Artificial intelligence aerospace infrastructure agent system

Abstract

The invention relates to the technical field of satellites, and discloses an artificial intelligence aerospace infrastructure agent system, which comprises a satellite agent, and the satellite agent comprises an energy management module, a calculation module, a task execution module and a communication module. The artificial intelligence aerospace infrastructure agent method comprises the following steps: acquiring real-time states of a plurality of satellite agents, performing task scheduling and resource allocation by using an optimal control theory, optimizing a task allocation strategy in combination with a game theory model, and optimizing a communication path according to a channel capacity theorem, so as to obtain an artificial intelligence space infrastructure agent; and the task sequence and the resource allocation strategy are adjusted in real time in the task execution process, so that efficient collaboration and task completion are ensured. According to the method, task scheduling is optimized by combining the optimal control theory and the game theory, a communication path is optimized by adopting the channel capacity theorem, the task execution sequence is adjusted in real time, adaptive resource allocation is performed, efficient cooperation between satellite agents is ensured, and the task execution efficiency, the resource utilization rate and the system stability are effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of satellite technology, and in particular to an artificial intelligence aerospace infrastructure intelligent system. Background Art

[0002] At present, resource scheduling and task execution in satellite systems generally rely on traditional scheduling algorithms. Most of these traditional methods are based on predetermined rules and static scheduling strategies, and cannot dynamically adjust resource allocation and task execution order according to the actual status of the satellite in real time; although some systems have introduced task priority scheduling, they still cannot fully consider the collaborative relationship between satellite agents; for complex multi-task, multi-satellite collaborative task scheduling environments, this static scheduling method leads to unreasonable resource allocation, task delays and low execution efficiency.

[0003] Existing technologies also have certain shortcomings in optimizing inter-satellite communications. Traditional communication path selection is usually fixed and fails to fully consider changes in real-time channel quality. The utilization rate of inter-satellite communication resources is not high, especially when the number of satellites is large and the mission complexity is high, channel congestion and data transmission delays are particularly prominent. This limitation leads to delays in information transmission during mission execution, affecting the efficiency of overall mission completion. Although some solutions have introduced channel selection algorithms, there is still a lack of a mechanism for dynamic optimization based on real-time feedback and mission requirements.

[0004] In addition, the existing technology is still relatively simple in terms of task allocation and computing resource management; most systems ignore the impact of real-time states such as satellite energy consumption and computing load on task execution, and fail to achieve intelligent task allocation and load optimization; during task execution, resource waste or uneven load often occurs, resulting in the energy or computing resources of the satellite intelligent body not being used most effectively, further slowing down the execution progress of the task; therefore, the present invention proposes an artificial intelligence aerospace infrastructure intelligent body system to solve the shortcomings of the existing technology. Summary of the invention

[0005] In response to the shortcomings of the existing technology, the present invention provides an artificial intelligence aerospace infrastructure intelligent agent system, which solves the problem that resource scheduling, task execution and communication optimization between satellite intelligent agents cannot be adjusted dynamically in real time, ensuring efficient energy utilization and task completion under multi-task and multi-satellite collaborative work.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: an artificial intelligence aerospace infrastructure agent system, including a satellite agent, wherein the satellite agent includes: Energy management module, used to manage the energy status of satellite intelligent body; A computing module for managing the computing load of satellite agents; Task execution module, used to execute tasks and track task progress; The communication module is used for exchanging information between satellite intelligent agents; wherein a plurality of satellite intelligent agents are interconnected and collaborated via the communication module.

[0007] Preferably, the energy management module includes: an energy consumption control unit for executing and calculating loads according to the missions of the satellite agent; The energy recovery unit is used to recover energy during task execution.

[0008] Preferably, the calculation module includes: A computing resource allocation unit is used to allocate resources according to task requirements and computing load; The computational load optimization unit is used to optimize the computational load of the satellite agent during mission execution.

[0009] Preferably, the task execution module includes: A task allocation unit, used to dynamically allocate tasks according to the energy status, computing load and communication bandwidth of the satellite agent; The task scheduling unit is used to adjust the task execution order according to the real-time status and task priority of the satellite intelligent body.

[0010] Preferably, the communication module includes: A channel selection unit, used to dynamically select the optimal communication path between satellite agents according to the channel quality; The data transmission unit is used to realize data transmission through the selected communication path and monitor the data transmission quality.

[0011] Preferably, the energy consumption of the satellite agent , calculate the load and task progress The following dynamic equations are satisfied: Energy consumption equation: ;

[0012] in: Satellite Agent At the moment The rate of change of energy; Satellite Agent At the moment energy state; Satellite Agent At the moment Power consumption; Satellite Agent Calculating Load The corresponding energy conversion efficiency coefficient; Satellite Agent In executing task assignment Additional energy consumption coefficient when Satellite Agent At the moment task allocation decisions; Calculate the load equation: ;

[0013] in: Satellite Agent At the moment The rate of change of the calculated load; Satellite Agent At the moment The computational load; Satellite Agent At the moment task allocation decisions; is a function of the computational load, depending on the task allocation decision and the current load; Task progress equation: ;

[0014] in: Satellite Agent At the moment The rate of change of task progress; Satellite Agent At the moment The progress of the task; Satellite Agent At the moment The computational load; Satellite Agent The task processing efficiency indicates the impact of computing load on task progress.

[0015] Preferably, the optimal control strategy of the satellite agent By solving the following optimal control problem: ; in, are weight coefficients, respectively controlling energy consumption , calculate the load and task progress Impact on the objective function; Satellite Agent At the moment energy state; Satellite Agent At the moment The computational load; Satellite Agent At the moment The progress of the task; Satellite Agent The control strategy includes power distribution , Task Allocation and communication strategies .

[0016] Preferably, the task collaboration among the satellite agents is optimized through a game theory model, and the equilibrium solution of the game satisfies the following conditions: ; in, Satellite Agent The optimal profit function is Satellite Agent The profit function, Satellite Agent The task allocation strategy, Satellite Agent Communication strategies, Represents the strategies of other satellite agents.

[0017] Preferably, the communication between the satellite agents is optimized by the following channel capacity theorem: ; in, Satellite Agent Satellite Agent The communication rate between Satellite Agent The transmission power, Satellite Agent Satellite Agent The channel gain between is the noise power spectral density.

[0018] The present invention also provides an artificial intelligence aerospace infrastructure agent method, comprising the following steps: S1. Obtain the real-time status of multiple satellite agents, including energy status, computing load and task progress; S2. According to the real-time status of the satellite intelligent body, use the optimal control theory to perform task scheduling and resource allocation to solve the optimal control strategy; S3, based on the game theory model, optimize the task allocation strategy through collaboration and game among satellite agents; S4, optimize the communication path between satellite agents according to the channel capacity theorem and select the optimal communication strategy; S5. During the task execution process, adjust the task execution order and resource allocation strategy in real time.

[0019] The present invention provides an artificial intelligence aerospace infrastructure intelligent agent system. It has the following beneficial effects: The present invention adopts a task scheduling method that combines optimal control theory and game theory models to achieve the technical effect of dynamically optimizing task allocation among multiple satellite intelligent agents. Compared with the prior art solutions that only rely on simple scheduling algorithms, the present invention can effectively consider the real-time resource conditions of satellite intelligent agents, avoid excessive concentration or waste of resources, and thus improve the efficiency of task completion and resource utilization.

[0020] The present invention combines the communication optimization strategy of the channel capacity theorem and the real-time feedback mechanism, thereby achieving the effect of optimizing the inter-satellite communication path in a complex environment; compared with the fixed communication path strategy commonly used in the prior art, the present invention can flexibly adjust the communication strategy according to the inter-satellite distance, channel quality and mission requirements, thereby solving the problems of communication congestion and data transmission delay, and ensuring efficient and stable information transmission in mission collaboration.

[0021] The present invention adopts real-time monitoring and adaptive adjustment mechanism to optimize the task execution order, so as to achieve the effect of flexibly responding to changes in the task execution process; compared with the traditional method, the present invention can dynamically adjust the task execution order according to the real-time task progress and resource status, avoiding task delays caused by lagging task progress or resource bottlenecks, thereby improving the execution stability and flexibility of the system.

[0022] The present invention optimizes resource allocation by collaboratively working with multiple satellite intelligent bodies, thereby achieving the technical effect of improving the overall system efficiency in a multi-task environment; compared with the prior art solution that only considers the resource allocation of a single satellite, the present invention achieves the technical effect of improving the overall system efficiency in a multi-task environment by collaboratively optimizing between satellites, so that the entire system can reasonably dispatch the computing resources, energy and communication bandwidth of each satellite while executing multiple tasks in parallel, thereby solving the resource conflict problem in multi-satellite collaboration and improving the overall efficiency of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 is a system architecture diagram of the present invention; Figure 2 It is a schematic diagram of the energy management module architecture of the present invention; Figure 3 It is a schematic diagram of the computing module architecture of the present invention; Figure 4 It is a schematic diagram of the task execution module architecture of the present invention; Figure 5It is a schematic diagram of the communication module architecture of the present invention; Figure 6 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] See also Figure 1-5 , an embodiment of the present invention provides an artificial intelligence aerospace infrastructure agent system, including a satellite agent, wherein the satellite agent includes: Energy management module, used to manage the energy status of satellite intelligent body; In the artificial intelligence aerospace infrastructure intelligent agent system of the present invention, the energy management module is one of the core functional modules of each satellite intelligent agent, and is responsible for monitoring and optimizing the energy utilization efficiency of the satellite; in order to ensure that each satellite intelligent agent can complete the task efficiently, the energy management module needs to combine task execution, computing load and communication status, and adjust the satellite's energy consumption and recovery strategy in real time to avoid excessive energy consumption or waste.

[0026] In the foregoing discussion, we have described how the energy management module can extend the satellite's operating time and improve energy utilization efficiency through energy consumption control and energy recovery mechanisms; in this embodiment, the energy management module includes two main sub-modules: an energy consumption control unit and an energy recovery unit; these two sub-modules work together to ensure that the satellite's energy consumption is optimized as much as possible, while being able to reasonably recover and store energy during the execution of the satellite mission.

[0027] The goal of the energy consumption control unit is to calculate and adjust the power allocation of the satellite according to the needs of satellite mission execution; when performing a mission, the satellite must consume a certain amount of energy to complete operations such as calculation and communication; therefore, at different mission stages, the satellite's energy consumption is closely related to factors such as mission execution, computing load and communication bandwidth.

[0028] In this embodiment, the energy consumption control unit adjusts the energy consumption according to the real-time state of the satellite through the following dynamic equation: ; in: Satellite Agent At the moment The energy consumption rate (i.e. the change in energy per unit time); For satellite At the moment The power allocation, i.e. the power required by the satellite to perform its mission; For satellite The energy conversion efficiency when performing computing tasks; generally speaking, the higher the satellite computing load, the more energy it consumes. Changes in value can reflect changes in computational load; For satellite At the moment The computational load represents the amount of computational tasks that the satellite needs to handle; For satellite The energy consumption coefficient when performing communication tasks characterizes the energy consumption during the communication process; For satellite At the moment The task allocation decision represents the degree of resource allocation of the satellite agent during task execution.

[0029] Through this formula, the energy consumption control unit can accurately calculate the power allocation of the satellite agent at each moment, and then adjust the energy consumption of the satellite; for example, when the satellite needs to process complex computing tasks or perform high-intensity communications, and The increase in power consumption will lead to an increase in power distribution. Dynamic adjustments are made based on these changes.

[0030] The energy recovery unit is responsible for recovering excess energy during the satellite's mission and storing it in the battery. During the mission, if the satellite's computing load or communication requirements are low, the satellite will not need to consume a large amount of energy. At this time, the system can start the recovery mode and store the excess energy for subsequent use.

[0031] In a specific implementation, the energy recovery unit monitors the energy status of the satellite and calculate load , to determine whether there is enough margin for energy recovery; when the satellite is in a low-load state (for example, entering standby mode after completing a mission, or performing low-intensity computing tasks), the unit will activate the energy recovery mechanism and store unnecessary energy in the battery, thereby ensuring that the energy requirements of the satellite's subsequent missions are met.

[0032] In a possible implementation, the recycling strategy is calculated according to the following formula: ; in: Indicates satellite At the moment The rate of change of energy; It's a satellite At the moment The battery energy status reflects the current remaining energy of the satellite; It's a satellite At the moment The computational load reflects whether the satellite is in a low load state; It means that the satellite is in a state of low computing load, that is, the satellite is in an idle period or a light mission load stage; Indicates the current status of the battery (for example, the remaining percentage of battery capacity).

[0033] if When the recycling conditions are met, the energy recovery unit will start the recovery mechanism and convert unnecessary energy into reserve battery energy. At this time, the system will transfer the excess energy into the battery until the battery is fully charged or reaches the set maximum charge. Through this mechanism, the satellite can maximize energy storage under low load conditions and avoid unnecessary energy waste.

[0034] When multiple satellite agents work together, energy management is not limited to the energy control of a single satellite, but also involves collaboration between satellites. In order to optimize the energy distribution of the entire system, the energy management module needs to perform global optimization scheduling based on the communication requirements and task execution status between satellites.

[0035] Specifically, the energy management module coordinates energy between satellites based on the following factors: The satellite's remaining energy and current mission requirements; Communication bandwidth and computing resources between satellites; The mission priority and execution order of each satellite.

[0036] When multiple satellites collaborate, the energy management module not only needs to ensure that each satellite has enough energy to perform its tasks, but also needs to coordinate energy resources between different satellites to avoid the situation where the energy of a certain satellite is too low and the task cannot be continued. For example, satellite A may consume a lot of energy when performing computing tasks, while satellite B has sufficient energy and can share part of its energy with satellite A to ensure that the overall operation of the system is not affected.

[0037] As an option, the energy management module can also be adaptively adjusted through machine learning algorithms. In this way, the system continuously optimizes the satellite's energy management strategy based on each satellite's historical mission data, energy consumption records and communication conditions, using algorithms such as deep learning or reinforcement learning. In this way, the satellite agent can automatically adjust its energy use strategy under different mission and environmental conditions to achieve better energy utilization.

[0038] For example, based on the experience of past mission execution, the energy management module can intelligently predict the energy requirements of future missions, so as to plan power allocation and energy recovery in advance; this adaptive learning mechanism enables the satellite intelligent body to make more accurate and efficient energy scheduling in complex environments, thereby further extending the satellite's working time and improving the stability and efficiency of mission execution.

[0039] A computing module for managing the computing load of satellite agents; In the artificial intelligence aerospace infrastructure intelligent agent system of the present invention, the computing module is an important component of each satellite intelligent agent, which is responsible for the computing load management of the satellite and needs to reasonably allocate computing resources according to the needs of task execution; each satellite intelligent agent uses the computing module to ensure the rational use of computing resources, avoid excessive computing load or waste of resources, and thus improve the efficiency of task execution.

[0040] The aforementioned energy management module has described how to control and optimize the energy consumption of the satellite, and the computing module further optimizes the satellite's computing resource allocation on this basis to ensure the high efficiency of task execution; the computing module and the energy management module cooperate with each other to ensure that resources are evenly allocated and achieve global optimal scheduling; in this embodiment, the computing module includes two core functional parts: a computing resource allocation unit and a computing load optimization unit, which work together to ensure that the satellite intelligent body can complete the task efficiently and stably.

[0041] The computing resource allocation unit is a key component in the computing module. Its function is to dynamically adjust the computing resources of the satellite agent according to the computing requirements of the task. Each satellite agent requires certain computing resources when performing a task. The computing resource allocation unit dynamically analyzes the task load and reasonably allocates computing resources to ensure efficient completion of the task and avoid overload or waste of computing resources.

[0042] In this embodiment, computing resource allocation is dynamically adjusted using the following formula: ; in: Satellite Agent At the moment The computing load change rate reflects the computing load change of the satellite during the mission execution; is a description of the satellite agent When executing a task, a function that calculates the load variation; Satellite Agent At the moment The computational load represents the amount of computing resources used by the satellite, reflecting the total amount of computing tasks being performed by the satellite; For satellite At the moment The task allocation decision represents the amount of tasks currently assigned to the satellite, that is, the computing requirements of the tasks.

[0043] The formula shows that the change of satellite computing load is closely related to the task allocation decision; when the satellite performs a task, the computing load The computing load will increase with the increase of the task allocation amount; conversely, if the task amount decreases, the computing load will decrease accordingly; by adjusting the task allocation strategy ,The computing resource allocation unit can optimize the computing load of each satellite in real time.

[0044] The computational load optimization unit is responsible for adjusting and optimizing the computational load distribution of the satellite. Since the computing power of satellite agents is limited, it is crucial to reasonably allocate and optimize computing resources. The computational load optimization unit monitors the computational load of each satellite in real time to ensure that the computational load of each satellite agent is balanced, thereby avoiding performance bottlenecks caused by excessive load on a certain satellite.

[0045] In some embodiments, load optimization is performed by: Load balancing: based on the calculation load of each satellite and mission requirements The computing module will dynamically adjust the task allocation to keep the computing load of each satellite balanced; load balancing between satellites can effectively avoid excessive energy consumption or task execution delays due to excessive computing of a satellite.

[0046] Dynamic scheduling: When the computing load of a satellite is too high, the computing module adjusts the task allocation decision. , transferring part of the tasks to satellite agents with lower computational load, thereby achieving load balancing and ensuring the smooth execution of the tasks.

[0047] For example, when multiple satellites work together, the computing module will adjust the distribution of tasks based on the mission load and remaining computing power of each satellite to ensure a reasonable distribution of computing load among satellites; if the computing load of a satellite exceeds its maximum carrying capacity, the system will automatically allocate part of the computing tasks to other satellites to avoid mission execution failure due to excessive load on a single satellite.

[0048] In this embodiment, the task scheduling unit is responsible for dynamically adjusting the execution order of tasks according to the computing power of the satellite, the priority of the task, and the energy status between satellites; the optimization of task scheduling can ensure that tasks are completed smoothly according to priority and avoid excessive concentration of resources on a certain satellite, resulting in waste of computing resources or computing bottlenecks.

[0049] The optimization strategies of the task scheduling unit include: Priority scheduling: When multiple tasks are to be processed at the same time, the task scheduling unit will arrange the order of task execution based on the urgency, importance and computing requirements of the tasks; high-priority tasks will be allocated computing resources first.

[0050] Time window scheduling: If there are time window constraints between certain tasks, the task scheduling unit will reasonably arrange the computing resource allocation of the tasks according to the time requirements of the tasks to ensure that the tasks are completed on time.

[0051] For example, the scientific missions performed by satellites may be time-sensitive and therefore need to be scheduled according to the time limit of the mission; for background computing tasks that have no time limit, the system can flexibly arrange computing resources to avoid affecting high-priority tasks.

[0052] The feedback mechanism of load balancing and task scheduling is the core part of the computing module. During actual operation, the computing load and task requirements of the satellite will change dynamically. Therefore, the computing module needs to adjust the allocation of computing resources in real time according to the feedback information during task execution. The feedback mechanism enables the satellite agent to adjust the computing load and resource allocation according to the actual operating conditions.

[0053] For example, when the execution of a task lags behind or there is a shortage of computing resources, the computing module will prioritize allocating more computing resources to the lagging task based on feedback information to ensure its completion on time; and when certain tasks progress smoothly, the computing module can avoid resource waste by reducing resource allocation.

[0054] As an option, an adaptive computing resource allocation mechanism can automatically optimize the computing resource allocation strategy through a machine learning algorithm based on the data of historical satellite mission execution. In this way, the computing module learns based on information such as the computing load and task execution time of each satellite in different tasks, and predicts the amount of computing resources required when a new task arrives. This method can effectively reduce human intervention and improve the accuracy of computing resource allocation.

[0055] For example, by analyzing the satellite's resource usage under different tasks, the computing module can predict the computing requirements of new tasks and allocate resources in advance; through this adaptive mechanism, the satellite agent can automatically adjust computing resources under different task loads and working environments to achieve higher computing efficiency.

[0056] When multiple satellites work together, the computing module also needs to consider the collaborative computing needs between satellites; in some embodiments, satellite agents exchange real-time mission status and computing load information through the communication module, and the computing module dynamically adjusts the task allocation between satellites according to the needs of collaborative computing; collaborative computing ensures that when the mission load is too heavy or the computing power is insufficient, the satellite agents can collaborate to complete the mission, avoiding mission delays or failures due to excessive load on a single satellite.

[0057] Task execution module, used to execute tasks and track task progress; In the artificial intelligence aerospace infrastructure intelligent body system of the present invention, the task execution module is one of the core components of each satellite intelligent body, responsible for receiving, allocating and scheduling tasks to ensure efficient execution of tasks; the task execution module needs to consider factors such as the satellite's computing resources, energy status, and communication bandwidth when executing tasks to ensure that the execution of tasks is both efficient and does not cause waste of resources or system overload; the task execution module works closely with the energy management module, the computing module, and the communication module to achieve optimal scheduling of system resources and smooth completion of tasks.

[0058] In this embodiment, the task execution module is mainly composed of two sub-modules: a task allocation unit and a task scheduling unit; the task allocation unit is responsible for allocating tasks according to the real-time status of the satellite intelligent body, while the task scheduling unit is responsible for dynamically adjusting the order of task execution according to task priority and resource constraints during task execution.

[0059] The core task of the task allocation unit is to reasonably allocate tasks according to the real-time status of each satellite intelligent body, such as energy, computing load and communication bandwidth; task allocation not only considers the computing requirements of the task, but also the energy status and computing load of the satellite to avoid excessive consumption or overload of resources.

[0060] In this embodiment, the decision of the task allocation unit is made by the following formula: ; in: Indicates satellite At the moment Mission progress, indicating the progress of the mission the satellite is performing; Indicates satellite At the moment The task allocation decision represents the amount of tasks assigned to the satellite; Indicates satellite At the moment The energy status of the satellite determines the energy resources available to it; For satellite At the moment The computing load determines the amount of tasks the satellite can handle; Indicates satellite At the moment The communication bandwidth status indicates the satellite's communication resource usage; It is a task allocation function, which comprehensively considers factors such as satellite energy, computing load, communication bandwidth, etc. to determine the task allocation strategy.

[0061] The formula shows that the amount of tasks allocated Will affect the progress of the task , and task allocation is affected by the energy state , calculate the load and communication bandwidth The task allocation unit will dynamically adjust the task allocation based on this information to ensure that the satellite will not be overloaded due to too many tasks, and to avoid delays in task execution when resources are sufficient.

[0062] In some embodiments, the task allocation unit will also dynamically adjust the task allocation strategy according to the priority of the task; if some tasks have high priority, the task allocation unit will prioritize allocating more computing resources and energy to these tasks; in this way, the task allocation unit can ensure that critical tasks can be completed first under the resource constraints of the satellite.

[0063] The task scheduling unit is mainly responsible for adjusting the execution order of tasks according to task priority, task execution progress and the real-time resource status of the satellite (such as computing resources, energy, etc.) during the task execution process; task scheduling is not only to adjust the task order according to the task priority, but also needs to consider the dependencies between tasks, the constraints of system resources and the collaborative work of multiple satellites.

[0064] In this embodiment, the task scheduling unit determines the order of task execution through the following strategies: Priority scheduling: The task scheduling unit will first arrange the execution of tasks according to the priority of each task; the priority of a task is usually determined by the urgency or importance of the task; high-priority tasks will be executed first.

[0065] Task dependency scheduling: For tasks with strong dependencies (such as some tasks need to rely on the output of other tasks), the task scheduling unit will schedule them according to the order of tasks to ensure that the dependencies between tasks are met.

[0066] Dynamic Scheduling: During the mission execution process, the task scheduling unit will dynamically adjust the order of task execution based on the satellite's resource usage (such as energy, computing load, and communication bandwidth); if a task is executing slowly, the scheduling unit will prioritize allocating more resources to the task to speed up its completion.

[0067] For example, if a satellite agent The satellite agent is performing a high priority task and is progressing slowly. The task scheduling unit may schedule some tasks to the satellite agent. , to speed up the completion of high priority tasks.

[0068] One of the core elements of the mission execution module is its collaborative optimization with other modules (such as energy management module, computing module, and communication module); the mission execution module will adjust the task allocation and scheduling strategies in real time according to the feedback information during the mission execution process; for example, if the energy status of the satellite changes during the mission execution process, the mission execution module will re-evaluate the execution progress of the current task and adjust the task allocation and execution order according to the new energy situation.

[0069] The feedback mechanism enables the mission execution module to optimize task allocation and scheduling strategies according to the actual progress of the mission, ensuring that the mission can be completed at the most appropriate time; if the satellite's energy is insufficient to support high-load mission execution, the mission execution module will reduce the computational load of the task based on feedback information and transfer part of the task to other satellites with more sufficient energy.

[0070] As an option, the mission execution module can be adaptively optimized through intelligent algorithms such as machine learning. By learning from historical mission execution data, the mission execution module can predict the computing requirements, execution time, and possible delays of different types of tasks, so as to make task allocation and scheduling plans in advance. Intelligent algorithms can dynamically adjust task allocation strategies according to the characteristics of the mission and the satellite's historical execution data, thereby improving the efficiency and accuracy of mission execution.

[0071] For example, satellite agents can analyze the execution data of historical tasks, learn the resource requirements of different types of tasks, and predict the computing load and execution progress of tasks based on the real-time status of task execution, thereby achieving accurate scheduling and allocation of resources.

[0072] When multiple satellites work together, the mission execution module not only considers the mission execution of a single satellite, but also the collaboration between satellites; for example, satellites may share computing resources or communication bandwidth, or some tasks may require multiple satellites to collaborate to complete; the mission execution module will dynamically adjust the priority of task execution and resource allocation method according to the collaboration needs between satellites and the characteristics of the mission.

[0073] In some embodiments, the task execution module adjusts the task coordination mode according to the communication status and computing load between satellites; for example, if the computing load of a satellite is high, the task execution module can assign part of the computing tasks to other satellites to achieve collaborative computing, thereby improving the efficiency of the entire system.

[0074] A communication module, used for exchanging information between satellite agents; wherein a plurality of satellite agents are interconnected and collaborated through the communication module; In the artificial intelligence aerospace infrastructure intelligent agent system of the present invention, the communication module is an important component of each satellite intelligent agent, and is responsible for real-time data transmission and information exchange between satellite intelligent agents; since satellite intelligent agents need to frequently share information such as task execution progress, energy status, computing load, etc., the design of the communication module is crucial to the efficient operation of the entire system; the main function of the communication module is to dynamically adjust the data transmission strategy according to the channel quality, task requirements and resource status of the satellite to ensure stable and efficient transmission of information.

[0075] The communication module works closely with other modules (such as the energy management module, computing module and task execution module) to ensure smooth information flow, thereby achieving optimal scheduling of tasks and resources by satellite intelligent bodies in collaborative work.

[0076] The channel selection unit is a key component in the communication module and is responsible for selecting the most appropriate communication path. In a multi-satellite intelligent system, the role of the channel selection unit is to select the optimal communication path based on parameters such as channel quality, transmission power, and distance between satellites to maximize data transmission efficiency. In this embodiment, channel selection is optimized by the following formula: ; in: Indicates satellite and satellite The channel capacity between satellites, that is, the amount of data that can be transmitted between satellites, reflects the transmission efficiency of the channel; is the transmission power of satellite c, indicating that Data transmission power for communication; For satellite With satellite The channel gain between them is usually determined by factors such as the distance between satellites and the quality of the communication link; is the noise power spectral density, which represents the background noise in the channel and affects the quality of the communication channel.

[0077] The channel selection unit evaluates the communication quality of each path by calculating the channel capacity of each communication path; the larger the channel capacity, the better the channel quality and the higher the data transmission rate; the channel selection unit will select the optimal path according to the real-time needs during task execution to ensure the stability and efficiency of information transmission.

[0078] Generally speaking, when the satellite's computing load is high or the mission is urgent, the channel selection unit will give priority to communication paths with higher bandwidth to ensure that mission data can be transmitted in a timely manner; when the load is low, the channel selection unit can select lower-power paths to save energy and resources.

[0079] The data transmission unit is responsible for implementing data transmission between satellites based on the decisions of the channel selection unit; the main task of the data transmission unit is to dynamically adjust the data transmission rate according to the channel capacity and the resource status of the satellite (such as computing load and energy conditions) to ensure smooth and timely transmission of information.

[0080] In this embodiment, the change of the data transmission rate is controlled by the following formula: ; in: Indicates satellite and satellite The change of data transmission rate between reflects the real-time transmission scheduling of data; A function representing the change in data transmission rate; For satellite and satellite The channel capacity between represents the transmission efficiency of the channel; For satellite The remaining computing power of the satellite determines Data processing capabilities available for communications; For satellite mission progress, indicating the satellite The progress of the task currently being processed.

[0081] The formula shows that the data transmission rate Affected by channel capacity, satellite computing power and mission progress; when the satellite's computing load is high or the mission progress is slow, the data transmission rate will be reduced accordingly to reduce the consumption of computing and communication bandwidth; when the satellite's computing load is light or the mission progress is fast, the data transmission rate can be appropriately increased to ensure that data can be transmitted in time.

[0082] As an option, the data transmission unit can also adjust the transmission strategy according to the real-time energy status of the satellite; if the energy of a satellite is low, the data transmission rate will be appropriately reduced to avoid excessive energy consumption; when the energy is sufficient, the system can increase the data transmission rate to speed up the transmission of information.

[0083] Adaptive communication strategy is one of the important features of the communication module. Through the real-time feedback mechanism, the communication module can adjust the communication strategy according to the actual operation of the satellite. The feedback mechanism includes two main parts: on the one hand, the communication module will adjust the communication path and transmission rate according to information such as the satellite's computing load, energy status and mission progress; on the other hand, the communication module will also adjust the communication strategy according to problems such as delays and communication failures in mission execution to ensure smooth transmission of information.

[0084] Specifically, the feedback mechanism can be adaptively adjusted according to the following aspects: Computational load feedback: When the computational load of the satellite increases, the communication module can choose a path with lower bandwidth and higher latency, thereby reducing the consumption of computing resources.

[0085] Energy status feedback: If the satellite’s energy is insufficient to support high-bandwidth data transmission, the communication module will dynamically select a lower-power communication path or adjust the data transmission rate to save energy.

[0086] Task progress feedback: If the execution progress of some tasks lags behind, the communication module will prioritize allocating more bandwidth and communication resources to these tasks to ensure that the tasks can be completed on time.

[0087] In one possible implementation, the communication module will make adjustments based on the real-time transmission quality, the noise level of the channel, and the need for collaboration between satellites; for example, if the channel quality is poor, the data transmission unit will automatically select a more stable relay path or adjust the transmission rate to avoid data transmission failure due to poor channel quality.

[0088] When multiple satellites work together, the communication module is not only responsible for communication between individual satellites, but also needs to handle the collaborative work between satellites; satellite intelligent bodies optimize resource sharing through collaborative communication to ensure efficient cooperation between multiple satellites.

[0089] In some embodiments, the communication module can be adaptively adjusted based on the cooperation requirements between satellites; for example, the satellite and satellite When performing the same mission, the communication modules dynamically adjust the data flow between them to avoid repeated use of communication paths and excessive concentration of resources; this collaborative communication strategy enables the system to utilize communication resources between satellites more efficiently.

[0090] When multiple satellite agents work together, the communication module is also responsible for ensuring the smooth flow of information; when collaborating between satellites, the system will optimize the path and timing of information transmission based on the communication capabilities between satellites and the urgency of task execution, thereby achieving efficient execution of the task.

[0091] See also Figure 6 ,The artificial intelligence aerospace infrastructure agent method includes the following steps: S1. Obtain the real-time status of multiple satellite agents, including energy status, computing load and task progress; Before executing a mission, the system first obtains the real-time status information of each satellite agent; specifically, the energy consumption of the satellite , calculate the load and task progress It is a key parameter for evaluating the current operating capability of the satellite; the energy status reflects the remaining energy of the satellite, the computational load represents the processing capacity of the satellite, and the mission progress shows the degree of completion of the satellite in mission execution; real-time acquisition of this information is helpful for subsequent task scheduling and resource optimization decisions, so that the task allocation of each satellite can be dynamically adjusted according to the current resource status.

[0092] S2. According to the real-time status of the satellite intelligent body, use the optimal control theory to perform task scheduling and resource allocation to solve the optimal control strategy; Based on the real-time status information of the satellite, the system uses optimal control theory for resource allocation and task scheduling; through the optimal control algorithm, the power allocation, computing resource scheduling and task allocation strategy of each satellite during the mission execution can be determined; the goal of the optimal control strategy is to maximize task completion, minimize resource consumption, and ensure that the task is completed on time; in actual operation, the system will dynamically calculate and adjust the task allocation strategy according to the energy status, computing load and task progress of each satellite to ensure the optimal utilization of resources.

[0093] S3, based on the game theory model, optimize the task allocation strategy through collaboration and game among satellite agents; Task allocation among satellite agents depends not only on the internal resource situation of each satellite, but also on the strategies of other satellites. In the framework of game theory, each satellite agent makes decisions based on its payoff function and the strategies of other satellites. Game models help satellite agents optimize their task allocation and coordinate task scheduling under mutual influence to ensure optimal global task completion. The equilibrium solution in the game will provide satellite agents with the best task allocation plan, thereby improving the resource utilization efficiency of the entire system.

[0094] S4, optimize the communication path between satellite agents according to the channel capacity theorem and select the optimal communication strategy; When data is transmitted between satellites, the system needs to select the optimal communication path based on factors such as channel quality, distance between satellites, and bandwidth requirements. According to the channel capacity theorem, the communication module selects the best path by evaluating the capacity of each channel. This step involves consideration of variables such as signal power, channel gain, and noise to ensure the speed and stability of data transmission between satellites. The optimization of communication strategies not only improves data transmission efficiency, but also effectively reduces latency, especially in the collaborative process of multiple tasks and large amounts of data.

[0095] S5. During the task execution process, adjust the task execution order and resource allocation strategy in real time; During the task execution process, the system needs to dynamically adjust the task execution order and resource allocation strategy based on real-time feedback information; this is to cope with possible changes in task execution, such as changes in satellite energy consumption, delayed task progress, communication problems, etc.; by real-time monitoring of the task execution status, the system will adjust the task allocation strategy and resource scheduling according to the task priority and resource changes; this ensures that even if the system status changes, the task can still be completed efficiently and smoothly, avoiding system performance degradation due to resource bottlenecks or task delays.

[0096] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. Artificial intelligence space infrastructure agent system, characterized by: The invention comprises a satellite intelligent agent, wherein the satellite intelligent agent comprises: Energy management module, used to manage the energy status of satellite intelligent body; A computing module for managing the computing load of satellite agents; Task execution module, used to execute tasks and track task progress; Communication module, used for information exchange between satellite agents; Wherein, a plurality of satellite intelligent bodies are interconnected and cooperated via a communication module.

2. The artificial intelligence aerospace infrastructure agent system according to claim 1, characterized in that: The energy management module comprises: an energy consumption control unit for executing and calculating loads according to the missions of the satellite agent; The energy recovery unit is used to recover energy during task execution.

3. The artificial intelligence aerospace infrastructure agent system according to claim 1, characterized in that: The calculation module comprises: A computing resource allocation unit is used to allocate resources according to task requirements and computing load; The computational load optimization unit is used to optimize the computational load of the satellite agent during mission execution.

4. The artificial intelligence aerospace infrastructure agent system according to claim 1, characterized in that: The task execution module includes: A task allocation unit, used to dynamically allocate tasks according to the energy status, computing load and communication bandwidth of the satellite agent; The task scheduling unit is used to adjust the task execution order according to the real-time status and task priority of the satellite intelligent body.

5. The artificial intelligence aerospace infrastructure agent system according to claim 1, characterized in that: The communication module comprises: A channel selection unit, used to dynamically select the optimal communication path between satellite agents according to the channel quality; The data transmission unit is used to realize data transmission through the selected communication path and monitor the data transmission quality.

6. The artificial intelligence aerospace infrastructure agent system according to claim 1, characterized in that: The energy consumption of the satellite agent , calculate the load and task progress The following dynamic equations are satisfied: Energy consumption equation: ;in: Satellite Agent At the moment The rate of change of energy; Satellite Agent At the moment energy state; Satellite Agent At the moment Power consumption; Satellite Agent Calculating Load The corresponding energy conversion efficiency coefficient; Satellite Agent In executing task assignment Additional energy consumption coefficient when Satellite Agent At the moment task allocation decisions; Calculate the load equation: ;in: Satellite Agent At the moment The rate of change of the calculated load; Satellite Agent At the moment The computational load; Satellite Agent At the moment task allocation decisions; is a function of the computational load, depending on the task allocation decision and the current load; Task progress equation: in: Satellite Agent At the moment The rate of change of task progress; Satellite Agent At the moment The progress of the task; Satellite Agent At the moment The computational load; Satellite Agent The task processing efficiency indicates the impact of computing load on task progress.

7. The artificial intelligence space infrastructure agent system according to claim 6, characterized in that: The optimal control strategy of the satellite agent By solving the following optimal control problem: ;in, are weight coefficients, respectively controlling energy consumption , calculate the load and task progress Impact on the objective function; Satellite Agent At the moment energy state; Satellite Agent At the moment The computational load; Satellite Agent At the moment The progress of the task; Satellite Agent The control strategy includes power distribution , Task Allocation and communication strategies .

8. The artificial intelligence space infrastructure agent system according to claim 1, characterized in that: The task collaboration between the satellite agents is optimized through a game theory model, and the equilibrium solution of the game satisfies the following conditions: ;in, Satellite Agent The optimal profit function is Satellite Agent The profit function, Satellite Agent The task allocation strategy Satellite Agent Communication strategies, Represents the strategies of other satellite agents.

9. The artificial intelligence space infrastructure agent system according to claim 1, characterized in that: The communication between the satellite agents is optimized by the following channel capacity theorem: ;in, Satellite Agent Satellite Agent The communication rate between Satellite Agent The transmission power, Satellite Agent Satellite Agent The channel gain between is the noise power spectral density.

10. An artificial intelligence space infrastructure agent method, applied to an artificial intelligence space infrastructure agent system according to any one of claims 1 to 9, characterized in that: The following steps are involved: S1. Obtain the real-time status of multiple satellite agents, including energy status, computing load and task progress; S2. According to the real-time status of the satellite intelligent body, use the optimal control theory to perform task scheduling and resource allocation to solve the optimal control strategy; S3, based on the game theory model, optimize the task allocation strategy through collaboration and game among satellite agents; S4, optimize the communication path between satellite agents according to the channel capacity theorem and select the optimal communication strategy; S5. During the task execution process, adjust the task execution order and resource allocation strategy in real time.

Images