Satellite remote control and telemetering AI safety monitoring method and device

Through AI security monitoring methods, satellite remote control and telemetry information are preprocessed and neural network monitoring, which solves the problem of inefficiency of traditional methods, realizes real-time security monitoring and rapid response, and enhances the satellite's security management and attack pattern recognition capabilities.

CN118900145APending Publication Date: 2024-11-05SHANGHAI AEROSPACE COMP TECH INST
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410935337.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-12
Publication Date
2024-11-05

AI Technical Summary

Technical Problem

Traditional satellite security monitoring methods are inefficient, prone to missing important security information, and are difficult to deal with complex security threats and malicious attacks.

Method used

AI security monitoring method is used to obtain remote control and telemetry information from the bus for pre-processing, and use the trained neural network model for monitoring and processing, to determine whether there are abnormalities in the information, and use the central processor for processing and response.

Benefits of technology

Real-time monitoring of satellite operation status, automatically discover potential safety hazards, and respond quickly, enhancing the satellite's security management capabilities, accurately identifying complex attack modes, and improving the guarantee of satellite safe operation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118900145A_ABST
    Figure CN118900145A_ABST
Patent Text Reader

Abstract

The invention provides an AI safety monitoring method and device for satellite remote control and telemetering, and the method comprises the steps: obtaining original information from a bus, carrying out the preprocessing of the original information, and obtaining the preprocessed information, and the preprocessing comprises the steps: carrying out the partial numerical change and vacancy supplement of data in a binary form; the preprocessed information is input into a trained neural network model, AI monitoring processing is conducted through the trained neural network model, a judgment result is obtained, and the judgment result is used for judging whether the original information is abnormal or not; and sending a judgment result to a central processing unit (CPU) for processing. Therefore, the running state of the satellite can be monitored in real time, whether current remote control is abnormal or not can be judged through the previous remote control and telemetering time sequence of the satellite, potential safety hazards are automatically found, quick response is made, the satellite AI safety monitoring method based on the time sequence is achieved, and safety management of the satellite is enhanced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of satellite safety monitoring, and in particular to an AI (Artificial Intelligence) safety monitoring method and device for satellite remote control and telemetry. Background Art

[0002] With the rapid development of science and technology, satellites are playing an increasingly important role in the fields of communication, navigation, meteorological observation, etc. At the same time, satellite systems are also facing increasing security risks. Since satellites are in a complex space environment, they are easily interfered by various factors, which affects their normal operation.

[0003] At present, traditional satellite security monitoring methods are unable to cope with increasingly complex security threats. Traditional methods usually rely on manual monitoring and analysis, which is not only inefficient but also prone to missing important security information. In addition, hackers and malicious attackers attack satellite systems, making the satellite security situation even more severe.

[0004] Therefore, it is urgent to design an AI security monitoring method for satellite remote control and telemetry to improve the security of satellite systems. Summary of the invention

[0005] In view of the defects in the prior art, the purpose of the present invention is to provide an AI safety monitoring method and device for satellite remote control and telemetry.

[0006] In a first aspect, an embodiment of the present application provides an AI safety monitoring method for satellite remote control and telemetry, including:

[0007] Acquire original information from the bus, wherein the original information includes: remote control information and telemetry information;

[0008] Preprocessing the original information to obtain preprocessed information, wherein the preprocessing method includes: partially changing the values ​​of binary data and supplementing missing bits;

[0009] Inputting the preprocessed information into a trained neural network model, and performing AI monitoring processing by the trained neural network model to obtain a judgment result, wherein the judgment result is used to determine whether the original information has an abnormality;

[0010] The judgment result is sent to a CPU (Central Processing Unit / Processor) for processing.

[0011] Optionally, after acquiring the original information from the bus, the method further includes:

[0012] All-0 information is inserted between each remote control message and telemetry message to indicate the time interval.

[0013] Optionally, preprocessing the original information to obtain preprocessed information includes:

[0014] The original information is represented in binary form, and the value 1 in the binary data is kept unchanged, and 0 in the binary data is changed to -1, so as to obtain the changed data;

[0015] The changed data is supplemented with missing bits so that the data is expanded from M bits to N bits, wherein the missing bits are supplemented with 0, M and N are positive integers, and M <N。

[0016] Optionally, before inputting the preprocessed information into a pre-trained neural network model, the method further comprises:

[0017] Construct a neural network model based on LSTM (Long Short-Term Memory); wherein the neural network model includes: a K-layer recurrent neural network and an L-layer fully connected neural network, K and L are positive integers; the neural network model includes: an input gate, a forget gate, an output gate, and a t The hidden state at time t is recorded, where the hidden state contains the LSTM's summary of all input information before time t.

[0018] Optionally, the method further comprises:

[0019] Download the original information obtained by the satellite from the bus to the ground terminal through the space-ground link;

[0020] The neural network model is trained by a GPU (Graphics Processing Unit) on the ground side to obtain a trained neural network model;

[0021] The trained neural network model is uploaded to the satellite's NPU (Neural Network Processing Unit) through the ground-to-space link, so that the NPU outputs a judgment result based on the preprocessed information and the trained neural network model; wherein the NPU is used to determine whether the current remote control is abnormal based on the satellite's past remote control and telemetry time series.

[0022] In a second aspect, an embodiment of the present application provides an AI safety monitoring device for satellite remote control and telemetry, including:

[0023] An acquisition module, used for acquiring original information from the bus, wherein the original information includes: remote control information and telemetry information;

[0024] A preprocessing module, used for preprocessing the original information to obtain preprocessed information, wherein the preprocessing method includes: partially changing the numerical value of the binary data and supplementing the missing bits;

[0025] An AI monitoring and processing module is used to input the pre-processed information into a trained neural network model, and the trained neural network model performs AI monitoring processing to obtain a judgment result, wherein the judgment result is used to determine whether the original information has an abnormality;

[0026] The sending module is used to send the judgment result to the central processing unit CPU for processing.

[0027] Optionally, it also includes: an information insertion module, which is used to insert all-0 information between each remote control information and telemetry information after obtaining the original information from the bus to indicate the time interval.

[0028] Optionally, the preprocessing module is specifically used to:

[0029] The original information is represented in binary form, and the value 1 in the binary data is kept unchanged, and 0 in the binary data is changed to -1, so as to obtain the changed data;

[0030] The changed data is supplemented with missing bits so that the data is expanded from M bits to N bits, wherein the missing bits are supplemented with 0, M and N are positive integers, and M <N。

[0031] Optionally, the above device further includes:

[0032] A model building module is used to build a neural network model based on a long short-term memory network LSTM before inputting the pre-processed information into a pre-trained neural network model; wherein the neural network model includes: a K-layer recurrent neural network and an L-layer fully connected neural network, K and L are positive integers; the neural network model includes: an input gate, a forget gate, an output gate, and through c t The hidden state at time t is recorded, where the hidden state contains the LSTM's summary of all input information before time t.

[0033] Optionally, the device further comprises:

[0034] The download module is used to download the original information obtained by the satellite from the bus to the ground terminal through the ground-to-space link;

[0035] A training module, used to train the neural network model through a graphics processing unit (GPU) on the ground side to obtain a trained neural network model;

[0036] An upload module is used to upload the trained neural network model to the satellite's embedded neural network processor NPU through a ground-to-space link, so that the NPU can output a judgment result based on the preprocessed information and the trained neural network model; wherein the NPU is used to determine whether the current remote control is abnormal based on the satellite's past remote control and telemetry time series.

[0037] In a third aspect, an embodiment of the present application provides an AI safety monitoring device for satellite remote control and telemetry, including: a processor and a memory, wherein the memory stores executable program instructions, and when the processor calls the program instructions in the memory, the processor is used to:

[0038] Execute the steps of the AI ​​safety monitoring method for satellite remote control and telemetry as described in any one of the first aspects.

[0039] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium for storing a program, which, when executed, implements the steps of the AI ​​security monitoring method for satellite remote control and telemetry as described in any one of the first aspects.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] In this application, the original information is obtained from the bus, wherein the original information includes: remote control information and telemetry information; the original information is preprocessed to obtain the preprocessed information, wherein the preprocessing method includes: partial numerical changes to binary data, and missing supplements; the preprocessed information is input into a trained neural network model, and the trained neural network model performs AI monitoring and processing to obtain a judgment result, wherein the judgment result is used to determine whether the original information is abnormal; the judgment result is sent to the central processing unit CPU for processing. In this way, the operating status of the satellite can be monitored in real time, and the current remote control can be judged whether it is abnormal through the satellite's past remote control and telemetry time series, and potential safety hazards can be automatically discovered and responded quickly, realizing a satellite AI safety monitoring method based on time series, and enhancing the safety management of satellites. In addition, the method in this application also has powerful data processing and analysis capabilities, can accurately identify various complex attack modes, and provide strong protection for the safe operation of satellites. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings in the following descriptions are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without creative work. By reading the detailed description of the non-limiting embodiments with reference to the following drawings, other features, purposes and advantages of the present invention will become more obvious:

[0043] Figure 1 A flowchart of an AI safety monitoring method for satellite remote control and telemetry provided according to an embodiment of the present application;

[0044] Figure 2 A flowchart of another satellite remote control and telemetry AI safety monitoring method provided according to an embodiment of the present application;

[0045] Figure 3 A flowchart of bus information preprocessing provided according to an embodiment of the present application;

[0046] Figure 4 A flowchart of an AI monitoring processing algorithm provided according to an embodiment of the present application;

[0047] Figure 5 A diagram of a ground-ground collaborative training reasoning architecture provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0049] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there can be a central component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there can be a central component at the same time.

[0050] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0051] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein, for example, can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0052] The following specific embodiments are used to describe in detail the technical solution of the present invention and how the technical solution of the present application solves the above technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0053] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0054] For example, Figure 1 A flowchart of an AI safety monitoring method for satellite remote control and telemetry provided according to an embodiment of the present application is shown in FIG. Figure 1 As shown, the method in this embodiment may include the following steps:

[0055] Step S101: Acquire original information from the bus.

[0056] In this embodiment, the original information includes: remote control information and telemetry information.

[0057] Exemplarily, after acquiring the original information from the bus, all-0 information is inserted between each remote control information and telemetry information to indicate the time interval, thereby obtaining the remote control information and telemetry information containing the time series.

[0058] Step S102: pre-process the original information to obtain pre-processed information.

[0059] In this embodiment, the preprocessing method includes: partially changing the numerical values ​​of the binary data and supplementing the missing bits.

[0060] Exemplarily, the original information is represented in binary form, and the value 1 in the binary data is kept unchanged, and the 0 in the binary data is changed to -1 to obtain the changed data; the changed data is supplemented with missing bits so that the data is expanded from M bits to N bits, wherein the missing bits are supplemented with 0, M and N are positive integers, and M <N。

[0061] Step S103: input the preprocessed information into the trained neural network model, and the trained neural network model performs AI monitoring processing to obtain a judgment result.

[0062] In this embodiment, the judgment result is used to determine whether the original information has an abnormality.

[0063] Exemplarily, before the preprocessed information is input into the pre-trained neural network model, a neural network model based on a long short-term memory network LSTM is first constructed; wherein the neural network model includes: a K-layer recurrent neural network and an L-layer fully connected neural network, K and L are positive integers; the neural network model includes: an input gate, a forget gate, an output gate, and through c t The hidden state at time t is recorded, where the hidden state contains the LSTM's summary of all input information before time t.

[0064] In this embodiment, the original information obtained by the satellite from the bus can be downloaded to the ground end through the space-ground link; the neural network model is trained by the graphics processor GPU of the ground end to obtain a trained neural network model; the trained neural network model is uploaded to the satellite's embedded neural network processor NPU through the space-ground link, so that the NPU can output a judgment result based on the preprocessed information and the trained neural network model; wherein the NPU is used to determine whether the current remote control is abnormal based on the satellite's past remote control and telemetry time series.

[0065] Step S104: Send the judgment result to the central processing unit CPU for processing.

[0066] In this embodiment, the original information is obtained from the bus, where the original information includes: remote control information and telemetry information; the original information is preprocessed to obtain preprocessed information, where the preprocessing method includes: partially changing the numerical values of binary-form data and supplementing missing positions; the preprocessed information is input into a trained neural network model, and the trained neural network model performs AI monitoring processing to obtain a judgment result, where the judgment result is used to determine whether there is an abnormality in the original information; the judgment result is sent to the central processing unit CPU for processing. Thus, the operating state of the satellite can be monitored in real time, whether the current remote control is abnormal can be judged through the satellite's past remote control and telemetry time series, potential security hazards can be automatically discovered, and a quick response can be made, realizing a satellite AI security monitoring method based on time series and enhancing the security management of the satellite. In addition, the method in this application also has strong data processing and analysis capabilities, can accurately identify various complex attack patterns, and provides a strong guarantee for the safe operation of the satellite.

[0067] Figure 2 FIG. is a flowchart of another satellite remote control and telemetry AI security monitoring method provided according to an embodiment of the present application. As Figure 2 shown, first, information is obtained from the bus, including remote control and telemetry information; the information is preprocessed, keeping the binary 1 unchanged, converting the binary 0 to -1, and supplementing 0 for missing positions to match the subsequent interface; the preprocessed information is subjected to AI monitoring processing based on the LSTM neural network, and a judgment is made on whether the current information is abnormal based on the past information; finally, the CPU disposes according to the judgment result.

[0068] In this embodiment, all-0 information is inserted between each remote control / telemetry to represent the time interval.

[0069] Figure 3 FIG. is a flowchart of the bus information preprocessing provided according to an embodiment of the present application. As Figure 3 shown, first, the binary 1 in the bus original information remains unchanged, and the binary 0 is converted to -1 to implement normalization processing; then the data is expanded from 384 bits to 512 bits, and the missing positions are filled with 0 to match the subsequent interface.

[0070] It should be noted that the above-described flowcharts are merely illustrative, and the missing position filling with 0 operation can be expanded from M bits to N bits (M and N are positive integers, and M < N). Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0071] Figure 4 FIG. is a flowchart of the AI monitoring processing algorithm provided according to an embodiment of the present application. As Figure 4As shown, in this embodiment, the algorithm includes: 3 layers of LSTM and 2 layers of fully connected neural network, and LSTM adopts a unidirectional design. Among them, the LSTM neural network includes an input gate, a forget gate, and an output gate. t Record the hidden state at time t, which contains the LSTM's summary of all input information before time t.

[0072] It should be noted that the flowchart described above is only schematic, wherein the neural network may be a K-layer recurrent neural network and an L-layer fully connected neural network (K and L are positive integers, and the recurrent neural network may also be RNN (Recurrent Neural Network), GRU (Gate Recurrent Unit), etc.). Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0073] Figure 5 This is a diagram of the ground-ground collaborative training reasoning architecture provided according to an embodiment of the present application, such as Figure 5 As shown, this embodiment mainly includes: the bus information acquired by the satellite is downloaded to the ground through the ground-ground link; the GPU on the ground trains the bus information to obtain a model (i.e., a neural network model) for AI monitoring and processing; the AI ​​model (i.e., the neural network model) is uploaded to the NPU of the satellite through the ground-ground link; the NPU acquires information from the bus, performs AI monitoring and processing on it, and obtains a judgment result; the CPU handles it according to the judgment result.

[0074] In this embodiment, the ground uses a GPU to implement the training part of the ground-ground collaborative training inference architecture. The input is bus information and the output is an AI model (i.e., a neural network model).

[0075] In this embodiment, the satellite implements the reasoning part of the ground-ground collaborative training reasoning architecture through the NPU. The input is the AI ​​model (i.e., the neural network model) and bus information, and the output is the judgment result.

[0076] This embodiment can monitor the operating status of the satellite in real time, and can determine whether the current remote control is abnormal through the satellite's past remote control and telemetry time series, automatically discover potential safety hazards, and respond quickly, realizing a satellite AI safety monitoring method based on time series, and enhancing satellite safety management. In addition, the method in this application also has powerful data processing and analysis capabilities, can accurately identify various complex attack modes, and provide strong protection for the safe operation of satellites.

[0077] Exemplarily, an embodiment of the present application further provides an AI safety monitoring device for satellite remote control and telemetry, including:

[0078] An acquisition module, used for acquiring original information from the bus, wherein the original information includes: remote control information and telemetry information;

[0079] A preprocessing module, used for preprocessing the original information to obtain preprocessed information, wherein the preprocessing method includes: partially changing the numerical value of the binary data and supplementing the missing bits;

[0080] An AI monitoring and processing module is used to input the pre-processed information into a trained neural network model, and the trained neural network model performs AI monitoring processing to obtain a judgment result, wherein the judgment result is used to determine whether the original information has an abnormality;

[0081] The sending module is used to send the judgment result to the central processing unit CPU for processing.

[0082] Exemplarily, the above-mentioned device also includes: an information insertion module, which is used to insert all-0 information between each remote control information and telemetry information after obtaining the original information from the bus to indicate the time interval.

[0083] Exemplarily, the preprocessing module is specifically used for:

[0084] The original information is represented in binary form, and the value 1 in the binary data is kept unchanged, and 0 in the binary data is changed to -1, so as to obtain the changed data;

[0085] The changed data is supplemented with missing bits so that the data is expanded from M bits to N bits, wherein the missing bits are supplemented with 0, M and N are positive integers, and M <N。

[0086] Exemplarily, the above device further includes:

[0087] A model building module is used to build a neural network model based on a long short-term memory network LSTM before inputting the pre-processed information into a pre-trained neural network model; wherein the neural network model includes: a K-layer recurrent neural network and an L-layer fully connected neural network, K and L are positive integers; the neural network model includes: an input gate, a forget gate, an output gate, and through c t The hidden state at time t is recorded, where the hidden state contains the LSTM's summary of all input information before time t.

[0088] Exemplarily, the device further comprises:

[0089] The download module is used to download the original information obtained by the satellite from the bus to the ground terminal through the ground-to-space link;

[0090] A training module, used to train the neural network model through a graphics processing unit (GPU) on the ground side to obtain a trained neural network model;

[0091] An upload module is used to upload the trained neural network model to the satellite's embedded neural network processor NPU through a ground-to-space link, so that the NPU can output a judgment result based on the preprocessed information and the trained neural network model; wherein the NPU is used to determine whether the current remote control is abnormal based on the satellite's past remote control and telemetry time series.

[0092] An embodiment of the present application also provides an AI safety monitoring device for satellite remote control and telemetry, including: a processor and a memory, wherein the memory stores executable program instructions, and when the processor calls the program instructions in the memory, the processor is used to: execute the steps in the above-mentioned method embodiments.

[0093] The embodiment of the present application also provides a computer-readable storage medium for storing a program, and when the program is executed, the steps in the above-mentioned method embodiments are implemented.

[0094] It should be noted that those skilled in the art will appreciate that various aspects of the present invention may be implemented as systems, methods or program products. Therefore, various aspects of the present invention may be specifically implemented in the following forms, namely: complete hardware implementation, complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to as "circuit", "module" or "platform" herein.

[0095] In addition, an embodiment of the present application further provides a computer-readable storage medium, in which computer-executable instructions are stored. When at least one processor of a user device executes the computer-executable instructions, the user device executes the above-mentioned various possible methods. Among them, the computer-readable medium includes a computer storage medium and a communication medium, wherein the communication medium includes any medium that facilitates the transmission of a computer program from one place to another. The storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer. An exemplary storage medium is coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an ASIC. In addition, the ASIC can be located in a user device. Of course, the processor and the storage medium can also exist in a communication device as discrete components.

[0096] The present application also provides a program product, which includes a computer program. The computer program is stored in a readable storage medium. At least one processor of the server can read the computer program from the readable storage medium. At least one processor executes the computer program so that the server implements any method of the above-mentioned embodiments of the present invention.

[0097] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0098] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art may make various modifications or variations within the scope of the claims, which do not affect the essence of the present invention.

Claims

1. An AI safety monitoring method for satellite remote control and telemetry, characterized in that: include: Acquire original information from the bus, wherein the original information includes: remote control information and telemetry information; Preprocessing the original information to obtain preprocessed information, wherein the preprocessing method includes: partially changing the values ​​of binary data and supplementing missing bits; Inputting the preprocessed information into a trained neural network model, and performing AI monitoring processing by the trained neural network model to obtain a judgment result, wherein the judgment result is used to determine whether the original information has an abnormality; The judgment result is sent to the central processing unit CPU for processing.

2. The AI ​​safety monitoring method for satellite remote control and telemetry according to claim 1 is characterized in that: After acquiring the original information from the bus, the method further includes: All-0 information is inserted between each remote control message and telemetry message to indicate the time interval.

3. The AI ​​safety monitoring method for satellite remote control and telemetry according to claim 1 is characterized in that: Preprocessing the original information to obtain preprocessed information includes: The original information is represented in binary form, and the value 1 in the binary data is kept unchanged, and 0 in the binary data is changed to -1, so as to obtain the changed data; The changed data is supplemented with missing bits so that the data is expanded from M bits to N bits, wherein the missing bits are supplemented with 0, M and N are positive integers, and M <N。 4. The AI ​​safety monitoring method for satellite remote control and telemetry according to any one of claims 1 to 3, characterized in that: Before inputting the preprocessed information into the pre-trained neural network model, the method further includes: Construct a neural network model based on long short-term memory network LSTM; wherein the neural network model includes: K layers of recurrent neural network and L layers of fully connected neural network, K and L are positive integers; the neural network model includes: input gate, forget gate, output gate, and through c t The hidden state at time t is recorded, where the hidden state contains the LSTM's summary of all input information before time t.

5. The AI ​​safety monitoring method for satellite remote control and telemetry according to claim 4 is characterized in that: The method further comprises: Download the original information obtained by the satellite from the bus to the ground terminal through the space-ground link; The neural network model is trained by a graphics processor (GPU) on the ground side to obtain a trained neural network model; The trained neural network model is uploaded to the satellite's embedded neural network processor NPU through the ground-to-space link, so that the NPU outputs a judgment result based on the preprocessed information and the trained neural network model; wherein the NPU is used to determine whether the current remote control is abnormal based on the satellite's past remote control and telemetry time series.

6. An AI safety monitoring device for satellite remote control and telemetry, characterized in that: include: An acquisition module, used for acquiring original information from the bus, wherein the original information includes: remote control information and telemetry information; A preprocessing module, used for preprocessing the original information to obtain preprocessed information, wherein the preprocessing method includes: partially changing the numerical value of the binary data and supplementing the missing bits; An AI monitoring and processing module is used to input the pre-processed information into a trained neural network model, and the trained neural network model performs AI monitoring processing to obtain a judgment result, wherein the judgment result is used to determine whether the original information has an abnormality; The sending module is used to send the judgment result to the central processing unit CPU for processing.

7. The AI ​​safety monitoring device for satellite remote control and telemetry according to claim 6 is characterized in that: Also includes: The information insertion module is used to insert all-0 information between each remote control information and telemetry information after obtaining the original information from the bus to indicate the time interval.

8. The AI ​​safety monitoring device for satellite remote control and telemetry according to claim 6 is characterized in that: The preprocessing module is specifically used for: The original information is represented in binary form, and the value 1 in the binary data is kept unchanged, and 0 in the binary data is changed to -1, so as to obtain the changed data; The changed data is supplemented with missing bits so that the data is expanded from M bits to N bits, wherein the missing bits are supplemented with 0, M and N are positive integers, and M <N。 9. An AI safety monitoring device with satellite remote control and telemetry, characterized in that: include: A processor and a memory, wherein the memory stores executable program instructions, and when the processor calls the program instructions in the memory, the processor is used to: Execute the steps of the AI ​​safety monitoring method for satellite remote control and telemetry as described in any one of claims 1 to 5.

10. A computer-readable storage medium for storing a program, characterized in that: When the program is executed, the steps of the AI ​​safety monitoring method for satellite remote control and telemetry described in any one of claims 1 to 5 are implemented.