A pan-Beidou terminal equipment management method and system

Abstract

The present application discloses a pan-Beidou terminal equipment management method and system, which belongs to the field of Beidou terminal equipment management technology. The method comprises the following steps: step S1, building a modular framework based on microservice architecture to ensure that each function is implemented through an independent module; step S2, designing a loosely coupled module environment, developing module units, and combining and configuring modules through user permissions to build a personalized workflow; step S3, integrating remote sensing monitoring and Beidou positioning technology, real-time collection and pre-processing of natural resource data, including geological disaster monitoring data, positioning terminal location and status data; step S4, providing natural resource analysis, geological disaster risk assessment, tracking and trajectory analysis, visualization and report generation, as well as event triggering and alarm mechanism underlying function implementation; step S5, supporting function expansion through API interface, allowing users to access third-party services or develop new modules, and supporting the import and export of workflows.

Description

Technical Field

[0001] The present application relates to the technical field of Beidou terminal equipment management, and more specifically, to a pan-Beidou terminal equipment management method and system. Background Art

[0002] With the gradual improvement and widespread application of the Beidou satellite navigation system, Beidou terminal devices are becoming more and more popular in various fields. These devices are not only used for positioning and navigation, but also play an important role in many industries such as traffic management, logistics monitoring, and agricultural management. However, with the surge in the number of terminal devices, the management, maintenance, and data processing of the equipment are facing unprecedented challenges.

[0003] Existing Beidou terminal equipment management methods mostly rely on manual management and simple database records, which have the following problems: First, the lack of real-time monitoring of equipment status makes it difficult to find and handle faults in a timely manner; second, the storage and analysis capabilities of equipment data are limited, which cannot effectively support the needs of the big data era; third, the lack of a unified management platform leads to data islands between different devices, making information sharing and collaborative work difficult. In addition, the update and upgrade management of equipment also lacks an effective mechanism and cannot adapt to the rapidly changing market needs.

[0004] At present, many industries are still facing problems such as insufficient equipment performance monitoring, poor data integration, and low management efficiency when using Beidou terminal equipment. These problems not only affect the efficiency of equipment use, but may also lead to waste of resources and increase of safety hazards.

[0005] To sum up, how to effectively manage Beidou terminal equipment and improve the equipment's utilization efficiency and data processing capabilities has become a technical problem that needs to be solved urgently. Summary of the invention

[0006] In order to overcome a series of defects in the prior art, the purpose of this application is to provide a pan-Beidou terminal equipment management method for the above-mentioned problems, including the following steps.

[0007] Step S1, build a modular framework based on microservice architecture to ensure that each function is implemented through an independent module.

[0008] Step S2, design a loosely coupled module environment, develop module units, combine and configure modules through user permissions to build a personalized workflow.

[0009] Step S3, integrating remote sensing monitoring and Beidou positioning technology, real-time collection and pre-processing of natural resource data, including geological disaster monitoring data, positioning terminal location and status data.

[0010] Step S4 provides the underlying functional implementation of natural resource analysis, geological disaster risk assessment, tracking and trajectory analysis, visualization and report generation, as well as event triggering and alarm mechanism.

[0011] Step S5, supporting function extension through API interface, allowing users to access third-party services or develop new modules, and supporting the import and export of workflows.

[0012] Furthermore, in step S1, a modular framework based on the microservice architecture is built through the following steps.

[0013] Step S101, split the function into multiple independent modules.

[0014] Step S102, designing a standardized API interface, allowing direct calling of existing functional modules, or supporting the development and access of new modules to achieve flexible custom expansion.

[0015] Step S103, configure the database and design an efficient data management system to support data persistence, fast query, index creation and data stream interaction of different modules to ensure efficient data processing.

[0016] Step S104, introduce a role authority management mechanism to strictly control the user's access rights to function modules and data.

[0017] Step S105, deploy edge computing nodes to process large-scale real-time data streams, implement data preprocessing and localized storage, effectively reduce the load on the main server, and improve the real-time processing efficiency of data.

[0018] Furthermore, in step S2, the implementation process of the loosely coupled module environment includes the following steps.

[0019] Step S201, develop a standardized module library, implement functional encapsulation and classification of modules, and design a permission control interface to support module calls based on user roles.

[0020] Step S202, creating an independent parameter configuration interface for each module, supporting dynamic definition of input / output data flows, and providing a configuration tool to customize the execution logic of the module.

[0021] Step S203, connect different modules through permission verification, define data flow and execution order, and form a complete workflow.

[0022] Step S204 introduces a permission-based intelligent matching mechanism to recommend available module combinations and workflow templates according to the user's permission level and operation requirements, helping the user to quickly build a workflow that meets the permissions.

[0023] Furthermore, the core of permission-based intelligent matching is to achieve accurate matching between user needs and functional modules, including the following key links.

[0024] Abstract user needs into a set of demand keywords K={k1,k2,...,k μ}, where k1, k2, ..., k μ represents different keywords of user needs, and μ represents the number of keywords of user needs.

[0025] For each module M w Assign a feature label set T(M w )={t1,t2,...,t m}, where t1, t2,...,t m Represents module M w The feature label of m represents the module M w The number of feature labels.

[0026] By calculating the relevance score Relevance(M w )=|K∩T(M w )∣ / ∣T(M w )|×P(Auth) to evaluate M w The degree of matching with user requirements, where P(Auth) represents the user's authority coefficient for the module.

[0027] According to the task type T selected by the user task , from the preset standard module set M task ={M1, M2,...,M o} to quickly build workflows that meet the needs of specific scenarios, where M1, M2,...,M o Represents different modules; o represents the number of modules in the preset standard module set.

[0028] Enrich the keyword set entered by the user through synonym expansion or keyword derivation to more comprehensively capture user needs.

[0029] Furthermore, permission-based intelligent matching provides personalized module suggestions for workflow construction by analyzing user behaviors and preferences, including the following steps.

[0030] Guide users to enter or select specific requirements and provide initial recommendations based on requirements and user permissions.

[0031] According to the permission level, the modules that are not available are adjusted or restricted, and the intelligent matching mechanism is continuously optimized based on the user’s feedback. The mathematical model is as follows: Let the user’s permission to access the module be Auth(Mw), and the probability of accepting the recommended module be P(Accept(Mw)). w )), adjust the module's recommendation weight W(M) according to the acceptance or rejection operation w ), the adjustment formula is: W(M w ) new =W(M w ) old ×P(Accept(M w ))×Auth(Mw), through this mechanism, the weight of modules that are frequently accepted and have permissions will be increased, and vice versa, thereby improving the accuracy of recommendations.

[0032] Furthermore, in step S3, the implementation of real-time collection and preprocessing of natural resource data includes the following steps.

[0033] Step S301, based on the adaptability of the time window, regularly acquire high-resolution remote sensing images covering the preset natural resource monitoring area and geological disaster risk area to ensure that the acquired image data can effectively support resource identification and risk assessment.

[0034] Step S302, preprocessing the acquired remote sensing image, including image correction, denoising and enhancement, to improve image quality.

[0035] Step S303, deploy Beidou positioning terminals at the boundaries and key monitoring points of the natural resource monitoring area to collect positioning data in real time to track the dynamic changes of natural resources and update resource status information in a timely manner.

[0036] Step S304, deploy Beidou positioning terminals in high-risk areas for geological disasters, monitor ground displacement and changes in real time, and combine meteorological data to ensure timely identification of potential geological disaster risks. The data of Beidou positioning terminals are transmitted according to the JT808 protocol.

[0037] Furthermore, the specific implementation of natural resource analysis includes the following steps: Step S401, identifying and classifying natural resources to determine the spatial distribution range of each type of resources; Step S402, evaluating the distribution characteristics, dynamic changes and potential risks of different types of resources, determining the boundaries and key monitoring points of each type of natural resources, and deploying Beidou positioning terminals; Step S403, based on the data collected by the Beidou positioning terminal, continuously monitoring the dynamic changes of natural resources to achieve accurate positioning of the resource range and real-time updating of the status.

[0038] The specific implementation of geological disaster risk assessment includes the following steps: Step S404, using regularly acquired remote sensing images to identify potential risk areas and quantify risk levels; Step S405, real-time monitoring of ground displacement and changes in high-risk areas through Beidou positioning terminals, and combining monitoring data with risk assessment results to achieve real-time updating and early warning of risk levels.

[0039] The specific implementation of tracking and trajectory analysis includes the following steps: Step S406, locating the known positioning terminal, obtaining its real-time location information and updating it dynamically in parallel, including longitude and latitude, timestamp, speed and direction information; Step S407, establishing a historical trajectory database to support users to retrieve the historical movement trajectory of a specific terminal based on a time range, and providing a trajectory playback function to intuitively display the historical operation status of the trajectory; Step S408, integrating geographic location, real-time status and emergency event information, and providing real-time scheduling decision support and optimal path planning suggestions.

[0040] The specific implementation of visualization and report generation includes the following steps: Step S409, by rendering the natural resource distribution and terminal positioning data into vector tiles in real time, a multi-level and multi-dimensional dynamic display is performed in the user interface, supporting users to flexibly adjust the hierarchical structure and display method of data visualization according to actual needs; Step S410, automatically integrating natural resource monitoring, geological disaster warning and scheduling information to generate a customizable report, and providing a variety of export format options (such as PDF, Excel, etc.) to meet different data analysis needs.

[0041] The specific implementation of the event triggering and alarm mechanism includes the following steps: Step S411, real-time monitoring of the location changes of the positioning terminal through geographic fence technology, triggering an alarm notification when the selected or unknown positioning terminal leaves or enters the fence; Step S412, combining geological disaster warning data, automatically generating alarm information and pushing it to relevant users; Step S413, introducing an abnormal behavior detection mechanism based on a rule engine, automatically triggering an event when the positioning terminal exhibits abnormal behavior, starting the corresponding analysis process or sending a notification.

[0042] Furthermore, in step S401, the remote sensing image is classified at the pixel level using the UNet semantic segmentation model, including the following steps: extracting features of the input remote sensing image I through a convolutional neural network to obtain a feature map F=CNN(I); inputting the extracted feature map F into the UNet semantic segmentation model to generate a pixel classification probability map P=UNet(F) of the same size as the original image, wherein P represents the probability distribution of each pixel belonging to a different resource type; performing threshold segmentation on the probability map to obtain a boundary map B of each resource category. q =Threshold(P q ), where B qrepresents the boundary area of ​​resource category q; Threshold is a threshold segmentation function used to transform the probability map P q Transformed into clear category boundaries; P q Represents the probability graph associated with resource category q.

[0043] In step S402, the dynamic changes of natural resources are evaluated by the following formula: future =f(R t ,R t+1 ,...,R g ), where R future is the changing trend of natural resources, R t ,R t+1 ,...,R g is the resource range in different periods; f() is the time series prediction function; g represents the last moment in the time series of the resource range.

[0044] In step S403, the dynamic changes of natural resources are continuously monitored by the following formula: ΔR=R t+1 -L t , where ΔR is the change in natural resources; L t Represents the position coordinates of the Beidou positioning terminal at time t; R t+1 is the resource range at time t+1.

[0045] Step S405 includes the following steps.

[0046] The surface displacement vector ΔL is calculated by the following formula i :ΔL i =L t+1 i -L t i , where L t i and L t+1 i are the positions of Beidou positioning terminal i at time t and t+1 respectively.

[0047] The amplitude of the ground displacement |ΔL is calculated by the following formula i |:|ΔL i ∣=[(Δx i ) 2 +(Δy i ) 2 +(Δz i ) 2 ] 1 / 2 , where (Δx i , Δy i , Δz i) represents the displacement change of Beidou positioning terminal i in three spatial dimensions.

[0048] The high-risk area is modeled as a graph structure G=(V,E), where V represents the node set, E represents the edge set, and node v γ ∈V represents the Beidou positioning terminal, edge e γj ∈E represents node v γ and v j The interaction between them.

[0049] By passing information, node features are updated. γ The feature at time t is denoted as h γ (t) , the update formula is: γ (t+1) =σ[W (t) ∑ j∈N(γ) h j (t) +b (t) ], where W (t) and b (t) is a learnable parameter, N(γ) is the neighbor node of node γ, σ is the activation function; h γ (t+1) is the updated feature representation, h j (t) Represents node v j Feature representation at time t.

[0050] By learning the mutual influence between nodes, the geological disaster risk of the node is predicted and the disaster risk level is evaluated. Finally, the node represents h γ (T) The disaster risk prediction P is obtained through the fully connected layer γ , expressed as: P γ =Sigmoid(W final h γ (T) +b final ), where P γ is the risk probability of node γ, indicating the possibility of disasters occurring in the area; W final Represents the weight matrix of the final fully connected layer in the convolutional neural network, which is used to represent the final node h γ (T) Converted to disaster risk prediction γ ; b final Represents the bias term of the final fully connected layer in the convolutional neural network.

[0051] Further, in step S5, the functional extension is configured to be implemented.

[0052] Access new data sources, algorithm models or external systems through open API interfaces.

[0053] The development of custom modules allows users to integrate new functional modules into existing systems through APIs to achieve dynamic expansion of functions.

[0054] Importing and exporting modules allows users to export configured workflows and import them into other projects or systems, facilitating team collaboration and deployment.

[0055] Template saving allows users to save commonly used modules as reusable templates to improve subsequent deployment efficiency.

[0056] Sharing of module libraries allows downloading modules or workflow templates from other users, reducing duplication of work and improving collaboration efficiency.

[0057] Dynamic loading of modules: dynamically load the required functional modules according to the actual needs of users, or unload unnecessary modules according to the scenarios and performance requirements.

[0058] The purpose of this application is also to provide a pan-Beidou terminal equipment management system, including the following modules.

[0059] The remote sensing data acquisition module is responsible for collecting satellite remote sensing data of the selected area, performing preprocessing, and providing the corresponding calling interface.

[0060] The positioning data acquisition module is responsible for collecting the location and movement information of the relevant positioning terminals, performing preprocessing, and providing the corresponding calling interface.

[0061] The natural resource identification module, based on processed remote sensing data, uses semantic segmentation technology to determine the boundaries of various natural resources in the selected area, and provides corresponding output data and calling interfaces.

[0062] The geological disaster risk assessment module uses processed remote sensing data to identify high-risk areas for geological disasters through convolutional neural networks, and provides corresponding output data and calling interfaces.

[0063] The natural resource analysis module accurately determines the location and area of ​​various natural resources by deploying positioning terminals at the original boundaries of various natural resources, and compares them with the newly acquired remote sensing semantic segmentation data.

[0064] The geological disaster determination module uses a group of positioning terminals deployed in high-risk areas to identify potential geological disaster locations through a graph neural network.

[0065] The terminal positioning module locates known positioning terminals in real time and dynamically updates their location information, including latitude and longitude, timestamp, speed and direction information.

[0066] The historical trajectory query module records the historical trajectory of the positioning terminal, queries the movement path within a specific time period by time range, and supports historical status playback.

[0067] The scheduling optimization module plans and schedules the paths of the devices carrying the positioning terminals based on the destination, the location of the positioning terminals and the road conditions.

[0068] The visualization module uses dynamic vector tile technology to display the input parameters of the connection module on the map of the user interface, supports the visualization and filtering of spatiotemporal data, and adjusts the view hierarchy and display method according to needs.

[0069] The report generation module generates customized reports based on the parameters selected by the user or the parameters output by the selected module, and exports them to PDF, Excel or other formats as required.

[0070] The geo-fence module is responsible for setting geo-fences for the selected area and triggering alarm notifications when the positioning terminal leaves or enters the fence.

[0071] The geological disaster early warning module automatically generates warning information and pushes it to relevant users based on the output data of the geological disaster risk assessment module and the geological disaster determination module.

[0072] The abnormal warning module for positioning terminals monitors abnormal behaviors of positioning terminals through a built-in rule engine and automatically starts related analysis processes or sends notifications.

[0073] The loosely coupled workflow building module provides a modular combination environment. Through module combination and configuration based on user permissions, personalized workflows are built, and an open API interface is provided to access new data sources, algorithm models or external systems.

[0074] Custom development modules allow users to integrate new functional modules into existing systems through APIs to achieve dynamic expansion of functions.

[0075] The permission management module is responsible for the permission control and export functions of the module, allowing users to export the configured workflow and reuse it in other projects or systems. It supports saving commonly used modules as templates and provides a permission-based shared module library for other users to download, reducing duplication of work.

[0076] Compared with the prior art, the present application has the following beneficial effects.

[0077] This application uses a microservice architecture and a visual programming environment to achieve modular management and personalized workflow construction, integrate remote sensing monitoring and Beidou positioning technology, realize real-time data collection, analysis, early warning and flexible expansion, and improve management efficiency and data processing capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 This is a flow chart of a pan-Beidou terminal device management method disclosed in an embodiment of the present application.

[0079] Figure 2 This is a structural diagram of a pan-Beidou terminal equipment management system disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0080] In order to make the purpose, technical scheme and advantages of the implementation of the present invention clearer, the technical scheme in the embodiment of the present invention will be described in more detail below in conjunction with the drawings in the embodiment of the present invention. In the drawings, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The described embodiments are part of the embodiments of the present invention, not all of the embodiments.

[0081] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0082] The embodiments and directional terms described below with reference to the accompanying drawings are exemplary and intended to be used to explain the present invention, but should not be construed as limiting the present invention.

[0083] like Figure 1 As shown, a pan-Beidou terminal equipment management method includes the following steps.

[0084] Step S1, build a modular framework based on microservice architecture to ensure that each function is implemented through an independent module.

[0085] Step S2, design a loosely coupled module environment, develop module units, combine and configure modules through user permissions to build a personalized workflow.

[0086] Step S3, integrating remote sensing monitoring and Beidou positioning technology, real-time collection and pre-processing of natural resource data, including geological disaster monitoring data, positioning terminal location and status data.

[0087] Step S4 provides the underlying functional implementation of natural resource analysis, geological disaster risk assessment, tracking and trajectory analysis, visualization and report generation, as well as event triggering and alarm mechanism.

[0088] Step S5, supporting function extension through API interface, allowing users to access third-party services or develop new modules, and supporting the import and export of workflows.

[0089] Furthermore, in step S1, a modular framework based on the microservice architecture is built through the following steps.

[0090] Step S101, split the function into multiple independent modules.

[0091] Step S102, designing a standardized API interface, allowing direct calling of existing functional modules, or supporting the development and access of new modules to achieve flexible custom expansion.

[0092] Step S103, configure the database and design an efficient data management system to support data persistence, fast query, index creation and data stream interaction of different modules to ensure efficient data processing.

[0093] Step S104, introduce a role authority management mechanism to strictly control the user's access rights to function modules and data.

[0094] Step S105, deploy edge computing nodes to process large-scale real-time data streams, implement data preprocessing and localized storage, effectively reduce the load on the main server, and improve the real-time processing efficiency of data.

[0095] Furthermore, in step S2, the implementation process of the loosely coupled module environment includes the following steps.

[0096] Step S201, develop a standardized module library, implement functional encapsulation and classification of modules, and design a permission control interface to support module calls based on user roles.

[0097] Step S202, creating an independent parameter configuration interface for each module, supporting dynamic definition of input / output data flows, and providing a configuration tool to customize the execution logic of the module.

[0098] Step S203, connect different modules through permission verification, define data flow and execution order, and form a complete workflow.

[0099] Step S204 introduces a permission-based intelligent matching mechanism to recommend available module combinations and workflow templates according to the user's permission level and operation requirements, helping the user to quickly build a workflow that meets the permissions.

[0100] In step S201, the module library adopts a loosely coupled plug-in architecture, provides permission-based, plug-and-play functional modules, supports various data retrieval, processing and function calls, to ensure that the functional modules can be dynamically loaded; each module represents an independent functional unit and is defined as a function or containerized service, with standardized input and output interfaces and permission control interfaces, supporting independent operation and permission-based calls.

[0101] In step 202, the parameter configuration interface implements the connection operation between modules through permission verification. When connecting two modules, the user permission is first verified, and then the nodes and edges in the data flow engine are dynamically updated to save the workflow configuration in real time. w ={p1,p2,...,p u}Module M w , these parameters can be adjusted within the user's authority range through the configuration interface to meet different needs, where p1, p2, ..., p u Indicates M w Multiple parameters of P w Module representation M w The parameter vector of the module M w The number of parameters in .

[0102] In step 203, the execution order and data transfer method are defined through permission verification. In the backend, a data flow engine that supports a directed acyclic graph (DAG) structure is designed to ensure that the data flow does not form a loop and supports asynchronous operations to improve processing efficiency. Data transfer between modules is represented by D wj =M out w →M in j ×Auth (w,j) , where D ij is the data flow, M out w is the output data of module w, which is passed to the input M of module j in j ,Auth (w,j) Represents the user's authority coefficient for the data flow from module w to j.

[0103] In step 204, the user can trigger the permission-based intelligent matching mechanism by entering the required keywords or selecting a specific task type. Each module and function is assigned a specific label or keyword, and a suitable module combination is recommended by matching the keywords of the user's requirements and verifying the user's permissions; in addition, common task flow templates based on permission levels are preset. For example, when the user selects "natural resource monitoring", an available complete workflow template including "remote sensing data collection", "natural resource analysis" and "data visualization" will be automatically recommended according to the user's permissions, so that the template can be directly applied or further adjusted within the user's permissions.

[0104] The core of the permission-based intelligent matching mechanism is to accurately match user needs with functional modules. Its technical implementation involves several key links: First, the user's needs are abstracted into a set of demand keywords K={k1,k2,...,k μ}, where k1, k2, ..., kμ Represents different keywords of user needs, μ represents the number of keywords of user needs; at the same time, for each module M w Assign a feature label set T(M w )={t1,t2,...,t m}, where t1, t2,...,t m Represents module M w The feature label of m represents the module M w Based on the demand keyword set and the feature tag set, the relevance score Relevance (M w )=|K∩T(M w )∣ / ∣T(M w )|×P(Auth) to evaluate M w The matching degree of user needs, the higher the score, the more the module meets the user needs, where P (Auth) represents the user's authority coefficient for the module; according to the task type T selected by the user task , from the preset standard module set M task ={M1, M2,...,M o} to quickly build workflows that meet the needs of specific scenarios, where M1, M2,...,M o represents different modules; o represents the number of modules in the preset standard module set. In addition, to improve the accuracy of recommendations, the keyword set entered by the user is enriched through synonym expansion or keyword derivation. For example, when the user enters "geological disaster monitoring", it will automatically expand it to "landslide monitoring", "displacement detection", "early warning" and other related keywords, so as to more comprehensively capture user needs. Through this multi-dimensional intelligent matching and recommendation strategy, it can provide accurate and efficient module combination suggestions for users with different backgrounds and needs, significantly improving the efficiency and quality of workflow construction.

[0105] The permission-based intelligent matching mechanism plays a key role in the workflow construction process, and its implementation involves multiple technical links: First, guide users to input or select specific requirements, and provide initial recommendations based on the requirements and user permissions. The recommendation interface is designed with clear levels, displaying the module's functional description, permission requirements, and relevance scores based on the user's permission level, and supporting users to add modules to the workflow with one click within their permission range. Adjust or restrict unavailable modules according to permission levels, and continuously optimize the intelligent matching mechanism based on user feedback. Specifically, if some recommended modules are frequently rejected or exceed the user's permission range, the recommendation weights of these modules for users of that permission level will be reduced, thereby gradually optimizing the recommendation quality. This process can be represented by a mathematical model: let the user's permission to access the module be Auth(Mw), and the probability of accepting the recommended module be P(Accept(Mw)). w )), adjust the module's recommendation weight W(M) according to the acceptance or rejection operation w ), the adjustment formula is: W(M w ) new =W(M w ) old ×P(Accept(M w ))×Auth(Mw), through this mechanism, the weight of modules that are frequently accepted and for which users have permissions will be increased, and vice versa, thereby improving the accuracy of recommendations.

[0106] In addition, a variety of common task types are preset, and each task type predefines a set of typical module processes to help quickly build workflows. For example, when the user selects "natural resource monitoring", modules such as "remote sensing data collection", "natural resource identification", and "visualization display" will be automatically recommended. This recommendation process can be formally expressed as follows: For task type T task , recommended module set M recommended It consists of a standard module set and a user-defined module set, namely M recommended =M task +M custom , where M custom It is a custom module added according to the user's additional requirements.

[0107] As user needs increase and usage continues, the module library will continue to expand. After completing a specific task, users can save their own module combinations as templates and share them in the module library for other users to use directly. This user-driven module library expansion mechanism not only enriches the functions, but also improves the diversity and applicability of recommendations. Through this comprehensive intelligent recommendation and adaptive optimization strategy, it can provide more accurate and efficient workflow construction support for users with different backgrounds and needs, significantly improving user experience and work efficiency.

[0108] Furthermore, in step S3, the implementation of real-time collection and preprocessing of natural resource data includes the following steps.

[0109] Step S301, based on the adaptability of the time window, regularly acquire high-resolution remote sensing images covering the preset natural resource monitoring area and geological disaster risk area to ensure that the acquired image data can effectively support resource identification and risk assessment.

[0110] Step S302, preprocessing the acquired remote sensing image, including image correction, denoising and enhancement, to improve image quality.

[0111] Step S303, deploy Beidou positioning terminals at the boundaries and key monitoring points of the natural resource monitoring area to collect positioning data in real time to track the dynamic changes of natural resources and update resource status information in a timely manner.

[0112] Step S304, deploy Beidou positioning terminals in high-risk areas for geological disasters, monitor ground displacement and changes in real time, and combine meteorological data to ensure timely identification of potential geological disaster risks. The data of Beidou positioning terminals are transmitted according to the JT808 protocol.

[0113] Furthermore, in step S4, the specific implementation of natural resource analysis includes the following steps.

[0114] Step S401, identifying and classifying natural resources, thereby determining the spatial distribution range of each type of resources.

[0115] Step S402, evaluate the distribution characteristics, dynamic changes and potential risks of different types of resources, determine the boundaries and key monitoring points of various natural resources, and deploy Beidou positioning terminals.

[0116] Step S403, based on the data collected by the Beidou positioning terminal, the dynamic changes of natural resources are continuously monitored to achieve accurate positioning of the resource range and real-time update of the status.

[0117] In step S401, the UNet semantic segmentation model is used to classify the remote sensing image at the pixel level. Specifically, first, the convolutional neural network CNN is used to extract features from the input remote sensing image I to obtain a feature map F=CNN(I), which captures the multi-scale feature information of the image through multi-layer convolution and pooling operations; then, the extracted feature map F is input into the UNet semantic segmentation model to generate a pixel classification probability map P=UNet(F) of the same size as the original image, where P represents the probability distribution of each pixel belonging to different resource types (such as forests, grasslands, waters, etc.); finally, the probability map P is thresholded to obtain the boundary map B of each resource category. q =Threshold(P q), where B q represents the boundary area of ​​resource category q; Threshold is a threshold segmentation function used to transform the probability map P q Transformed into clear category boundaries; P q Represents the probability graph associated with resource category q.

[0118] In step S402, a time series analysis model is used to compare and analyze multiple remote sensing images to achieve dynamic change monitoring of natural resources. t ,R t+1 ,...,R g , predict future trends through time series analysis R future , expressed as: R future =f(R t ,R t+1 ,...,R g ), where f() is a time series prediction function implemented by the LSTM prediction model, and g represents the last moment in the time series of the resource range; based on the prediction results, the natural resource boundaries and key monitoring point locations are identified.

[0119] In step S403, the positioning data L collected by the Beidou positioning terminal t represents the position coordinates at time t, and R t+1 The resource range at the subsequent moment, the change in natural resources ΔR is calculated by the following formula: ΔR=R t+1 -L t By calculating the changes of multiple positioning points on the boundary, the movement range and change trend of the resource boundary can be determined.

[0120] Furthermore, in step S4, the specific implementation of geological disaster risk assessment includes the following steps.

[0121] Step S404: using the regularly acquired remote sensing images to identify potential risk areas and conduct quantitative analysis of risk levels.

[0122] Step S405, using the Beidou positioning terminal to monitor the displacement and changes of the ground in the high-risk area in real time, and combining the monitoring data with the risk assessment results to achieve real-time update and early warning of the risk level.

[0123] In step S404, a convolutional neural network (CNN) is used to process the remote sensing image, extract the geological features in the remote sensing image, and combine the terrain data (such as slope, soil moisture, and historical rainfall) to form multidimensional input data, which specifically includes the following steps: for a given input remote sensing image I and terrain data G, extract feature representation through a convolutional neural network, the formula is: F=CNN(I,G), where F is the feature mapping of the remote sensing image and the geological data; classify the risk level of each area through a fully connected layer, and output the risk level R of each area, the formula is: R=Softmax(WF+b), where W and b are the weights and bias terms of the fully connected layer, and Softmax outputs the probability distribution of each type of risk; based on the obtained probability distribution, calculate the risk score S of each area, the formula is: S=∑ d=1 n p d r d , where p d is the probability of risk category d, r d is the level of risk category d, and n represents the number of risk categories.

[0124] In S405, multiple Beidou positioning terminals are deployed in high-risk areas for geological disasters to collect surface displacement data in real time and analyze surface movement trends in combination with historical data to identify potential disaster triggering factors. The positions of Beidou positioning terminal i at time t and time t+1 are L t i and L t+1 i , surface displacement vector ΔL i The formula is: ΔL i =L t+1 i -L t i ; Amplitude of surface displacement |ΔL i | is: |ΔL i ∣=[(Δx i ) 2 +(Δy i ) 2 +(Δz i ) 2 ] 1 / 2 , where (Δx i , Δy i , Δz i ) represents the displacement change of Beidou positioning terminal i in three spatial dimensions; the high-risk area is modeled as a graph structure G=(V,E) using graph neural network (GNN), where V represents the node set and E represents the edge set. For node v γ ∈V represents the Beidou positioning terminal, edge e γj ∈E represents node vγ and v j The interaction between them; Graph Neural Network (GNN) updates node features through information transmission, node v γ The feature at time t is denoted as h γ (t) , the update formula is: γ (t+1) =σ[W (t) ∑ j∈N(γ) h j (t) +b (t) ], where W (t) and b (t) is a learnable parameter, N(γ) is the neighbor node of node γ, σ is the activation function; h γ (t+1) is the updated feature representation, h j (t) Represents node v j The convolutional neural network (CNN) predicts the geological disaster risk of the node and evaluates the disaster risk level by learning the mutual influence between nodes, such as the displacement correlation between multiple terminals. The final node representation h γ (T) The disaster risk prediction P is obtained through the fully connected layer γ , the formula is: γ =Sigmoid(W final h γ (T) +b final ), where P γ is the risk probability of node γ, indicating the possibility of disasters occurring in the area; W final Represents the weight matrix of the final fully connected layer in the convolutional neural network, which is used to represent the final node h γ (T) Converted to disaster risk prediction γ ; b final Represents the bias term of the final fully connected layer in the convolutional neural network.

[0125] Furthermore, in step S4, the specific implementation of tracking and trajectory analysis includes the following steps.

[0126] Step S406, positioning the known positioning terminal, obtaining its real-time position information and dynamically updating it in parallel, including latitude and longitude, timestamp, speed and direction information.

[0127] Step S407, establish a historical trajectory database to support users to retrieve the historical movement trajectory of a specific terminal based on a time range, and provide a trajectory playback function to intuitively display the historical operation status of the trajectory.

[0128] Step S408, integrating geographic location, real-time status and emergency event information to provide real-time scheduling decision support and optimal path planning suggestions.

[0129] Furthermore, in step S4, the specific implementation of visualization and report generation includes the following steps.

[0130] Step S409, by rendering the natural resource distribution and terminal positioning data into vector tiles in real time, a multi-level, multi-dimensional dynamic display is performed in the user interface, supporting users to flexibly adjust the hierarchical structure and display method of data visualization according to actual needs.

[0131] Step S410, automatically integrates natural resource monitoring, geological disaster warning and dispatch information to generate a customizable report, and provides multiple export format options (such as PDF, Excel, etc.) to meet different data analysis needs.

[0132] Furthermore, in step S4, the specific implementation of the event triggering and alarm mechanism includes the following steps.

[0133] Step S411, using geographic fence technology to monitor the location changes of the positioning terminal in real time, and triggering an alarm notification when a selected or unknown positioning terminal leaves or enters the fence.

[0134] Step S412, combining geological disaster warning data, automatically generating warning information and pushing it to relevant users.

[0135] Step S413, introducing an abnormal behavior detection mechanism based on a rule engine, when the positioning terminal exhibits abnormal behavior, automatically triggering an event, starting a corresponding analysis process or sending a notification.

[0136] Further, in step S5, the functional extension is configured to be implemented.

[0137] Access new data sources, algorithm models or external systems through open API interfaces.

[0138] The development of custom modules allows users to integrate new functional modules into existing systems through APIs to achieve dynamic expansion of functions.

[0139] Importing and exporting modules allows users to export configured workflows and import them into other projects or systems, facilitating team collaboration and deployment.

[0140] Template saving allows users to save commonly used modules as reusable templates to improve subsequent deployment efficiency.

[0141] Sharing of module libraries allows downloading modules or workflow templates from other users, reducing duplication of work and improving collaboration efficiency.

[0142] Dynamic loading of modules: dynamically load the required functional modules according to the actual needs of users, or unload unnecessary modules according to the scenarios and performance requirements.

[0143] like Figure 2 As shown, a pan-Beidou terminal equipment management system includes the following modules.

[0144] The remote sensing data acquisition module is responsible for collecting satellite remote sensing data of the selected area, performing preprocessing, and providing the corresponding calling interface.

[0145] The positioning data acquisition module is responsible for collecting the location and movement information of the relevant positioning terminals, performing preprocessing, and providing the corresponding calling interface.

[0146] The natural resource identification module, based on processed remote sensing data, uses semantic segmentation technology to determine the boundaries of various natural resources in the selected area, and provides corresponding output data and calling interfaces.

[0147] The geological disaster risk assessment module uses processed remote sensing data to identify high-risk areas for geological disasters through convolutional neural networks, and provides corresponding output data and calling interfaces.

[0148] The natural resource analysis module accurately determines the location and area of ​​various natural resources by deploying positioning terminals at the original boundaries of various natural resources, and compares them with the newly acquired remote sensing semantic segmentation data.

[0149] The geological disaster determination module uses a group of positioning terminals deployed in high-risk areas to identify potential geological disaster locations through a graph neural network.

[0150] The terminal positioning module locates known positioning terminals in real time and dynamically updates their location information, including latitude and longitude, timestamp, speed and direction information.

[0151] The historical trajectory query module records the historical trajectory of the positioning terminal, queries the movement path within a specific time period by time range, and supports historical status playback.

[0152] The scheduling optimization module plans and schedules the paths of the devices carrying the positioning terminals based on the destination, the location of the positioning terminals and the road conditions.

[0153] The visualization module uses dynamic vector tile technology to display the input parameters of the connection module on the map of the user interface, supports the visualization and filtering of spatiotemporal data, and adjusts the view hierarchy and display method according to needs.

[0154] The report generation module generates customized reports based on the parameters selected by the user or the parameters output by the selected module, and exports them to PDF, Excel or other formats as required.

[0155] The geo-fence module is responsible for setting geo-fences for the selected area and triggering alarm notifications when the positioning terminal leaves or enters the fence.

[0156] The geological disaster early warning module automatically generates warning information and pushes it to relevant users based on the output data of the geological disaster risk assessment module and the geological disaster determination module.

[0157] The abnormal warning module for positioning terminals monitors abnormal behaviors of positioning terminals through a built-in rule engine and automatically starts related analysis processes or sends notifications.

[0158] The loosely coupled workflow building module provides a modular combination environment. Through module combination and configuration based on user permissions, personalized workflows are built, and an open API interface is provided to access new data sources, algorithm models or external systems.

[0159] Custom development modules allow users to integrate new functional modules into existing systems through APIs to achieve dynamic expansion of functions.

[0160] The permission management module is responsible for the permission control and export functions of the module, allowing users to export the configured workflow and reuse it in other projects or systems. It supports saving commonly used modules as templates and provides a permission-based shared module library for other users to download, reducing duplication of work.

[0161] Finally, it should be pointed out that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Although the present invention has been described in detail with reference to the above embodiments, a person skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for managing Pan-Beidou terminal equipment, characterized in that: The following steps are involved: Step S1, building a modular framework based on microservice architecture; Step S2, designing a loosely coupled module environment, developing module units, and combining and configuring modules through user permissions; Step S3, integrating remote sensing monitoring and Beidou positioning technology to collect and pre-process natural resource data in real time, including geological disaster monitoring data, positioning terminal location and status data; Step S4, providing the underlying functional implementation of natural resource analysis, geological disaster risk assessment, tracking and trajectory analysis, visualization and report generation, and event triggering and alarm mechanism; Step S5, supporting function extension through API interface, allowing users to access third-party services or develop new modules, and supporting import and export of workflows; In step S2, the implementation process of the loosely coupled module environment includes the following steps: Step S201, developing a standardized module library, implementing module function encapsulation and classification, and designing a permission control interface to support module calls based on user roles; Step S202, creating an independent parameter configuration interface for each module, supporting dynamic definition of input / output data flows, and providing a configuration tool to customize the execution logic of the module; Step S203, connecting different modules through authority verification, defining data flow and execution sequence, and forming a complete workflow; Step S204, introducing a permission-based intelligent matching mechanism to recommend available module combinations and workflow templates according to the user's permission level and operation requirements; The core of permission-based intelligent matching is to achieve accurate matching between user needs and functional modules, including the following key links: Abstract user needs into a set of demand keywords K={k1,k2,...,k μ }, where k1, k2, ..., k μ represents different keywords of user needs, μ represents the number of keywords of user needs; For each module M w Assign a feature label set T(M w )={t1,t2,...,t m }, where t1, t2,...,t m Represents module M w The feature label of m represents the module M w The number of feature labels; By calculating the relevance score Relevance(M w )=|K∩T(M w )∣ / ∣T(M w )|×P(Auth) to evaluate M w The matching degree to user needs, where P(Auth) represents the user's authority coefficient to the module; According to the task type T selected by the user task , from the preset standard module set M task ={M1, M2,...,M o } recommends common module combinations, where M1, M2,...,M o represents different modules; o represents the number of modules in the preset standard module set; Enrich the keyword set entered by the user through synonym expansion or keyword derivation; The specific implementation of natural resource analysis includes the following steps: step S401, identifying and classifying natural resources; step S402, evaluating the distribution characteristics, dynamic changes and potential risks of different types of resources, determining the boundaries and key monitoring point locations of various types of natural resources, and deploying Beidou positioning terminals; step S403, based on the data collected by the Beidou positioning terminals, continuously monitoring the dynamic changes of natural resources, and realizing accurate positioning of the resource range and real-time status update; The specific implementation of geological disaster risk assessment includes the following steps: Step S404, using regularly acquired remote sensing images to identify potential risk areas and quantify risk levels; Step S405, using Beidou positioning terminals to monitor the displacement and changes of the ground in high-risk areas in real time, and combining the monitoring data with risk assessment results to achieve real-time update and early warning of risk levels; The specific implementation of tracking and trajectory analysis includes the following steps: Step S406, locate the known positioning terminal, obtain its real-time location information and dynamically update it in parallel, including latitude and longitude, timestamp, speed and direction information; Step S407, establish a historical trajectory database to support users to retrieve the historical movement trajectory of a specific terminal based on the time range, and provide a trajectory playback function to intuitively display the historical operation status of the trajectory; Step S408, integrate the geographic location, real-time status and emergency event information, and provide real-time scheduling decision support and optimal path planning suggestions; The specific implementation of visualization and report generation includes the following steps: Step S409, by rendering the natural resource distribution and terminal positioning data into vector tiles in real time, a multi-level and multi-dimensional dynamic display is performed in the user interface, which supports users to flexibly adjust the hierarchical structure and display method of data visualization according to actual needs; Step S410, automatically integrating natural resource monitoring, geological disaster warning and dispatch information to generate a customizable report, and providing multiple export format options; The specific implementation of the event triggering and alarm mechanism includes the following steps: Step S411, real-time monitoring of the location changes of the positioning terminal through the geographic fence technology, triggering an alarm notification when the selected or unknown positioning terminal leaves or enters the fence; Step S412, automatically generating alarm information and pushing it to relevant users in combination with geological disaster warning data; Step S413, introducing an abnormal behavior detection mechanism based on a rule engine, automatically triggering an event when the positioning terminal exhibits abnormal behavior, starting the corresponding analysis process or sending a notification; In step S401, the remote sensing image is classified at the pixel level using the UNet semantic segmentation model, including the following steps: extracting features from the input remote sensing image I through a convolutional neural network to obtain a feature map F=CNN(I); inputting the extracted feature map F into the UNet semantic segmentation model to generate a pixel classification probability map P=UNet(F) of the same size as the original image, where P represents the probability distribution of each pixel belonging to a different resource type; performing threshold segmentation on the probability map to obtain a boundary map B of each resource category. q =Threshold(P q ), where B q represents the boundary area of ​​resource category q; Threshold is a threshold segmentation function used to transform the probability map P q Transformed into clear category boundaries; P q represents the probability graph associated with resource category q; In step S402, the dynamic changes of natural resources are evaluated by the following formula: future =f(R t ,R t+1 ,...,R g ), where R future is the changing trend of natural resources, R t ,R t+1 ,...,R g is the resource range in different periods; f() is the time series prediction function; g represents the last moment in the time series of the resource range; In step S403, the dynamic changes of natural resources are continuously monitored by the following formula: ΔR=R t+1 -L t , where ΔR is the change in natural resources; L t Represents the position coordinates of the Beidou positioning terminal at time t; R t+1 is the resource range at time t+1; Step S405 includes the following steps: The surface displacement vector ΔL is calculated by the following formula i :ΔL i =L t+1 i -L t i , where L t i and L t+1 i are the positions of Beidou positioning terminal i at time t and t+1 respectively; The amplitude of the ground displacement |ΔL is calculated by the following formula i |:|ΔL i ∣=[(Δx i ) 2 +(Δy i ) 2 +(Δz i ) 2 ] 1 / 2 , where (Δx i , Δy i , Δz i ) represents the displacement change of Beidou positioning terminal i in three spatial dimensions; The high-risk area is modeled as a graph structure G=(V,E), where V represents the node set, E represents the edge set, and node v γ ∈V represents the Beidou positioning terminal, edge e γj ∈E represents node v γ and v j the interactions between them; By passing information, node features are updated. γ The feature at time t is denoted as h γ (t) , the update formula is: γ (t+1) =σ[W (t) ∑ j∈N(γ) h j (t) +b (t) ], where W (t) and b (t) is a learnable parameter, N(γ) is the neighbor node of node γ, σ is the activation function; h γ (t+1) is the updated feature representation, h j (t) Represents node v j Feature representation at time t; By learning the mutual influence between nodes, the geological disaster risk of the node is predicted and the disaster risk level is evaluated. Finally, the node represents h γ (T) The disaster risk prediction P is obtained through the fully connected layer γ , expressed as: P γ =Sigmoid(W final h γ (T) +b final ), where P γ is the risk probability of node γ, indicating the possibility of disasters occurring in the area; W final Represents the weight matrix of the final fully connected layer in the convolutional neural network, which is used to represent the final node h γ (T) Converted to disaster risk prediction γ ; b final Represents the bias term of the final fully connected layer in the convolutional neural network.

2. A Pan-Beidou terminal equipment management method according to claim 1, characterized in that: In step S1, a modular framework based on microservice architecture is built through the following steps: Step S101, splitting the functions into multiple independent modules; Step S102, designing a standardized API interface to allow direct calling of existing functional modules or support the development and integration of new modules; Step S103, configure the database and design an efficient data management system to support data persistence, fast query, index creation and data flow interaction of different modules; Step S104, introducing a role authority management mechanism to strictly control the user's access rights to functional modules and data; Step S105, deploy edge computing nodes to process large-scale real-time data streams and implement data preprocessing and localized storage.

3. A Pan-Beidou terminal equipment management method according to claim 1, characterized in that: The permission-based intelligent matching mechanism provides personalized module suggestions for workflow construction by analyzing user behavior and preferences, including the following steps: Guide users to enter or select specific requirements and provide initial recommendations based on requirements and user permissions; According to the permission level, the modules that are not available are adjusted or restricted, and the intelligent matching mechanism is continuously optimized based on the user’s feedback. The mathematical model is as follows: Let the user’s permission to access the module be Auth(Mw), and the probability of accepting the recommended module be P(Accept(Mw)). w )), adjust the module's recommendation weight W(M) according to the acceptance or rejection operation w ), the adjustment formula is: W(M w ) new =W(M w ) old ×P(Accept(M w ))×Auth(Mw).

4. A Pan-Beidou terminal equipment management method according to claim 1, characterized in that: In step S3, the implementation of real-time collection and preprocessing of natural resource data includes the following steps: Step S301, regularly acquiring high-resolution remote sensing images covering preset natural resource monitoring areas and geological disaster risk areas according to the adaptability of the time window; Step S302, preprocessing the acquired remote sensing image, including image correction, denoising and enhancement; Step S303, deploying Beidou positioning terminals at the boundaries and key monitoring points of the natural resource monitoring area to collect positioning data in real time; Step S304, deploy Beidou positioning terminals in high-risk areas for geological disasters to monitor ground displacement and changes in real time, wherein the data of the Beidou positioning terminals are transmitted according to the JT808 protocol.

5. A Pan-Beidou terminal equipment management method according to claim 1, characterized in that: In step S5, the functional extension is configured to implement: Access new data sources, algorithm models or external systems through open API interfaces; Development of custom modules, allowing users to integrate new functional modules into existing systems through APIs; Import and export of modules, allowing users to export configured workflows and import them into other projects or systems; Template saving allows users to save commonly used modules as reusable templates; Sharing of module libraries, allowing downloading of other users’ modules or workflow templates; Dynamic loading of modules, dynamically loading functional modules according to actual needs of users.

6. A Pan-Beidou terminal equipment management system, used to implement a Pan-Beidou terminal equipment management method according to any one of claims 1 to 5, characterized in that: Includes the following modules: The remote sensing data acquisition module is responsible for collecting satellite remote sensing data of the selected area, preprocessing it, and providing the corresponding calling interface; The positioning data collection module is responsible for collecting the location and movement information of the relevant positioning terminals, performing preprocessing, and providing the corresponding calling interface; The natural resource identification module uses semantic segmentation technology to determine the boundaries of various natural resources in the selected area based on processed remote sensing data, and provides corresponding output data and calling interfaces; The geological disaster risk assessment module uses processed remote sensing data to identify high-risk areas for geological disasters through convolutional neural networks, and provides corresponding output data and call interfaces; The natural resource analysis module accurately determines the location and area of ​​various natural resources by deploying positioning terminals at the original boundaries of various natural resources, and compares them with the newly acquired remote sensing semantic segmentation data; The geological disaster determination module uses a group of positioning terminals deployed in high-risk areas to identify potential geological disaster locations through a graph neural network; Terminal positioning module, which locates known positioning terminals in real time and dynamically updates their location information, including latitude and longitude, timestamp, speed and direction information; The historical trajectory query module records the historical trajectory of the positioning terminal, queries the movement path within a specific time period by time range, and supports historical status playback; The scheduling optimization module plans and schedules the paths of the devices carrying the positioning terminals based on the destination, the location of the positioning terminals and the road conditions; The visualization module uses dynamic vector tile technology to display the input parameters of the connection module on the map of the user interface, supports the visualization and filtering of spatiotemporal data, and adjusts the view hierarchy and display method according to needs; Report generation module, which generates customized reports based on the parameters selected by the user or the parameters output by the selected module, and exports them to PDF, Excel or other formats as required; The geo-fence module is responsible for setting geo-fences for the selected area and triggering alarm notifications when the positioning terminal leaves or enters the fence; The geological disaster early warning module automatically generates warning information and pushes it to relevant users based on the output data of the geological disaster risk assessment module and the geological disaster determination module; The abnormal warning module of the positioning terminal monitors the abnormal behavior of the positioning terminal through the built-in rule engine, and automatically starts the relevant analysis process or sends a notification; Loosely coupled workflow building module provides a modular combination environment, builds personalized workflows through module combination and configuration based on user permissions, and provides an open API interface; Custom development modules allow users to integrate new functional modules into existing systems through APIs; The permission management module is responsible for the permission control and export functions of the module, allowing users to export the configured workflow and reuse it in other projects or systems. It supports saving commonly used modules as templates and provides a permission-based shared module library for other users to download.

Images