internet of Things can be understood by the big language model. In the process of using the semantic network of the Internet of Things, the big language model is introduced and used as the agent of the semantic network of the Internet of Things. The semantic network of the Internet of Things communicates with the Internet of Things devices.

[0108] The device service requirements are adopted by using a large language model to perform device semantic retrieval on the semantic network of the Internet of Things stored in the semantic model library of the Internet of Things. Specifically, based on the large language model, the device retrieval paradigm (prompt) is adopted to determine the semantic information of the device service requirements to realize the semantic understanding of the requirements, and then the semantic network of the Internet of Things is semantically retrieved using the extracted retrieval semantics to obtain candidate Internet of Things devices. If there are at least two candidate Internet of Things devices, the supplementary description of the requirements associated with the device service requirements is obtained, and the supplementary description of the requirements is used as the Internet of Things constraint condition. Based on the Internet of Things constraint condition, the candidate Internet of Things devices are searched again to obtain the target Internet of Things devices that meet both the device service requirements and the Internet of Things constraint conditions. The target Internet of Things device is returned as the device semantic retrieval result through the large language model. It is worth noting that the construction process of the semantic network of the Internet of Things is independent of the use process of the semantic network of the Internet of Things, that is, it is not necessary to rebuild the semantic network of the Internet of Things every time the semantic network of the Internet of Things is used.

[0109] Figure 4B is an application example of an Internet of Things device control method based on a large language model provided according to an embodiment of the present disclosure; see Figure 4B , using the equipment service demand as “please make a cup of coffee” for illustration.

[0110] As an agent of the semantic network of the Internet of Things, the large language model obtains the device service demand "Please make a cup of coffee". The large language model retrieves machines that can make coffee from the semantic network of the Internet of Things, and retrieves machines cm1, cm2, and cm3 that can make coffee. Further, the large language model selects cm1 from cm1, cm2, and cm3 as the target IoT device that meets the device service demand "Please make a cup of coffee".

[0111] Optionally, the large language model obtains a supplementary description of requirements associated with the device service requirements, uses the supplementary description of requirements as an IoT constraint, and selects cm1 from cm1, cm2, and cm3 based on the IoT constraint as a target IoT device that meets the device service requirements and the IoT constraint.

[0112] Next, the control description data of cm1 is obtained from the semantic network of the Internet of Things through the large language model, and the control description data of cm1 is parsed to obtain the device call interface and interface call command of cm1;

[0113] The large language model determines the interface call parameters of the device call interface according to the device service requirements, and based on the interface call parameters and the interface call command, calls the device call interface of cm1 to control cm1 to make coffee. Optionally, the large language model configures appropriate interface call parameters for the interface call command based on "Please make a cup of coffee", and then sends the configured interface call command to cm1 through the semantic network of the Internet of Things to control cm1 to make coffee. Optionally, after cm1 executes the interface call command, the command execution result is fed back to the large language model through the semantic network of the Internet of Things.

[0114] Figure 5 The present invention is a schematic diagram of a structure of an IoT device control device based on a large language model provided according to an embodiment of the present invention. The present invention is applicable to the case of controlling IoT devices. The device can be implemented by software and / or hardware, and the device can implement the IoT device control method based on a large language model described in any embodiment of the present invention.

[0115] like Figure 5 As shown, the Internet of Things device control device 500 based on the large language model includes:

[0116] The demand acquisition module 501 is used to acquire the device service demand for the Internet of Things;

[0117] A semantic detection module 502 is used to perform semantic retrieval on the semantic network of the Internet of Things based on the pre-trained large language model and the device service requirements, and determine a target Internet of Things device that meets the device service requirements and control description data of the target Internet of Things device;

[0118] The device control module 503 is used to control the target IoT device to provide services based on the device service requirements and the control description data of the target IoT device.

[0119] The technical solution disclosed in the present invention constructs a semantic network of the Internet of Things that can be understood by a large language model. The large language model is used to control the Internet of Things devices. With the help of the language understanding ability of the large language model, the device service requirements are understood through the large language model, and semantic retrieval is performed on the semantic network of the Internet of Things to determine the target Internet of Things devices that meet the device service requirements and the control description data of the target Internet of Things devices. Based on the device service requirements and the control description data of the target Internet of Things devices, the target Internet of Things devices are controlled to provide services. Users only need to express their device service requirements, and the target Internet of Things devices that can meet the device service requirements can provide services, and the control process of the target Internet of Things devices no longer requires human participation. Compared with the related art, determining the target Internet of Things devices that meet the device service requirements and controlling the target Internet of Things devices both require human participation. The Internet of Things device control process based on the large language model provided by the embodiment of the present invention is simpler and more intelligent, which can meet the flexible control requirements of Internet of Things devices and is conducive to providing users with an experience in using Internet of Things devices.

[0120] Optionally, the semantic network is obtained by adding semantic information to device description data of the IoT devices in the basic network using device technical data of the IoT devices.

[0121] Optionally, the semantic information of the device description data is determined in the following manner: based on preset description rules, the device description data and a first speech template, a semantic description paradigm is constructed for the pre-trained large language model; the semantic description paradigm and the device technical data of the Internet of Things device are input into the large language model; the semantic information of the device description data is output through the large language model; wherein the semantic information of the device description data is in natural language and complies with preset description rules.

[0122] Optionally, the basic network of the Internet of Things is constructed in the following manner: determine the device association relationship between the Internet of Things devices in the Internet of Things and the Internet of Things devices; based on the device association relationship between the Internet of Things devices, define the ontology structure of the basic network; based on the device description data of the Internet of Things devices, instantiate the ontology structure to obtain the basic network of the Internet of Things; wherein the device description data at least includes control description data.

[0123] Optionally, the semantic detection module 502 includes: a retrieval paradigm construction sub-module, used to construct a device retrieval paradigm for a pre-trained large language model based on the device service requirements and the second speech template; a semantic retrieval sub-module, used to use the device retrieval paradigm based on the large language model to perform semantic retrieval on the semantic network of the Internet of Things to determine the target Internet of Things device that meets the device service requirements; a control description data acquisition sub-module, used to acquire control description data of the target Internet of Things device from the semantic network of the Internet of Things based on the large language model.

[0124] Optionally, the semantic retrieval submodule includes: a candidate device determination unit, used to perform semantic retrieval on the semantic network of the Internet of Things based on the large language model and adopt the device retrieval paradigm to obtain candidate Internet of Things devices; a demand supplement acquisition unit, used to obtain a demand supplement description associated with the device service demand if there are at least two candidate Internet of Things devices; and a target device determination unit, used to select a target Internet of Things device that meets the device service demand from the candidate Internet of Things devices based on the large language model and adopt the demand supplement description.

[0125] Optionally, the semantic detection module 502 includes: a semantic information determination submodule, used to determine the semantic information of the device service demand based on the large language model and using the device retrieval paradigm; a semantic similarity determination submodule, used to determine the semantic similarity between the semantic information of the device description data in the semantic network and the semantic information of the device service demand based on the large language model; and a target device determination submodule, used to determine the target Internet of Things device that meets the device service demand based on the large language model and the semantic similarity.

[0126] Optionally, the device control module 503 includes: a data parsing submodule, used to parse the control description data of the target IoT device to obtain the device call interface and device call command of the target IoT device; a calling parameter determination submodule, used to determine the interface calling parameters of the device calling interface according to the device service requirements; and a device control submodule, used to call the device calling interface of the target IoT device based on the interface calling parameters and the device calling command to control the target IoT device to provide services.

[0127] The Internet of Things device control device based on a large language model provided in the embodiments of the present disclosure can execute the Internet of Things device control method based on a large language model provided in any embodiment of the present disclosure, and has the corresponding functional modules and beneficial effects for executing the Internet of Things device control method based on a large language model.

[0128] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0129] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0130] Figure 6 A schematic block diagram of an example electronic device 500 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0131] like Figure 6 As shown, the electronic device 600 includes a computing unit 601, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 602 or a computer program loaded from a storage unit 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 can also be stored. The computing unit 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0132] Multiple components in the electronic device 600 are connected to the I / O interface 605, including: an input unit 606, such as a keyboard, a mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a disk, an optical disk, etc.; and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 609 allows the electronic device 600 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0133] The computing unit 601 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 601 performs the various methods and processes described above, such as the IoT device control method based on a large language model. For example, in some embodiments, the IoT device control method based on a large language model may be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as a storage unit 608. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 600 via the ROM 602 and / or the communication unit 609. When the computer program is loaded into the RAM 603 and executed by the computing unit 601, one or more steps of the IoT device control method based on the large language model described above may be performed. Alternatively, in other embodiments, the computing unit 601 may be configured to execute the IoT device control method based on a large language model in any other appropriate manner (for example, by means of firmware).

[0134] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0135] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable large language model-based IoT device control device, so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0136] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0137] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0138] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0139] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0140] Artificial intelligence is a discipline that studies how to use computers to simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, planning, etc.). It includes both hardware-level and software-level technologies. Artificial intelligence hardware technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, and big data processing; artificial intelligence software technologies mainly include computer vision technology, speech recognition technology, natural language processing technology, machine learning / deep learning technology, big data processing technology, knowledge graph technology, and other major directions.

[0141] Cloud computing refers to a technology system that uses network access to elastically scalable shared physical or virtual resource pools. Resources can include servers, operating systems, networks, software, applications, and storage devices, and can be deployed and managed on demand and in a self-service manner. Cloud computing technology can provide efficient and powerful data processing capabilities for technical applications such as artificial intelligence and blockchain, as well as model training.

[0142] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0143] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for controlling an Internet of Things device based on a large language model, the method comprising: Obtain device service requirements for the Internet of Things; Based on the pre-trained large language model, the device service requirements are used to perform semantic retrieval on the semantic network of the Internet of Things to determine the target Internet of Things device that meets the device service requirements and the control description data of the target Internet of Things device; Based on the device service requirement and the control description data of the target IoT device, the target IoT device is controlled to provide a service.

2. The method according to claim 1, wherein: The semantic network is obtained by adding semantic information to the device description data of the Internet of Things devices in the basic network using the device technical data of the Internet of Things devices.

3. The method according to claim 2, wherein: The semantic information of the device description data is determined in the following manner: Based on preset description rules, the device description data and the first speech template, construct a semantic description paradigm for the pre-trained large language model; Inputting the semantic description paradigm and the device technical data of the IoT device into the large language model; The semantic information of the device description data is outputted through the large language model; wherein the semantic information of the device description data is a natural language and complies with a preset description rule.

4. The method according to claim 2, wherein: The basic network of the Internet of Things is constructed in the following way: Determine an IoT device in the IoT and a device association relationship between the IoT devices; Based on the device association relationship between the IoT devices, define the ontology structure of the basic network; The ontology structure is instantiated based on the device description data of the Internet of Things device to obtain the basic network of the Internet of Things; wherein the device description data at least includes control description data.

5. The method according to claim 1, wherein: Based on the pre-trained large language model, the device service requirements are used to perform semantic retrieval on the semantic network of the Internet of Things to determine the target Internet of Things device that meets the device service requirements and the control description data of the target Internet of Things device, including: Building a device retrieval paradigm for a pre-trained large language model according to the device service requirements and the second speech template; Based on the large language model, the device retrieval paradigm is used to perform semantic retrieval on the semantic network of the Internet of Things to determine the target Internet of Things device that meets the device service requirements; Based on the large language model, control description data of a target Internet of Things device is acquired from the semantic network of the Internet of Things.

6. The method according to claim 5, wherein: Based on the large language model and using the device retrieval paradigm, a semantic retrieval is performed on the semantic network of the Internet of Things to determine a target Internet of Things device that meets the device service requirements, including: Based on the large language model, the device retrieval paradigm is used to perform semantic retrieval on the semantic network of the Internet of Things to obtain candidate Internet of Things devices; If there are at least two candidate IoT devices, obtaining a supplementary description of the requirements associated with the device service requirements; The requirement supplementary description is adopted based on the large language model to select a target Internet of Things device that meets the device service requirement from the candidate Internet of Things devices.

7. The method according to claim 5, wherein: Based on the pre-trained large language model, the device service requirements are used to perform semantic retrieval on the semantic network of the Internet of Things to determine the target Internet of Things device that meets the device service requirements, including: Determine semantic information of the device service requirement by using the device retrieval paradigm based on the large language model; Determine the semantic similarity between the semantic information of the device description data in the semantic network and the semantic information of the device service requirements based on the large language model; Based on the large language model, a target Internet of Things device that meets the device service requirement is determined according to the semantic similarity.

8. The method according to claim 1, wherein: Based on the device service requirement and the control description data of the target IoT device, controlling the target IoT device to provide a service includes: Parsing the control description data of the target IoT device to obtain a device calling interface and an interface calling command of the target IoT device; Determining interface call parameters of the device call interface according to the device service requirements; Based on the interface calling parameter and the interface calling command, a device calling interface of the target Internet of Things device is called to control the target Internet of Things device to provide a service.

9. An IoT device control device based on a large language model, the device comprising: Demand acquisition module, used to obtain device service requirements for the Internet of Things; A semantic detection module, which is used to perform semantic retrieval on the semantic network of the Internet of Things based on the pre-trained large language model and the device service requirements, and determine the target Internet of Things device that meets the device service requirements and the control description data of the target Internet of Things device; The device control module is used to control the target IoT device to provide services based on the device service requirements and the control description data of the target IoT device.

10. The device according to claim 9, wherein: The semantic network is obtained by adding semantic information to the device description data of the Internet of Things devices in the basic network using the device technical data of the Internet of Things devices.

11. The device according to claim 10, wherein: The semantic information of the device description data is determined in the following manner: based on preset description rules, the device description data and a first speech template, a semantic description paradigm is constructed for the pre-trained large language model; the semantic description paradigm and the device technical data of the Internet of Things device are input into the large language model; the semantic information of the device description data is output through the large language model; wherein the semantic information of the device description data is a natural language and complies with the preset description rules.

12. The device according to claim 10, wherein: The basic network of the Internet of Things is constructed in the following manner: determining a device association relationship between an Internet of Things device in the Internet of Things and the Internet of Things device; and defining an ontology structure of the basic network based on the device association relationship between the Internet of Things devices; The ontology structure is instantiated based on the device description data of the Internet of Things device to obtain the basic network of the Internet of Things; wherein the device description data at least includes control description data.

13. The device according to claim 9, wherein: Semantic detection module, including: A retrieval pattern building submodule, used to build a device retrieval pattern for a pre-trained large language model according to the device service requirements and the second speech template; A semantic search submodule, configured to perform semantic search on the semantic network of the Internet of Things based on the large language model and the device search paradigm, and determine a target Internet of Things device that meets the device service requirement; The control description data acquisition submodule is used to acquire the control description data of the target Internet of Things device from the semantic network of the Internet of Things based on the large language model.

14. The device according to claim 13, wherein: Semantic retrieval submodule, including: A candidate device determination unit, configured to perform semantic retrieval on the semantic network of the Internet of Things using the device retrieval paradigm based on the large language model to obtain candidate Internet of Things devices; A demand supplement acquisition unit, configured to acquire a demand supplement description associated with the device service demand if there are at least two candidate IoT devices; A target device determination unit is used to select a target IoT device that meets the device service requirement from the candidate IoT devices based on the large language model and using the supplementary description of the requirement.

15. The device according to claim 13, wherein: Semantic detection module, including: A semantic information determination submodule, configured to determine the semantic information of the device service requirement based on the large language model and using the device retrieval paradigm; A semantic similarity determination submodule, used for determining the semantic similarity between the semantic information of the device description data in the semantic network and the semantic information of the device service requirements based on the large language model; The target device determination submodule is used to determine the target Internet of Things device that meets the device service requirement based on the large language model and according to the semantic similarity.

16. The device according to claim 9, wherein: Equipment control module, including: A data parsing submodule, used to parse the control description data of the target IoT device to obtain a device calling interface and an interface calling command of the target IoT device; A calling parameter determination submodule, used to determine the interface calling parameters of the device calling interface according to the device service requirements; The device control submodule is used to call the device call interface of the target Internet of Things device based on the interface call parameters and the interface call command to control the target Internet of Things device to provide services.

17. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the Internet of Things device control method based on a large language model as described in any one of claims 1-8.

18. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to enable a computer to execute the Internet of Things device control method based on a large language model according to any one of claims 1-8.

19. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the Internet of Things device control method based on a large language model according to any one of claims 1 to 8.

Images