Method and system for automatically generating remote control instruction in satellite interface data list

Abstract

The invention provides a method and a system for automatically generating a remote control instruction for a satellite interface data list. The method comprises the following steps: S1, filling various interconnection information in the satellite interface data list; s2, reading the satellite interface data list under the catalog to obtain product element information; s3, key features are extracted from the product element information; s4, compiling a remote control instruction template and verifying the remote control instruction template, if the remote control instruction template passes verification, storing the remote control instruction template in a training set of the machine learning model and continuing to execute the step S5, otherwise, returning to the step S1; and S5, automatically generating a remote control instruction list. According to the invention, a satellite interface data sheet does not need to be manually extracted and recorded, and detailed information including signal content, subsystem name, equipment name, equipment code, electric connector code, electric connector model, destination, contact number, voltage, current, polarity, signal type, line type, remark description and electric connector code can be automatically identified and read by software.

Description

Technical Field

[0001] The present invention relates to the field of satellite communication technology, and in particular to a method and system for automatically generating remote control commands from satellite interface data sheets, and in particular to a method and system for automatically generating remote control commands from satellite interface data sheets based on machine learning. Background Art

[0002] With the advancement of satellite communication technology, communications between satellites and ground stations have become increasingly complex. Traditional methods for generating remote control commands rely on manually parsing the satellite interface data format and manually writing the corresponding remote control commands. This method is not only time-consuming and labor-intensive, but also prone to errors. Furthermore, faced with diverse satellite models and changing mission requirements, traditional methods struggle to quickly adapt to new situations, limiting the flexibility and efficiency of satellite missions.

[0003] Patent document CN112995191A discloses a universal satellite remote control command generation method and system based on class derivation. The method includes steps for generating a remote control package and a frame function abstract class. This method can progressively derive satellite remote control commands, enabling universal processing and generation of satellite remote control commands. However, this patent document likely focuses more on the framework and methodology for command generation, with less attention paid to the specific generation and verification of command content.

[0004] Patent document CN119227690A discloses a knowledge graph-based method and management system for constructing remote control commands for low-orbit satellite clusters. This system designs a knowledge ontology-based command metadata model and implements remote control command generation and management for low-orbit satellites by constructing a multi-level command generation mechanism ("concept layer-capability layer-command layer"). However, compared to this application, it focuses more on cluster management and the systematic construction of commands.

[0005] Patent document CN105094829B discloses a method for automatically generating satellite measurement and control information flow diagrams. This method solves the time-consuming and error-prone nature of traditional manual drawing of satellite measurement and control information flow diagrams, enabling timely or synchronous updates of the information flow diagram and IDS data sources. This technology can indirectly improve the efficiency and accuracy of remote control command generation.

[0006] Patent document CN116366121A discloses a method and apparatus for generating satellite remote control command codes. The method involves obtaining the factory command set for a satellite subsystem and then organizing, modeling, storing, configuring, and generating it. This allows for batch assembly of satellite remote control commands. However, the method focuses more on batch command generation and provides less attention to intelligent and personalized command generation.

[0007] Patent document CN115202846B discloses a method and apparatus for determining remote control commands for an in-orbit satellite. The method involves determining a remote control command template for execution by a satellite, sorting satellite position data, and appending it to a payload command template to generate a remote control command set. However, the method focuses on command generation in specific scenarios and has limited applicability to other scenarios. Summary of the Invention

[0008] In view of the defects in the prior art, the purpose of the present invention is to provide a method and system for automatically generating remote control commands from satellite interface data sheets.

[0009] According to the present invention, a method for automatically generating remote control instructions from a satellite interface data sheet includes:

[0010] Step S1: Fill in various interconnection information in the satellite interface data sheet according to preset standards;

[0011] Step S2: Read the satellite interface data sheet in the directory and obtain product element information therefrom;

[0012] Step S3: extract key features from product element information;

[0013] Step S4: Compile a remote control command template according to the operation task of the satellite interface data sheet and verify it. If the verification passes, store it in the training set of the machine learning model and continue to step S5. Otherwise, return to step S1;

[0014] Step S5: Automatically generate a remote control command table based on the remote control command template and the trained machine learning model.

[0015] Preferably, the satellite system contact data verification system software is used to read the satellite interface data sheet under the directory to obtain product element information therefrom.

[0016] Preferably, step S3 includes dimensional sorting of product element information, applying word segmentation technology and named entity recognition technology to build a machine learning model, and automatically extracting key features from the product element information.

[0017] Preferably, the step S3 further includes using word segmentation technology to segment the text data, using named entity recognition to locate and classify important elements, and formulating additional post-processing rules to correct and supplement the results of named entity recognition.

[0018] Preferably, step S4 includes the following sub-steps:

[0019] Step S4.1: Parameter configuration of key feature content and generation of remote control instructions;

[0020] Step S4.2: Compile a remote control command template according to the operational tasks in the satellite interface data sheet. Compile remote control commands related to key features at the head or tail of the remote control command template according to the operational regulations of each type of interconnected information. Embed the relevant remote control command general model in the platform according to the specific tasks of the interconnected information. Combine the two according to the operational sequence to form the remote control command template for this type of product element information.

[0021] Step S4.3: When the remote control instruction template passes the verification, it is stored in the machine learning model training set.

[0022] Preferably, the step S4.2 includes:

[0023] Based on the trained machine learning model, real-time satellite interface data is input and the corresponding remote control command sequence is output. The monitoring system is deployed to continuously receive the latest telemetry data from the satellite to ensure that the information input into the machine learning model is up to date; whenever a new telemetry update is received, it is immediately applied to the machine learning model.

[0024] Preferably, the step of training the machine learning model includes:

[0025] Use weak classifiers to generate results;

[0026] Multiple results are integrated into a consensus function to obtain the final result using a voting scheme.

[0027] Preferably, step S5 includes the following sub-steps:

[0028] Step S5.1: Based on the trained machine learning model, real-time satellite interface data is input, the remote control command template is automatically filled in, and the corresponding remote control command sequence is output;

[0029] Step S5.2: Automatically generate a remote control command table according to the remote control command sequence.

[0030] Preferably, a user interface is provided for template selection, data sheet upload, instruction table generation preview and export functions; the user sets the input and output directories of the interface data sheet file through the user interface, and views the program reading in real time, integrates product element information, and the progress and processing structure of the production remote control instruction table.

[0031] According to the present invention, a system for automatically generating remote control instructions from a satellite interface data sheet includes:

[0032] Module M1: Fill in various interconnection information in the satellite interface data sheet according to the preset standards;

[0033] Module M2: reads the satellite interface data sheet in the directory and obtains product element information from it;

[0034] Module M3: Extract key features from product element information;

[0035] Module M4: Compiles a remote control command template according to the operation task of the satellite interface data sheet and verifies it. If the verification passes, it is stored in the training set of the machine learning model and continues to execute module M5. Otherwise, it returns to module M1.

[0036] Module M5: Automatically generate remote control command tables based on remote control command templates and trained machine learning models.

[0037] Preferably, the satellite system contact data verification system software is used to read the satellite interface data sheet under the directory to obtain product element information therefrom.

[0038] Preferably, the module M3 includes dimensional sorting of product element information, applying word segmentation technology and named entity recognition technology to build a machine learning model, and automatically extracting key features from the product element information.

[0039] Preferably, the module M3 also includes using word segmentation technology to segment text data, using named entity recognition to locate and classify important elements, and formulating additional post-processing rules to correct and supplement the results of named entity recognition.

[0040] Preferably, the module M4 includes the following submodules:

[0041] Module M4.1: configure parameters for key features and generate remote control commands;

[0042] Module M4.2: Compile remote control command templates according to the operational tasks in the satellite interface data sheet. Compile remote control commands related to key features at the head or tail of the remote control command template according to the operational regulations of each type of interconnected information. Embed the relevant remote control command general model in the platform according to the specific tasks of the interconnected information. Combine the two according to the operational sequence to form the remote control command template for this type of product element information.

[0043] Module M4.3: When the remote control command template passes the verification, it is stored in the machine learning model training set.

[0044] Preferably, the module M4.2 includes:

[0045] Based on the trained machine learning model, real-time satellite interface data is input and the corresponding remote control command sequence is output. The monitoring system is deployed to continuously receive the latest telemetry data from the satellite to ensure that the information input into the machine learning model is up to date; whenever a new telemetry update is received, it is immediately applied to the machine learning model.

[0046] Preferably, the modules of the machine learning model training set include:

[0047] Use weak classifiers to generate results;

[0048] Multiple results are integrated into a consensus function to obtain the final result using a voting scheme.

[0049] Preferably, the module M5 includes the following submodules:

[0050] Module M5.1: Based on the trained machine learning model, it inputs real-time satellite interface data, automatically fills in the remote control command template, and outputs the corresponding remote control command sequence;

[0051] Module M5.2: Automatically generate remote control command table according to remote control command sequence.

[0052] Preferably, a user interface is provided for template selection, data sheet upload, instruction table generation preview and export functions; the user sets the input and output directories of the interface data sheet file through the user interface, and views the program reading in real time, integrates product element information, and the progress and processing structure of the production remote control instruction table.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] 1. The present invention eliminates the need for manual extraction and entry of satellite interface data sheets, and can enable the software to automatically identify and read detailed information including signal content, subsystem name, device name, device code, electrical connector code, electrical connector model, destination, contact number, voltage, current, polarity, signal type, line type, remarks, and electrical connector code.

[0055] 2. The present invention constructs a machine learning model by applying word segmentation technology and named entity recognition technology (NER), which can automatically extract key features from product element information.

[0056] 3. The present invention realizes automatic software screening of remote control command related information and classification of remote control commands through keyword matching and a training model based on sample data, thereby improving the recognition accuracy of remote control commands and necessary parameters.

[0057] 4. The present invention provides a user-friendly software operation interface, supports customized output formats, and enables the software to automatically generate remote control command tables, effectively improving the efficiency, reliability and compatibility of the remote control command tables.

[0058] Other beneficial effects of the present invention will be explained through the introduction of specific technical features and technical solutions in the specific implementation methods. Those skilled in the art should be able to understand the beneficial technical effects brought about by the introduction of these technical features and technical solutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Other features, objects and advantages of the present invention will become more apparent upon reading the detailed description of non-limiting embodiments with reference to the following drawings:

[0060] Figure 1 This is a block diagram of the method for automatically generating remote control instructions from satellite interface data based on machine learning in the present invention.

[0061] Figure 2 This is a diagram of the machine learning model architecture in an embodiment of the present invention. DETAILED DESCRIPTION

[0062] The present invention will be described in detail below with reference to specific embodiments. The following examples will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those skilled in the art, several changes and improvements can be made without departing from the scope of the present invention. These all fall within the scope of protection of the present invention.

[0063] Reference Figure 1 As shown, a method for automatically generating remote control commands from a satellite interface data sheet includes:

[0064] Step S1: Fill in various interconnection information in the interface data sheet as required according to the satellite interface data sheet specification document;

[0065] Step S2: Using the satellite system contact data verification system software to read the satellite interface data sheet in the directory, and obtain product element information therefrom;

[0066] Step S3: Dimensionally sort the product element information, apply word segmentation technology and named entity recognition technology to build a machine learning model, and automatically extract key features from the product element information;

[0067] Step S4: Parameter configuration of key feature content to generate remote control instructions;

[0068] Step S5: Compile a command template based on the operational tasks in the satellite interface data sheet. Compile remote control commands related to key features at the head or tail of the remote control command template according to the operational specifications of each type of interconnected information. Embed the relevant remote control command general model in the platform based on the specific tasks of the interconnected information. Finally, combine the two according to the operational sequence to form the remote control command template for this type of product element information.

[0069] Step S6: When the remote control command template passes the verification, it is stored in the machine learning model training set;

[0070] Step S7: Based on the trained model, real-time satellite interface data is input, the remote control command template is automatically filled in, and the corresponding remote control command sequence is output.

[0071] Step S8: Automatically generate a remote control command table according to the remote control command sequence.

[0072] After filling in the satellite interface data sheet in step S1, a process of verifying the integrity and format of the data sheet is further included.

[0073] The satellite interface data sheet read in step S2 includes product element information such as signal content, subsystem name, device name, device code, electrical connector code, electrical connector model, destination, contact number, voltage, current, polarity, signal type, line type, remarks, electrical connector code, etc.

[0074] In step S3, for textual data, word segmentation techniques are used to segment the data, and named entity recognition (NER) is used to locate and classify important elements. For patterns that are difficult to capture directly through machine learning, additional post-processing rules can be developed to correct and supplement the NER results.

[0075] In step S5, the trained model is fed with real-time satellite interface data and outputs a corresponding sequence of remote control commands. The deployed monitoring system continuously receives the latest telemetry data from the satellites, ensuring that the information input to the model is up to date. Whenever a new telemetry update is received, it is immediately applied to the model.

[0076] In step S6, the machine learning model training set includes two steps:

[0077] Step S61 uses weak classifier to generate results

[0078] Step S62 integrates the multiple results into a consistency function to obtain the final result using a voting scheme.

[0079] Specifically: Suppose there are m base classifiers C1, C2, ..., C m , the prediction results of each classifier are y1,y2,…,y m ,∈{s tr0 ,s tr1 ,s tr2 ,...,s trn}, where stri is the first string generated by the i-th model, i = 0, 1, 2...., and each string is regarded as a separate category, thereby converting the text generation voting problem into a multi-classification voting problem.

[0080] The remote control command table in step S8 provides multiple output options, including PDF, HTML pages and machine-readable binary files, for subsequent remote control system planning and verification.

[0081] The automated process includes a user interface that provides functions such as template selection, data sheet upload, generated instruction table preview and export.

[0082] Through the software's visual interface, users can set the input and output directories of interface data files and view the progress and processing structure of program reading, integration of product element information, and production remote control instruction tables in real time.

[0083] The above are basic embodiments of the present invention. The technical solution of the present invention will be further described below through one or several preferred embodiments.

[0084] Example 1

[0085] Reference Figure 1 As shown, the method for automatically generating remote control commands from satellite interface data based on machine learning in this embodiment includes the following steps:

[0086] 1. According to the satellite interface data sheet specification document, fill in various interconnection information in the interface data sheet as required, further including the verification process of the data sheet integrity and format.

[0087] 2. Use the satellite system contact data verification system software to read the satellite interface data sheet in the directory, and read the satellite interface data sheet including signal content, subsystem name, equipment name, equipment code, electrical connector code, electrical connector model, destination, contact number, voltage, current, polarity, signal type, line type, remarks, electrical connector code and other product element information.

[0088] 3. For text data, use word segmentation technology to segment the data; use named entity recognition (NER) to locate and classify important elements. For some patterns that are difficult to capture directly through machine learning, additional post-processing rules can be developed to correct or supplement the NER results. By applying word segmentation technology and named entity recognition technology (NRE) to build a machine learning model, key features can be automatically extracted from product element information.

[0089] In this embodiment, word segmentation technology is used to cut text into usable data units.

[0090] Choose a suitable word segmentation tool: Choose a suitable word segmentation library or tool based on the language characteristics. For example, NLTK can be used for English; Jieba word segmenter can be used for Chinese.

[0091] The field of satellite communications has its own unique terminology and technical vocabulary, so special attention should be paid to preserving the integrity of this specialized vocabulary during word segmentation. Building a custom dictionary can enhance word segmentation and ensure that proper nouns are not mistakenly split into meaningless parts.

[0092] Satellite data often contains a variety of abbreviations, codes, or special symbols. Rules should be defined in advance to correctly parse these elements and avoid confusion during word segmentation.

[0093] In an embodiment of the present invention, named entity recognition is used to locate and classify important elements.

[0094] A customized NER model is trained using an annotated satellite communications corpus. This can be achieved through transfer learning, where an existing general model is fine-tuned to suit the needs of a specific domain.

[0095] Determine which types of entities are important to the task and give each type clear labels. For example, "PARAM_NAME" represents the parameter name, "TIMESTAMP" represents the timestamp, and "COMMAND_CODE" is used to identify the command code.

[0096] NER not only relies on individual words but also considers the surrounding context. By understanding sentence structure and logical relationships, recognition accuracy can be improved.

[0097] In the embodiment of the present invention, post-processing rules are used to formulate additional post-processing rules to correct or supplement the NER results for some patterns that are difficult to capture directly through machine learning.

[0098] For example, the identified timestamp is formatted according to a known format specification.

[0099] Consider a text description from a satellite interface data sheet: "The power supply voltage isset to 12V at 2024-03-01T15:30:00Z."

[0100] The word segmentation result may be the following fragment: ['The','power','supply','voltage','is','set','to','12V','at','2024-03-01T15:30:00Z'].

[0101] After NER processing:

[0102] Parameter name (PARAM_NAME): "power supply voltage"

[0103] Value: "12V"

[0104] Timestamp (TIMESTAMP): "2024-03-01T15:30:00Z"

[0105] 4. Combine a large number of annotated satellite operation cases as a training set for the machine learning model to train a model that can understand the satellite status and predict the required remote control commands;

[0106] Furthermore, if Figure 2 As shown in Figure 2, the machine learning model training set includes the following two steps:

[0107] The first step is to use weak classifier to generate results

[0108] The second step integrates multiple results into a consensus function to obtain the final result using a voting scheme.

[0109] Specifically: Suppose there are m base classifiers C1, C2, ..., C m , the prediction results of each classifier are y1,y2,…,y m ,∈{s tr0 ,s tr1 ,s tr2 ,...,s trn}, where stri is the first string generated by the i-th model, i = 0, 1, 2...., and each string is regarded as a separate category, thereby converting the text generation voting problem into a multi-classification voting problem.

[0110] The system inputs real-time satellite interface data and outputs the corresponding remote control command sequence. The deployment monitoring system continuously receives the latest telemetry data from the satellite, ensuring that the information input into the model is up to date. Whenever a new telemetry update is received, it is immediately applied to the model.

[0111] 5. Based on the trained model, input real-time satellite interface data, output the corresponding remote control command sequence, and deploy a monitoring system to continuously receive the latest telemetry data from the satellite to ensure the information input to the model.

[0112] 6. The remote control command sheet provides multiple output options, including PDF, HTML pages, and machine-readable binary files, based on the remote control command sequence. This output is used for subsequent remote control system planning and verification. The automated process includes a user interface that provides template selection, data sheet upload, generated command sheet preview, and export functionality. Through the software's visual interface, users can set the input and output directories for the interface data sheet file and view the program's progress and processing structure as it reads and integrates product element information, including the production of the remote control command sheet.

[0113] The present invention also provides a system for automatically generating remote control commands from a satellite interface data sheet. The system for automatically generating remote control commands from a satellite interface data sheet can be implemented by executing the process steps of the method for automatically generating remote control commands from a satellite interface data sheet. That is, those skilled in the art can understand the method for automatically generating remote control commands from a satellite interface data sheet as a preferred implementation of the system for automatically generating remote control commands from a satellite interface data sheet.

[0114] Specifically, a system for automatically generating remote control commands from satellite interface data includes:

[0115] Module M1: Fill in various interconnection information in the satellite interface data sheet according to the preset standards;

[0116] Module M2: reads the satellite interface data sheet in the directory and obtains product element information from it;

[0117] Module M3: Extract key features from product element information;

[0118] Module M4: Compiles a remote control command template according to the operation task of the satellite interface data sheet and verifies it. If the verification passes, it is stored in the training set of the machine learning model and continues to execute module M5. Otherwise, it returns to module M1.

[0119] Module M5: Automatically generate remote control command tables based on remote control command templates and trained machine learning models.

[0120] Use the satellite system contact data verification system software to read the satellite interface data sheet in the directory and obtain product element information from it.

[0121] The module M3 includes dimensional sorting of product element information, applying word segmentation technology and named entity recognition technology to build a machine learning model, and automatically extracting key features from the product element information.

[0122] The module M3 also includes using word segmentation technology to segment text data, using named entity recognition to locate and classify important elements, and formulating additional post-processing rules to correct and supplement the results of named entity recognition.

[0123] The module M4 includes the following submodules:

[0124] Module M4.1: configure parameters for key features and generate remote control commands;

[0125] Module M4.2: Compile remote control command templates according to the operational tasks in the satellite interface data sheet. Compile remote control commands related to key features at the head or tail of the remote control command template according to the operational regulations of each type of interconnected information. Embed the relevant remote control command general model in the platform according to the specific tasks of the interconnected information. Combine the two according to the operational sequence to form the remote control command template for this type of product element information.

[0126] Module M4.3: When the remote control command template passes the verification, it is stored in the machine learning model training set.

[0127] The module M4.2 includes:

[0128] Based on the trained machine learning model, real-time satellite interface data is input and the corresponding remote control command sequence is output. The monitoring system is deployed to continuously receive the latest telemetry data from the satellite to ensure that the information input into the machine learning model is up to date; whenever a new telemetry update is received, it is immediately applied to the machine learning model.

[0129] The modules of the machine learning model training set include:

[0130] Use weak classifiers to generate results;

[0131] Multiple results are integrated into a consensus function to obtain the final result using a voting scheme.

[0132] The module M5 includes the following submodules:

[0133] Module M5.1: Based on the trained machine learning model, it inputs real-time satellite interface data, automatically fills in the remote control command template, and outputs the corresponding remote control command sequence;

[0134] Module M5.2: Automatically generate remote control command table according to remote control command sequence.

[0135] A user interface is provided for template selection, data sheet upload, command sheet generation preview and export functions; users can set the input and output directories of interface data sheet files through the user interface, and view the program reading, integration of product element information and the progress and processing structure of the production remote control command sheet in real time.

[0136] Those skilled in the art will appreciate that, in addition to implementing the system and its various devices, modules, and units provided by the present invention in purely computer-readable program code, it is entirely possible to implement the same functions of the system and its various devices, modules, and units provided by the present invention in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered a hardware component, and the devices, modules, and units included therein for implementing various functions can also be considered as structures within the hardware component; the devices, modules, and units for implementing various functions can also be considered as both software modules implementing the method and structures within the hardware component.

[0137] The above describes specific embodiments of the present invention. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art may make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. The embodiments of this application and the features in the embodiments may be combined with each other in any manner unless there is a conflict.

Claims

1. A method for automatically generating remote control commands from satellite interface data sheets, characterized in that: include: Step S1: Fill in various interconnection information in the satellite interface data sheet according to preset standards; Step S2: Read the satellite interface data sheet in the directory and obtain product element information therefrom; Step S3: extract key features from product element information; Step S4: Compile a remote control command template according to the operation task of the satellite interface data sheet and verify it. If the verification passes, store it in the training set of the machine learning model and continue to step S5. Otherwise, return to step S1; Step S5: Automatically generate a remote control command table based on the remote control command template and the trained machine learning model.

2. The method for automatically generating remote control commands from satellite interface data sheets according to claim 1, characterized in that: Use the satellite system contact data verification system software to read the satellite interface data sheet in the directory and obtain product element information from it.

3. The method for automatically generating remote control commands from satellite interface data sheets according to claim 1, characterized in that: The step S3 includes dimensional sorting of product element information, applying word segmentation technology and named entity recognition technology to build a machine learning model, and automatically extracting key features from the product element information.

4. The method for automatically generating remote control commands from satellite interface data sheets according to claim 3, characterized in that: The step S3 also includes using word segmentation technology to segment text data, using named entity recognition to locate and classify important elements, and formulating additional post-processing rules to correct and supplement the results of named entity recognition.

5. The method for automatically generating remote control commands from satellite interface data sheets according to claim 3, characterized in that: The step S4 includes the following sub-steps: Step S4.1: Parameter configuration of key feature content and generation of remote control instructions; Step S4.2: Compile a remote control command template according to the operational tasks in the satellite interface data sheet. Compile remote control commands related to key features at the head or tail of the remote control command template according to the operational regulations of each type of interconnected information. Embed the relevant remote control command general model in the platform according to the specific tasks of the interconnected information. Combine the two according to the operational sequence to form the remote control command template for this type of product element information. Step S4.3: When the remote control instruction template passes the verification, it is stored in the machine learning model training set.

6. The method for automatically generating remote control commands from satellite interface data sheets according to claim 5, characterized in that: The step S4.2 includes: Based on the trained machine learning model, real-time satellite interface data is input and the corresponding remote control command sequence is output. The monitoring system is deployed to continuously receive the latest telemetry data from the satellite to ensure that the information input into the machine learning model is up to date; whenever a new telemetry update is received, it is immediately applied to the machine learning model.

7. The method for automatically generating remote control commands from satellite interface data sheets according to claim 5, characterized in that: The steps of training the machine learning model include: Use weak classifiers to generate results; Multiple results are integrated into a consensus function to obtain the final result using a voting scheme.

8. The method for automatically generating remote control commands from satellite interface data sheets according to claim 5, characterized in that: The step S5 includes the following sub-steps: Step S5.1: Based on the trained machine learning model, real-time satellite interface data is input, the remote control command template is automatically filled in, and the corresponding remote control command sequence is output; Step S5.2: Automatically generate a remote control command table according to the remote control command sequence.

9. The method for automatically generating remote control commands from satellite interface data sheets according to claim 1, characterized in that: A user interface is provided for template selection, data sheet upload, command sheet generation preview and export functions; users can set the input and output directories of interface data sheet files through the user interface, and view the program reading, integration of product element information and the progress and processing structure of the production remote control command sheet in real time.

10. A system for automatically generating remote control commands from satellite interface data sheets, characterized in that: include: Module M1: Fill in various interconnection information in the satellite interface data sheet according to the preset standards; Module M2: reads the satellite interface data sheet in the directory and obtains product element information from it; Module M3: Extract key features from product element information; Module M4: Compiles a remote control command template according to the operation task of the satellite interface data sheet and verifies it. If the verification passes, it is stored in the training set of the machine learning model and continues to execute module M5. Otherwise, it returns to module M1. Module M5: Automatically generate remote control command tables based on remote control command templates and trained machine learning models.

Images