Low-orbit aircraft remote control instruction generation and management system and method and storage medium

Abstract

The invention provides a low-orbit aircraft remote control instruction generation and management system which comprises an extraction module, an input module, a configuration module, a coding module, a sending module, an assembly module, a storage module, a modification module, a task management module and the like. The core problems of low automation degree, insufficient safety, poor flexibility and the like in instruction generation and management of the low-orbit aircraft in the prior art are solved. Compared with a traditional method, the method has the advantages that the efficiency is improved by more than 50%, the security reaches the quantum attack resistance level, the task scheduling flexibility is improved by 40%, and a whole-process solution is provided for complex instruction management of the low-orbit aircraft.

Description

Technical Field

[0001] The present invention belongs to the technical field of low-orbit aircraft control, and in particular to a low-orbit aircraft remote control instruction generation and management system, method and storage medium. Background Art

[0002] With the large-scale application of low-orbit spacecraft (orbital altitude 200km-2000km) in fields such as communications and remote sensing, their remote control command generation and management systems face multiple technical challenges. The existing technical system has significant bottlenecks in automation, security, and dynamic response capabilities, as shown below: (1) The traditional measurement and control process is inefficient. Traditional low-orbit spacecraft management relies on ground measurement and control station networks and manual operations, and requires the coordination of multiple links, including "demand submission - plan generation - command sending". Taking a single-satellite mission as an example, it is necessary to mobilize hundreds of personnel in command and dispatch, telemetry monitoring, orbit calculation, etc., and the mission response cycle is as long as several hours to several days. This model is difficult to meet the needs of multi-satellite coordination and dynamic mission adjustment in low-orbit constellations. For example, in emergency disaster monitoring or military reconnaissance scenarios, low-orbit spacecraft need to complete the imaging mode switching of multiple targets (such as sliding beamforming and multi-area stitching) within a single orbit, and the traditional process cannot update mission instructions in real time; (2) Insufficient full-time online measurement and control capabilities. Low-orbit aircraft application missions require "four anytime" capabilities (reporting, response, service, and feedback at any time), but the existing system still relies on the communication arc with limited coverage of ground-based measurement and control stations. For example, before a low-orbit aircraft leaves, it is necessary to inject instructions through the ground station, resulting in new tasks having to wait for the next circle to be executed, and emergency response delays of up to 3.8 seconds to several minutes. Although space-based measurement and control technologies (such as relay low-orbit aircraft and Beidou short messages) can expand coverage, they face problems such as link capacity limitations (such as Beidou's single-packet transmission success rate of 95.5%) and delay fluctuations (Doppler frequency deviation at a measured rate of 492kbps in the L-band channel); (3) There are hidden dangers in the secure transmission mechanism. Existing encryption technologies are mainly symmetric encryption (AES) and asymmetric encryption (RSA), but they face the risk of quantum computing cracking. For example, traditional key distribution relies on ground station transfer and is vulnerable to man-in-the-middle attacks; on-board resource limitations make it difficult for lightweight encryption algorithms to balance security and computational efficiency. In addition, low-orbit aircraft network nodes are dispersed and highly physically exposed. Attackers can use side-channel attacks (such as electromagnetic leakage) or tamper with onboard control systems to threaten data integrity and privacy; (4) Inefficient resource scheduling and signal processing. When multiple devices are networked, the low-orbit aircraft receiving system often has the problem of repeated signal reception. For example, multiple downlink signal processing devices capture the same low-orbit aircraft signal at the same time, resulting in a 20% to 30% decrease in resource utilization. In the low-orbit aircraft scenario with staring function, the rapid movement of terminal devices forces frequent switching of low-orbit aircraft (a single switch takes > 500ms), and there is a lack of predictive scheduling strategies based on the trajectory, resulting in a communication interruption rate of up to 15%; (5) Lack of intelligent decision-making capabilities. Currently, low-orbit aircraft operations are mainly based on preset rules and ground control, and the autonomous decision-making capabilities on board are weak. For example, obstacle avoidance instructions require the ground station to calculate the probability of orbital collision and then send it back, which takes several hours; data processing relies on the ground center, and the timeliness of polar ice sheet monitoring data is delayed by several months. Although some studies have attempted to introduce AI chips (such as NOVI SP240) and digital twin technology, fully autonomous dynamic decision-making has not yet been achieved, making it difficult to cope with highly dynamic space environments (such as debris avoidance and anti-interference communications).

[0003] Summary of technical bottlenecks: The existing system has significant defects in instruction generation, encryption, scheduling and verification, which are manifested as follows: low degree of automation: reliance on manual configuration and static templates / rules; weak security: traditional encryption methods are vulnerable to quantum attacks; poor scheduling flexibility: fixed priority strategies lead to low resource utilization; insufficient verification fault tolerance: manual intervention and rigid rule base.

[0004] These issues severely restrict the efficiency and safety of low-orbit spacecraft mission execution. Therefore, an instruction management system that integrates intelligent prediction, quantum encryption, and dynamic scheduling is urgently needed to overcome existing technical bottlenecks. Summary of the Invention

[0005] The present invention proposes a system, method, and storage medium for generating and managing remote control commands for low-orbit aircraft. These solutions address the existing issues of low automation, reliance on manual configuration and static templates / rules, weak security, with traditional encryption methods vulnerable to quantum attacks, and poor scheduling flexibility. These solutions meet the requirements for high efficiency, flexibility, and security in low-orbit aircraft mission management. The technical solution of the present invention is achieved as follows: A low-orbit aircraft remote control command generation and management system, including the following modules: An extraction module, which extracts instructions from historical instruction sets and subsystems of low-orbit vehicles and builds metadata models through deep learning; The input module converts user input into a structured instruction model through natural language processing (NLP); Configuration module, which optimizes command parameters based on dynamic mission requirements and low-orbit vehicle operating conditions; The encoding module uses quantum key distribution (QKD) technology to encrypt instructions; The sending module selects the optimal path to transmit the command code according to the status of the low-orbit aircraft communication network; Assemble modules, generate instruction clusters through genetic algorithms and optimize task scheduling; Storage module, which manages instruction clusters using cloud storage and caching algorithms; Modify the module and correct the instructions that fail the verification through the self-learning algorithm; The task management module combines reinforcement learning to dynamically adjust instruction priorities and resource allocation.

[0006] As a preferred technical solution, the extraction module includes: A convolutional neural network model is used to predict future command requirements based on historical command data, status data, and environmental data of low-orbit spacecraft. The specific formula is as follows:

[0007] in: is the set of instructions predicted at time t; is the historical command data of the low-orbit spacecraft at time t; is the status data of the low-orbit spacecraft at time t; For environmental data; A metadata model that defines the structured parameter types and constraints of instructions.

[0008] As an optimal technical solution, the input module realizes the automatic conversion of natural language to instruction model by minimizing the error between natural language instructions and structured instructions. The instruction parsing process:

[0009] Where: M is the instruction metadata model converted by NLP; Natural language instructions entered by the user; Structured instructions for the output of a machine learning model.

[0010] As a preferred technical solution, the configuration module dynamically matches the optimal parameters of the low-orbit aircraft operating conditions through a parameter optimization algorithm.

[0011] As a preferred technical solution, the encoding module adopts a hybrid encryption method to perform dual encryption processing on the original instructions using traditional encryption keys and quantum keys. The specific processing method is as follows:

[0012] in: is the encrypted instruction; For encryption operations; I is the original instruction; K is the traditional encryption key; QKD(K) is a quantum key generated based on quantum key distribution technology.

[0013] As a preferred technical solution, the sending module selects the communication path through a multi-objective optimization model. The routing optimization model is:

[0014] in: is the distance of the communication path; For communication delay; The reliability of the communication path; The weight coefficient for path optimization.

[0015] As a preferred technical solution, the assembly module optimizes the combination of instruction clusters through genetic algorithms. The priority scheduling goal is to maximize the weighted sum of task weight and urgency. The model is as follows:

[0016] Where P is the priority of the instruction; is the weight of the task; The urgency of the task.

[0017] As an optimal technical solution, the task management module dynamically adjusts the instruction execution strategy through the reinforcement learning model.

[0018] in: Q(s,a) is the value of performing action a in state s; r is the immediate reward; γ is the discount factor; For the next state, For the next action.

[0019] A method for generating and managing remote control instructions for a low-orbit aircraft uses a low-orbit aircraft remote control instruction generation and management system as described above to perform the extraction, entry, configuration, encoding, transmission, assembly, storage, modification, and task management tasks of a low-orbit aircraft.

[0020] A non-temporary storage medium is used to store a method for generating and managing remote control instructions for a low-orbit aircraft, so that the system can be run on a low-orbit aircraft.

[0021] Compared with the existing technology, this solution has the following beneficial effects: (1) Through the deep learning model of the extraction module and the natural language processing (NLP) technology of the input module, command demand prediction and automatic analysis of natural language input are achieved, reducing manual intervention by 90% and completely eliminating the inefficiency of traditional manual configuration; (2) The encoding module adopts a hybrid scheme of quantum key distribution (QKD) and traditional encryption, which reduces the risk of key leakage by several orders of magnitude while ensuring real-time performance (encryption and decryption delay <100ms), effectively resisting quantum computing attacks; (3) Based on the genetic algorithm of the assembly module (task priority optimization goal) and the reinforcement learning model of the task management module, dynamic matching of low-orbit spacecraft resources and mission requirements is achieved, the mission conflict rate is reduced from 15% to below 3%, and the emergency mission response time is shortened to seconds; (4) By modifying the module's self-learning verification algorithm (to minimize errors) and the storage module's cache optimization strategy, the success rate of self-correction of erroneous instructions is ≥98%. At the same time, it supports concurrent processing of tens of thousands of instructions per second, significantly improving the system's fault tolerance and throughput. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0023] Figure 1 This is a structural block diagram of a low-orbit aircraft remote control command generation and management system of the present invention. DETAILED DESCRIPTION

[0024] The following will clearly and completely describe the technical solutions of the present invention in conjunction with the embodiments of the present invention. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0025] Reference Figure 1This invention provides a system for generating and managing remote control commands for low-orbit aircraft. By integrating innovative technologies such as deep learning, quantum encryption, multi-layer command combination, and intelligent configuration, the system can efficiently and flexibly generate and manage remote control commands, ensuring their reliability and security. The following is a detailed description of each module: 1. Extraction module The extraction module's primary function is to extract applicable instructions from the historical instruction sets of each LEO vehicle and its subsystems and construct a metadata model. To enhance automation, this module employs deep learning technology, training a convolutional neural network (CNN) model to learn the most commonly used instruction sets from historical LEO data.

[0026] Command prediction model: Through deep learning models, it automatically predicts the command requirements of low-orbit aircraft at future times. The formula is as follows: , in: is the set of instructions predicted at time t; is the historical command data of the low-orbit spacecraft at time t; The state data of the low-orbit spacecraft at time t (such as position, speed, etc.); Environmental data (such as temperature, battery status, etc.).

[0027] 2. Input module

[0028] The input module enters the instructions and parameters obtained from the extraction module into the database according to the metadata model. Through natural language processing (NLP) technology, users can describe instructions in natural language, and the system will automatically convert them into a structured instruction model.

[0029] Instruction parsing process: , in: M is the instruction metadata model after NLP conversion; Natural language instructions entered by the user; Structured instructions for the output of a machine learning model.

[0030] 3. Configuration module

[0031] The configuration module allows users to flexibly configure commands and their parameters and generate single commands. The system uses an intelligent parameter adjustment algorithm to dynamically adjust command parameters based on the low-orbit vehicle's operating environment and mission requirements.

[0032] Parameter optimization goal: , in: Command parameters configured for the user; Optimization parameters recommended by the system.

[0033] 4. Encoding module

[0034] The encoding module is responsible for encoding instructions and generating instruction codes. By introducing quantum key distribution (QKD) technology, the security of instructions during transmission is ensured.

[0035] Quantum cryptography security: , in: is the encrypted instruction; I is the original instruction; K is the traditional encryption key; QKD(K) is a quantum key generated based on quantum key distribution technology.

[0036] 5. Sending module

[0037] The sending module sends the command code to the corresponding low-orbit aircraft. The system uses an intelligent routing algorithm to select the optimal path for data transmission based on the communication network status of the low-orbit aircraft.

[0038] Route optimization model: , in: is the distance of the communication path; For communication delay; The reliability of the communication path; , The weight coefficient for path optimization.

[0039] 6. Assemble the modules

[0040] The assembly module supports the combination and configuration of multiple instructions to generate instruction clusters. Genetic algorithms are used to optimize the combination of instructions to achieve optimal task scheduling.

[0041] Priority scheduling goals: , in: P is the priority of the instruction; is the weight of the task; The urgency of the task.

[0042] 7. Storage module

[0043] The storage module manages commonly used instruction clusters through cloud storage technology and improves access efficiency through cache optimization algorithms.

[0044] Storage efficiency formula: , in: is the number of cache hits; The total number of data requests.

[0045] 8. Modify the module

[0046] The modification module is used to modify instructions that have not passed verification until they pass verification. Through the self-learning algorithm, the system can automatically adjust the verification rules based on historical correction data.

[0047] Verification and correction targets: , in: Instructions that failed verification; This is the revised instruction.

[0048] 9. Task management and intelligent decision support

[0049] Deep learning and reinforcement learning algorithms are introduced to dynamically adjust the priority of instructions, resource allocation and task scheduling, and optimize the mission execution efficiency of low-orbit aircraft.

[0050] Reinforcement Learning Optimization Model: , in: Q(s,a) is the value of performing action a in state s; r is the immediate reward; γ is the discount factor; For the next state, For the next action.

[0051] Beneficial effects of this application

[0052] Aiming at the technical bottlenecks of existing low-orbit aircraft remote control command systems, this application has achieved the following core breakthroughs through technological integration and innovation: 1. Full process automation and multi-satellite collaboration efficiency improvement, The system digitizes the entire flight control process from design to execution. It supports multi-satellite parallel flight control design input, collaborative editing of flight events, and batch generation of execution files. For example, the system can output multi-satellite multi-event remote control commands and telemetry criteria with one click, and automatically integrate telemetry data to generate interpretation pages, significantly reducing manual editing and review time. This breakthrough directly addresses the automation shortcomings of traditional manual compilation of operation documents and processing instructions one by one. 2. Enhanced full-time online measurement and control and dynamic response capabilities. The system integrates space-based tracking and control (such as relaying low-orbit aircraft and BeiDou short messages) with ground-based networks to build a space-ground integrated tracking and control service system. The system supports full arc coverage of low-orbit aircraft through multi-method integrated tracking and control terminals, and reduces the response delay of new tasks from several hours to seconds. 3. Optimize resource scheduling and signal processing efficiency, An intelligent signal distribution algorithm is used to solve the problem of repeated signal reception in multi-device networks. By dynamically matching the trajectory of low-orbit aircraft with the location of ground terminals, the system can predict the signal coverage range and allocate the optimal receiving device, improving resource utilization and significantly enhancing the user experience. 4. Upgraded secure transmission and anti-interference capabilities. The introduction of a hybrid encryption mechanism and low-orbit signal enhancement technology provides dual guarantees for command transmission security. On the one hand, by increasing the signal power of the low-orbit aircraft, the anti-interference capability in complex terrain and electromagnetic environments is enhanced; on the other hand, the system supports dynamic encryption protocol switching between satellite and ground links to resist the risk of quantum computing attacks. 5. Intelligent verification and self-learning optimization,

[0053] A command verification model based on machine learning algorithms enables automatic identification and correction of abnormal commands. The system optimizes verification logic through historical data training, significantly improving the success rate of self-correction of erroneous commands. Furthermore, the system supports real-time feedback and parameter tuning of low-orbit vehicle operating data.

[0054] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A low-orbit aircraft remote control command generation and management system, characterized in that: Includes the following modules: An extraction module, which extracts instructions from historical instruction sets and subsystems of low-orbit vehicles and builds metadata models through deep learning; The input module converts user input into a structured instruction model through natural language processing (NLP); Configuration module, which optimizes command parameters based on dynamic mission requirements and low-orbit vehicle operating conditions; The encoding module uses quantum key distribution (QKD) technology to encrypt instructions; The sending module selects the optimal path to transmit the command code according to the status of the low-orbit aircraft communication network; Assemble modules, generate instruction clusters through genetic algorithms and optimize task scheduling; Storage module, which manages instruction clusters using cloud storage and caching algorithms; Modify the module and correct the instructions that fail the verification through the self-learning algorithm; The task management module combines reinforcement learning to dynamically adjust instruction priorities and resource allocation.

2. A low-orbit aircraft remote control command generation and management system according to claim 1, characterized in that: The extraction module includes: A convolutional neural network model is used to predict future command requirements based on historical command data, status data, and environmental data of low-orbit spacecraft. The specific formula is as follows: ， in: is the set of instructions predicted at time t; is the historical command data of the low-orbit spacecraft at time t; is the status data of the low-orbit spacecraft at time t; For environmental data; A metadata model that defines the structured parameter types and constraints of instructions.

3. A low-orbit aircraft remote control command generation and management system according to claim 1, characterized in that: The input module realizes the automatic conversion of natural language to instruction model by minimizing the error between natural language instructions and structured instructions. The instruction parsing process is as follows: ， Where: M is the instruction metadata model converted by NLP; Natural language instructions input by users; Structured instructions for the output of a machine learning model.

4. A low-orbit aircraft remote control command generation and management system according to claim 1, characterized in that: The configuration module dynamically matches the optimal parameters of the low-orbit aircraft operating conditions through a parameter optimization algorithm.

5. The system for generating and managing remote control instructions for low-orbit aircraft according to claim 1, characterized in that: The encoding module uses a hybrid encryption method to perform dual encryption processing on the original instruction using traditional encryption keys and quantum keys. The specific processing method is as follows: ， in: is the encrypted instruction; Cryptographic operations; I is the original instruction; K is the traditional encryption key; QKD(K) is a quantum key generated based on quantum key distribution technology.

6. A low-orbit aircraft remote control command generation and management system according to claim 1, characterized in that: The sending module selects the communication path through a multi-objective optimization model. The routing optimization model is: ， in: is the distance of the communication path; For communication delay; The reliability of the communication path; The weight coefficient for path optimization.

7. The system for generating and managing remote control instructions for low-orbit aircraft according to claim 1, characterized in that: The assembly module optimizes the combination of instruction clusters through genetic algorithms. The priority scheduling goal is to maximize the weighted sum of task weight and urgency. The model is as follows: ， Where P is the priority of the instruction; is the weight of the task; The urgency of the task.

8. The low-orbit aircraft remote control command generation and management system according to claim 1, characterized in that: The task management module dynamically adjusts the instruction execution strategy through the reinforcement learning model. ， in: Q(s,a) is the value of performing action a in state s; r is the immediate reward; γ is the discount factor; For the next state, For the next action.

9. A method for generating and managing remote control instructions for a low-orbit aircraft, characterized in that: A low-orbit aircraft remote control command generation and management system as described in claims 1 to 8 above is used to perform low-orbit aircraft extraction, entry, configuration, encoding, sending, assembly, storage, modification, and task management tasks.

10. A non-transitory storage medium for storing the method for generating and managing remote control instructions for a low-orbit aircraft according to claim 9, so that the system for generating and managing remote control instructions for a low-orbit aircraft according to any one of claims 1 to 8 can be run on a low-orbit aircraft.

Images