Semantic model construction method for cloud-edge collaborative system

By using the OPC UA protocol to build semantic models and SPC UA servers in the cloud-edge collaborative system of the space-based Internet of Things, the problem of difficult data on the device is solved, and the semantic interoperability between devices and cloud management is realized, which reduces the Development costs and edge node resource overhead.

CN114221690BActive Publication Date: 2024-11-19NAT UNIV OF DEFENSE TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111545044.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-16
Publication Date
2024-11-19
Estimated Expiration
2041-12-16

AI Technical Summary

Technical Problem

In the space-based Internet of Things, due to the inconsistent standards of various industries, the data on the device side is not easily understood by people and machines, which increases the time cost of upper-level developers. The lack of unified protocols leads to poor data interoperability, affecting the devices. connection and data usage.

Method used

The OPC UA protocol is used to build a semantic model in the cloud-edge collaborative system, and map it to the address space of the device side through the SPC UA server. The Mapper is used to convert the instructions of the MQTT protocol into the OPC UA protocol to realize semantic interoperability on the device side, and Through cloud server management of edge nodes, data interoperability between the cloud and device side is realized.

Benefits of technology

Improve interoperability between devices, reduce time costs for developers, realize the interconnection between human and machine regardless of hardware type, meet the wide-area connectivity and low power consumption requirements of space-based IoT, and reduce Resource overhead for edge nodes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114221690B_ABST
    Figure CN114221690B_ABST
Patent Text Reader

Abstract

The present application relates to a method for constructing a semantic model for a cloud-edge collaborative system, which is implemented in a space-based Internet of Things constructed by the cloud-edge collaborative system. The cloud-edge collaborative system includes a cloud server, a satellite as an edge node, and multiple device terminals. The method constructs a corresponding semantic model and an OPC UA server according to the OPC UA protocol on each device terminal, and maps the constructed semantic model to the OPC UA server to realize abstract representation of the information on the device terminal, thereby completing the semantic interoperability of the cloud server with each device terminal.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of space-based Internet of Things technology, and in particular to a method for constructing a semantic model for a cloud-edge collaborative system. Background Art

[0002] The Internet of Things is the trend of future network development and the core of the Fourth Industrial Revolution. However, the Internet of Things system that relies solely on ground equipment still has great defects. Due to extreme terrain or communication distance and cost restrictions, some remote areas have no network coverage. At present, the coverage rate of ground wireless networks is only 48%. The wide-area coverage of satellite broadcasting networks can solve the problem of ground coverage and the problem of communication interruption caused by natural disasters. Obviously, satellites have become an important part of the Internet of Things. At the same time, satellites are a powerful and feasible supplement to ground networks and future 5G / 6G communications.

[0003] Unlike ground-based network transmission, space-based networks have a wider communication range and can achieve global coverage. In addition, there is almost no need to consider terrain and space limitations when deploying sensors. In the face of various natural hazards and severe weather, space-based networks can still work normally.

[0004] In the traditional space-based IoT architecture, satellites play more of a role as a forwarder and relay, and the ground network and processing center are responsible for data processing, and the processed data is forwarded to users through various means such as mobile networks or the Internet. Considering the long transmission distance and extended transmission time between the ground cloud and the space-based edge nodes, the cloud-edge collaborative architecture is more suitable for the space-based background than the traditional cloud computing architecture.

[0005] For the space-based Internet of Things, the underlying protocols of the connected terminal devices have inconsistent standards, so interoperability is needed to achieve seamless operation on the device side, so as to give full play to the potential of wide-area connectivity of the space-based Internet of Things. However, the underlying equipment currently has no unified standards in various industries, which makes it difficult for people and machines to understand each other's data. This dispersion will increase the time cost of upper-level developers when using data. Therefore, in the space-based Internet of Things, it is necessary to consider how to improve the comprehensibility of data and enhance data interoperability so that people and machines can be connected to each other regardless of the type of hardware used. Summary of the invention

[0006] Based on this, it is necessary to provide a method for constructing a semantic model for a cloud-edge collaborative system that can enhance data interoperability and enable interconnection between people and machines, as well as machines and machines, regardless of the type of hardware on the device side.

[0007] A semantic model construction method for a cloud-edge collaborative system, wherein the semantic model construction method is implemented in the cloud-edge collaborative system, wherein the cloud-edge collaborative system is used to build a space-based Internet of Things, the cloud-edge collaborative system includes a cloud server, a satellite as an edge node, and multiple device terminals, the cloud server receives uploaded data from each device terminal through the edge node and sends instructions to each device terminal, the method includes:

[0008] In each device end, a semantic model corresponding to each device end is constructed according to the OPC UA protocol, and a corresponding SPC UA server is built according to the OPC UA protocol, and the semantic model is mapped to the SPC UA server address space in the corresponding device end;

[0009] When the cloud server sends instructions to each device end through the edge node, the instructions are sent to the edge node, and the edge node publishes the instructions to the MQTT server, and then the Mapper converts the instructions from the MQTT protocol into the OPC UA protocol, and sends the OPC UA protocol to the corresponding device end;

[0010] Each device obtains the corresponding instructions for the received OPC UA protocol through the semantic model in the SPC UA server, and performs corresponding operations according to the instructions.

[0011] In one embodiment, the method further comprises:

[0012] When each of the device ends uploads data to the cloud system, the received data is converted into the corresponding OPC UA protocol through the semantic model constructed by each device end, and then the OPC UA protocol is sent to the edge node;

[0013] The edge node publishes the received OPC UA protocol to the MQTT server, and then the Mapper converts the OPC UA protocol from the MQTT protocol into upload data in a format recognizable by the cloud server, and sends the upload data in this format to the cloud server for use by the cloud server.

[0014] In one embodiment, the method further comprises:

[0015] The cloud server writes a corresponding yaml file according to the semantic model of each device end, then constructs corresponding model metadata through each yaml file, and sends each model metadata to the edge node;

[0016] The edge node constructs corresponding device twins according to the model metadata sent by the cloud server, and stores each device twin in the edge node.

[0017] In one embodiment, when constructing a semantic model corresponding to each device end according to the OPC UA protocol, it includes:

[0018] Defining the device end according to its attributes and related data;

[0019] The defined device side is instantiated according to the OPC UA protocol, that is, the corresponding semantic model is constructed.

[0020] In one embodiment, when building a corresponding SPC UA server according to the OPC UA protocol, the following steps are sequentially included: server initialization, server creation, server parsing of structured files, server operation, and server deletion;

[0021] After the server is created, the constructed semantic model is mapped to the address space of the SPC UA server.

[0022] In one embodiment, the cloud-edge collaboration system adopts an overall design based on KubeEdge and Docker.

[0023] In one embodiment, a device plug-in framework EdgeCore is used between the cloud server and the edge node to report the data received by the edge node to the cloud server and to be responsible for the scheduling selection of the edge node.

[0024] A cloud-edge collaborative system, the cloud-edge collaborative system is used to build a space-based Internet of Things, the cloud-edge collaborative system includes a cloud server, a satellite as an edge node, and multiple device terminals, the cloud server receives uploaded data from each device terminal through the edge node and sends instructions to each device terminal;

[0025] In each device end, a semantic model corresponding to each device end is constructed according to the OPC UA protocol, and a corresponding SPC UA server is built according to the OPC UA protocol, and the semantic model is mapped to the SPC UA server address space in the corresponding device end;

[0026] When the cloud server sends instructions to each device end through the edge node, the instructions are sent to the edge node, and the edge node publishes the instructions to the MQTT server, and then the Mapper converts the instructions from the MQTT protocol into the OPC UA protocol, and sends the OPC UA protocol to the corresponding device end;

[0027] Each device obtains the corresponding instructions for the received OPC UA protocol through the semantic model in the SPC UA server, and performs corresponding operations according to the instructions.

[0028] The above-mentioned semantic model construction method for the cloud-edge collaborative system is implemented in the space-based Internet of Things constructed by the cloud-edge collaborative system, which includes a cloud server, a satellite as an edge node, and multiple device terminals. This method constructs a corresponding semantic model and an OPC UA server according to the OPC UA protocol on each device terminal, and maps the constructed semantic model to the OPC UA server to realize abstract representation of the information on the device terminal, thereby completing the semantic interoperability of the cloud server with each device terminal. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A schematic diagram of the overall system framework in one embodiment;

[0030] Figure 2 A schematic diagram of the original data mapping steps in an embodiment;

[0031] Figure 3 A schematic diagram of modeling a SX1302 gateway in one embodiment;

[0032] Figure 4 FIG. 1 is a schematic diagram of a configuration structure framework of an OPC UA server in an embodiment;

[0033] Figure 5 A schematic diagram of a server process is created in one embodiment;

[0034] Figure 6 A schematic diagram of a server operation function flow in one embodiment;

[0035] Figure 7 A schematic diagram of a process for constructing a Mapper image in an embodiment;

[0036] Figure 8 Schematic diagram of the workflow of a cloud-edge collaborative system in one embodiment. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0038] In view of the current space-based Internet of Things, the low-level devices in various industries have no unified standards, which makes it difficult for people and machines to understand the data. This dispersion will increase the time cost of upper-level developers when using data. In this embodiment, a semantic model construction method for a cloud-edge collaborative system is provided. The semantic model construction method is implemented in the cloud-edge collaborative system, where the cloud-edge collaborative system is used to build a space-based Internet of Things. The cloud-edge collaborative system includes a cloud server, a satellite as an edge node, and multiple device terminals. The cloud server receives uploaded data from each device terminal through the edge node and sends instructions to each device terminal. The method includes:

[0039] In each device end, a semantic model corresponding to each device end is constructed according to the OPC UA protocol, and a corresponding SPC UA server is built according to the OPC UA protocol, and the semantic model is mapped to the SPC UA server address space in the corresponding device end;

[0040] When the cloud server sends instructions to each device end through the edge node, the instruction is sent to the edge node, and the instruction is published to the MQTT server in the edge node. Then the Mapper converts the instruction from the MQTT protocol into the OPC UA protocol and sends the OPC UA protocol to the corresponding device end;

[0041] Each device obtains the corresponding instructions for the received OPC UA protocol through the semantic model in the SPC UA server, and performs corresponding operations according to the instructions.

[0042] It can be seen from this method that this method is actually implemented in the space-based Internet of Things constructed by the cloud-edge collaborative system, and it can be said that this method is inseparable from the system. At the same time, it can be seen from the above that the semantic model construction method proposed in this embodiment is not only a semantic model construction method, but also includes how to realize the semantic interoperability of the cloud server with each device end after the semantic model is constructed on each device end based on this method.

[0043] This method can realize the space-based Internet of Things built on the cloud-edge collaborative system to meet the requirements of application collaboration, service collaboration and resource collaboration. Application collaboration means that the application is managed by the cloud server and run by the edge node. Through open interface standards and platforms, an application development ecosystem is established, which brings together a rich variety of applications. Service collaboration refers to the establishment of standards for edge device data collection and reporting, so that applications can subscribe to various types of business data required from the cloud or edge with unified standards, and realize the collaboration of applications and data services. Resource collaboration refers to the placement of management services, such as monitoring and orchestration, on the cloud server, which greatly reduces the resource overhead of edge satellite nodes and meets the characteristics of light resources, low power consumption and low consumption of edge nodes in satellite scenarios.

[0044] In addition. For semantic modeling, considering that in the field of the Internet of Things, sensor devices collect data, usually each manufacturer is making its own customized solutions and cannot form a unified standard. Manufacturers who do data applications need to adapt to various manufacturers who do device data collection; manufacturers who do sensor data collection need to configure various platform manufacturers, resulting in repeated research and development and low efficiency. In order to improve efficiency, a set of flexible and universal space-based Internet of Things data collection and application architecture must be used to meet the collection of various types of sensor data and the development of various applications as much as possible. It must have the following characteristics:

[0045] (1) Unification: Supports unified management through the cloud;

[0046] (2) Compatibility: Supports adaptation of various IoT protocols and manages various terminal devices and software;

[0047] (3) Universality: Conforms to current mainstream container application management and orchestration specifications;

[0048] (4) Standardization: The cloud side and the edge side provide standard interfaces;

[0049] (5) Large scale: supports access and management of millions of devices;

[0050] (6) Low power consumption: low resource occupation and low power consumption.

[0051] Since the semantic model construction method in this application is based on the cloud-edge collaborative system, the system is first explained. In fact, the cloud-edge collaborative system is a space-based Internet of Things. The entire system platform framework is as follows Figure 1 As shown in the figure, it consists of three parts: a cloud server on the ground (the figure shows the ground cloud), a satellite as an edge node, that is, an edge satellite node, and multiple device terminals. CloudCore is integrated in the cloud server to manage edge nodes, and the data collected by the edge satellite nodes is stored in the ETCD database for unified computing and analysis in the cloud.

[0052] The framework adopts an overall design based on KubeEdge and Docker, combining multiple features such as lightweight, easy to manage, and open. Docker and KubeEdge are combined to build a portable and scalable cloud-edge collaborative computing platform for managing containerized workloads and services, which can promote declarative configuration and automation.

[0053] The platform is separated from the infrastructure and can run on mainstream operating systems such as Ubuntu, RHEL, and CentOS, and can support the deployment of diverse hardware such as ZYNQ-7000 and TX2. Therefore, the platform can be used to discover, monitor, and schedule heterogeneous edge hardware.

[0054] Through the form of a plug-in framework, the computing, storage, network and other resources of edge hardware are modeled and abstracted, allowing different hardware manufacturers to provide plug-in definitions and descriptions for their products, providing a unified resource capability description, deployment, and operation and maintenance management method.

[0055] In this embodiment, there are no excessive requirements for the hardware environment between the cloud server and the edge node, and it is sufficient to be able to run the most basic Linux system. The cloud server can use servers from manufacturers such as Huawei or DELL, and the edge node also supports a variety of heterogeneous devices, such as TX2, ZYNQ-7000 series or devices from other hardware manufacturers.

[0056] The platform provides a device plug-in framework EdgeCore at the edge satellite node to build a bridge between the ground cloud and the satellite edge. It is responsible for reporting satellite edge information to the ground cloud on the one hand, and for scheduling and selecting edge satellites on the other hand. EdgeCore can be deployed manually or automatically.

[0057] At the same time, considering the wide coverage of space-based Internet of Things, a satellite will connect to many sensing devices. The data forms of various devices are different, and different data have different semantics. They cannot be directly operated before the data is modeled accordingly. A unified information model architecture is needed to be compatible with data information of various modes. Therefore, this method uses semantic modeling to abstract the information of the device side to complete the semantic interoperability from the cloud to the device side.

[0058] Semantic interoperability affects the entire information lifecycle, both horizontally across individual devices and systems and vertically across different systems.

[0059] A semantic model is a linking structure onto which raw data can be mapped, and then data can be propagated between these structures to produce new data. After the processed data is mapped to the semantic structure, the data represented by these structures has a concrete meaning.

[0060] For example, if you map a floating point number to a structure that represents the Celsius temperature unit, then the floating point number now represents the Celsius temperature value, whereas before it was just a floating point value without any semantics. Figure 2 The mapping steps from raw data to semantic data are shown.

[0061] The essence of semantic modeling of data is to add metadata or contextual information to the original data, so that the original data has clearer and richer meanings, which is convenient for humans or machines to use. Take temperature sensors as an example. The data of temperature sensors produced by different manufacturers may have different units, which can be Celsius or Fahrenheit. Even if the unit is known in advance, it is far from enough to use its data. You also need to know the time, location and other contextual information of the entire temperature sensor measurement, which requires semantic modeling of the device.

[0062] The challenge of semantic modeling is to ensure that the information exchanged is not only understood by people at both ends of the ICT communication channel, but also by computer systems and their related software. Therefore, a unified standard is needed to regulate the modeling process and ensure that data between different industries can be understood and interoperable with each other. At present, IEC 61804 Parts 3 to 5 and IEC 62541 have constituted standards and ensured machine understandability, and have derived the Electronic Device Description (EDD) and Open Platform Communication Unified Architecture (OPC UA) protocols, respectively. EDD contains metadata, business logic, and user interface descriptions in a machine-interpretable format; OPC UA includes an information model. In addition to the metamodel, the information model also provides standardized serialization, making the type and instance descriptions (node ​​sets) machine-readable, and the information model is designed by individuals for devices connected to the OPC UA server.

[0063] As an application layer protocol in the OSI model, OPC UA has the advantages of cross-platform, unified access, security and reliability, unified information model and good scalability. In addition, the use of OPC UA protocol can easily access the space-based Internet of Things cloud-edge collaborative framework designed in this embodiment to realize the space-based open platform communication unified architecture (SPC UA).

[0064] Therefore, in this embodiment, the OPC UA protocol is used to perform semantic modeling on each device end, wherein the semantic modeling includes defining the device end according to its attributes and related data, and then instantiating the defined device end according to the OPC UA protocol, that is, building a corresponding semantic model.

[0065] Specifically, the basic modeling concept of SPC UA is the reference between nodes. Nodes can be divided into types and instances according to different node categories. Nodes can be connected using references, and each node may have different properties. NodeId is used to uniquely identify a server node. When the client queries, locates, and browses the address space, the server will return the NodeId. References between nodes can be considered as pointers to other nodes. References can be unidirectional or bidirectional. Pointers can reference nodes that do not exist on the server, or they can reference nodes on other servers. References are not nodes and do not have the same properties or characteristics as nodes. SPC UA uses reference types to describe the semantics of references. In the address space, reference types are also used as nodes.

[0066] Furthermore, when semantic modeling is performed on the device side, data is roughly divided into two categories: one is perception data, such as received data, which generally only needs to read field data without the upper-layer user sending control data to them; some are control data, such as power, mode, etc., which not only need to read field data, but also the upper-layer user can send control data to them. All these field data can be described and modeled through the SPC UA specification.

[0067] After defining the device type, you need to instantiate the device type. The instantiated device model is as follows: Figure 3 shown. Figure 3 In the example of the SX1302 gateway as the device side, the relationship between the nodes on the device side is identified. DeviceObjectType inherits the base type (BaseOjectType), and multiple basic types of nodes and methods are constructed under DeviceObjectType.

[0068] SPC UA supports building models through a variety of structured files, such as RDF, XML, and OWL. The constructed structured files are mapped to address spaces for user access.

[0069] After semantic modeling of each device end according to the OPC UA protocol, the corresponding SPC UA server is also built on each device end according to the OPC UA protocol, which includes: server initialization, server creation, server parsing of structured files, server operation and server deletion. After the server is created, the constructed semantic model is mapped to the address space of the SPC UA server. The main function of the SPC UA server is to implement the parsing of the device model and establish a connection with the edge end. The design and development of the SPC UA server is to use the UA stack provided by the SPC Foundation as the basis of the SPC UA architecture, and on this basis, use the API interface provided by the OPC UA open source project Open62541 to design and develop an OPC UA server with corresponding functions according to the requirements. Including, for example, the mapping of the SX1302 gateway model in the address space of the server in the form of a node, and the reading and writing of the node. The main function implementation of the SPC UA server is described in detail below.

[0070] The first step in developing an SPU UA server is to initialize the server and configure it accordingly. Open62541 provides the RegisterServer() method and the UA_ServerConfig() structure method for creating the server and the specific configuration information of the server, including the server IP address, network port number, access control, certificate verification, MDNS discovery, node life cycle callback, secure channel restrictions, session restrictions, operation restrictions, request restrictions, subscription restrictions, monitoring item restrictions, publishing request restrictions, and finally discovery services. Detailed descriptions are as follows Figure 4 As shown, pass the configured structure to the UA_ServerConfig() function to initialize the server. Once the initialization is completed, it cannot be modified during operation.

[0071] After completing the server initialization, Open62541 provides UA_Server_new() to create a server. Figure 5 As shown in the figure, when creating a server, the function will first check whether the server is initialized and create a new server according to the configuration information during initialization. According to the configuration information during initialization, the function will determine the required resource size and apply for the corresponding memory for the server. After completing the memory application work, the function will return a server object for subsequent use.

[0072] After completing the initialization and creation of the SPC UA server, the next step is to map the semantic model corresponding to the device side written previously to the SPC UA server address space, that is, the server parses the structured file. Here, the device side is also taken as an example of the SX1302 gateway. The specific steps are as follows:

[0073] 1) Configure Open62541.

[0074] 2) Generate custom information model code. Copy the previously generated structured files to the relevant directory and execute to generate related dependent files.

[0075] 3) Copy the relevant dependency files to the root directory.

[0076] After completing the mapping of the SX1302 gateway model, you can add UA_Server_run(server,&running).

[0077] The UA_Server_run(server,&running) function is used to run the SPC UA server. Figure 6 As shown, the function will first determine whether the server status flag created previously is 1. After completing the determination, the function will first start the network layer and process various events sent by the network layer, and determine the order of events by sampling the startup time of the server.

[0078] After completing all server work, the created server needs to be deleted. In Open62541, server deletion is implemented through the UA_Server_delete(server) function. Deleting an SPC UA server will first delete the internal data, including the server's configuration information and the device model previously mapped to the server's address space, and release memory. However, deleting a server will not cause the client connection to be lost. The server will only be deleted actively after the client actively cancels the connection to the server.

[0079] After the SPC UA server is built in each device and the corresponding semantic model is mapped to the server, the semantic interoperability of each device is completed. Ultimately, the semantic interoperability of the cloud server with each device needs to be realized, that is, when the cloud sends instructions to each device and receives data uploaded by each device, the edge node needs to perform protocol conversion. In the above, it is mentioned that when the cloud server sends instructions to each device through the edge node, the instruction is sent to the edge node, and the instruction is published to the MQTT server in the edge node. Then the Mapper converts the instruction from the MQTT protocol into the OPC UA protocol and sends the OPC UA protocol to the corresponding device. In each device, the semantic model in the SPC UA server is used to obtain the corresponding instruction for the received OPC UA protocol, and the corresponding operation is performed according to the instruction.

[0080] At the same time, it also includes that when each device uploads data to the cloud, the received data is converted into the corresponding OPC UA protocol through the semantic model constructed by each device, and then the OPC UA protocol is sent to the edge node; the edge node publishes the received OPC UA protocol to the MQTT server, and then the Mapper converts the OPC UA protocol from the MQTT protocol into the uploaded data in a format recognizable by the cloud server, and sends the uploaded data in this format to the cloud server for use by the cloud server.

[0081] In this embodiment, the cloud server writes the corresponding yaml file according to the semantic model of each device end, then constructs the corresponding model metadata through each yaml file, and sends each model metadata to the edge node.

[0082] The edge node builds the corresponding device twin based on the metadata of each model sent by the cloud server, and stores each device twin in the edge node.

[0083] Specifically, the cloud-edge collaborative system manages devices with the help of custom resource descriptors (CRDs) and the mapper to which the device is connected. CRDs are used to describe device metadata / status, and controllers are used to synchronize these device updates between edge nodes and cloud servers. Therefore, if you want to manage a device through the cloud, you need to define the CRD of the device first. CRD has the following functions.

[0084] 1) Describing device properties: Users can describe device properties and access mechanisms to interact with or control the device.

[0085] 2) Perform CRUD operations on the device side from the cloud server: Users can create, update, and delete device metadata from the cloud through the CRD API exposed by the server. The desired state of the device can be controlled through the CRD API.

[0086] 3) Reporting device attribute values: Mapper applications running at the edge can report the current values ​​of device attributes.

[0087] In this embodiment, two types of CRDs are supported, one is device model CRD and the other is device instance CRD.

[0088] To control the device, you must first configure the device model CRD and device instance CRD. The framework will generate a configmap file based on the configuration file. When running for the first time, Mapper will parse the configmap file and perform corresponding operations based on the parsed commands. During runtime, once the configuration changes, the configmap file will change accordingly, and the device twin will send the changed message to the MQTT agent. Mapper obtains the message by subscribing to the topic.

[0089] The device model describes the device attributes exposed by the device and the accessors to these attributes. The device model is like a reusable template that can be used to create and manage many devices. The device model CRD contains the name of each attribute, the description of the attribute, the read and write permission nodes of the attribute, as well as the node namespace and identifier.

[0090] Create an instance of the device (device instance CRD) based on the device model. The device instanceCRD represents the device model CRD referenced by the device instance CRD, defines the device specific protocol, and also defines the OPC UA protocol address and the specific edge node name issued.

[0091] When achieving semantic interoperability between the cloud and each device, it is also necessary to build a Mapper on the edge node, that is, the satellite, to realize protocol conversion between the cloud and the device.

[0092] After completing the creation of the device model and instance, you need to build a Mapper image and send it to the corresponding edge node to ensure that the edge node can connect to the device and parse the OPC UA protocol. The process of building a Mapper image is as follows: Figure 7 shown.

[0093] After the Mapper image is uploaded to the central warehouse, you need to configure the Deployment file to bind the Mapper and the configmap files generated by the device model CRD and device instance CRD to the corresponding edge nodes.

[0094] In the cloud-edge collaborative system, the workflow to achieve interoperability between each device and the cloud server can also be as follows: Figure 8 shown.

[0095] The above-mentioned semantic model construction method for the cloud-edge collaborative system is implemented in the space-based Internet of Things constructed by the cloud-edge collaborative system, which includes a cloud server, a satellite as an edge node, and multiple device terminals. This method constructs a corresponding semantic model and an OPC UA server according to the OPC UA protocol on each device terminal, and maps the constructed semantic model to the OPC UA server to realize abstract representation of the information on the device terminal, thereby completing the semantic interoperability of the cloud server with each device terminal.

[0096] This application also provides a cloud-edge collaboration system, the system structure of which can be referred to Figure 1 ,The cloud-edge collaborative system is used to build the space-based Internet of Things. The cloud-edge collaborative system includes a cloud server, a satellite as an edge node, and multiple device terminals. The cloud server receives the uploaded data from each device terminal through the edge node and sends instructions to each device terminal;

[0097] In each device end, a semantic model corresponding to each device end is constructed according to the OPC UA protocol, and a corresponding SPC UA server is built according to the OPC UA protocol, and the semantic model is mapped to the SPC UA server address space in the corresponding device end;

[0098] When the cloud server sends instructions to each device end through the edge node, the instructions are sent to the edge node, and the edge node publishes the instructions to the MQTT server, and then the Mapper converts the instructions from the MQTT protocol into the OPC UA protocol, and sends the OPC UA protocol to the corresponding device end;

[0099] Each device obtains the corresponding instructions for the received OPC UA protocol through the semantic model in the SPC UA server, and performs corresponding operations according to the instructions.

[0100] In one embodiment, the method further comprises:

[0101] When each of the device ends uploads data to the cloud system, the received data is converted into the corresponding OPC UA protocol through the semantic model constructed by each device end, and then the OPC UA protocol is sent to the edge node;

[0102] The edge node publishes the received OPC UA protocol to the MQTT server, and then the Mapper converts the OPC UA protocol from the MQTT protocol into upload data in a format recognizable by the cloud server, and sends the upload data in this format to the cloud server for use by the cloud server.

[0103] In one embodiment, the method further comprises:

[0104] The cloud server writes a corresponding yaml file according to the semantic model of each device end, then constructs corresponding model metadata through each yaml file, and sends each model metadata to the edge node;

[0105] The edge node constructs corresponding device twins according to the model metadata sent by the cloud server, and stores each device twin in the edge node.

[0106] In one embodiment, when constructing a semantic model corresponding to each device end according to the OPC UA protocol, it includes:

[0107] Defining the device end according to its attributes and related data;

[0108] The defined device side is instantiated according to the OPC UA protocol, that is, the corresponding semantic model is constructed.

[0109] In one embodiment, when building a corresponding SPC UA server according to the OPC UA protocol, the following steps are sequentially included: server initialization, server creation, server parsing of structured files, server operation, and server deletion;

[0110] After the server is created, the constructed semantic model is mapped to the address space of the SPC UA server.

[0111] In one embodiment, the cloud-edge collaboration system adopts an overall design based on KubeEdge and Docker.

[0112] In one embodiment, a device plug-in framework EdgeCore is used between the cloud server and the edge node to report the data received by the edge node to the cloud server and to be responsible for the scheduling selection of the edge node.

[0113] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0114] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0115] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A semantic model construction method for a cloud-edge collaborative system, characterized in that: The semantic model construction method is implemented in the cloud-edge collaborative system, wherein the cloud-edge collaborative system is used to build a space-based Internet of Things, the cloud-edge collaborative system includes a cloud server, a satellite as an edge node, and multiple device terminals, the cloud server receives uploaded data from each device terminal through the edge node and sends instructions to each device terminal, the method includes: In each device end, a semantic model corresponding to each device end is constructed according to the OPC UA protocol, wherein semantic modeling includes defining the device end according to its attributes and related data, and instantiating the defined device end according to the OPC UA protocol to construct the corresponding semantic model; a corresponding SPC UA server is also built according to the OPC UA protocol, which includes: server initialization, server creation, server parsing of structured files, server operation, and server deletion, and after the server is created, the semantic model is mapped to the SPC UA server address space in the corresponding device end, and the SPCUA server implements the parsing of the semantic model corresponding to the device end and establishes a connection with the edge node; When the cloud server sends instructions to each device end through the edge node, the instructions are sent to the edge node, and the edge node publishes the instructions to the MQTT server, and then the Mapper converts the instructions from the MQTT protocol into the OPC UA protocol, and sends the OPC UA protocol to the corresponding device end; Each device obtains the corresponding instructions for the received OPC UA protocol through the semantic model in the SPC UA server, and performs corresponding operations according to the instructions.

2. The semantic model construction method according to claim 1, characterized in that: The method further comprises: When each of the device ends uploads data to the cloud system, the received data is converted into the corresponding OPC UA protocol through the semantic model constructed by each device end, and then the OPC UA protocol is sent to the edge node; The edge node publishes the received OPC UA protocol to the MQTT server, and then the Mapper converts the OPC UA protocol from the MQTT protocol into upload data in a format recognizable by the cloud server, and sends the upload data in this format to the cloud server for use by the cloud server.

3. The semantic model construction method according to claim 2, characterized in that: The method further comprises: The cloud server writes a corresponding yaml file according to the semantic model of each device end, then constructs corresponding model metadata through each yaml file, and sends each model metadata to the edge node; The edge node constructs corresponding device twins according to the model metadata sent by the cloud server, and stores each device twin in the edge node.

4. The method for constructing a semantic model according to claim 1, characterized in that: When building a semantic model corresponding to each device end according to the OPC UA protocol, it includes: Defining the device end according to its attributes and related data; The defined device side is instantiated according to the OPC UA protocol, that is, the corresponding semantic model is constructed.

5. The method for constructing a semantic model according to any one of claims 1 to 4, characterized in that: The cloud-edge collaboration system adopts an overall design based on KubeEdge and Docker.

6. The method for constructing a semantic model according to claim 5, characterized in that: The device plug-in framework EdgeCore is used between the cloud server and the edge node to report the data received by the edge node to the cloud server and to be responsible for the scheduling selection of the edge node.