A satellite mission planning method, device, equipment and medium

Abstract

The present invention discloses a satellite mission planning method, device, equipment and medium, which relates to the field of artificial intelligence technology, in order to solve the problem of low reliability of satellite mission planning in the prior art. The method comprises: performing feature mining on a plurality of multi-source data by using a target satellite mission planning network model, obtaining multi-satellite description features corresponding to multi-satellite description data, multi-ground base station description features corresponding to multi-ground base stations, and multi-user task description features corresponding to multi-user task description data; based on the multi-satellite description features, the multi-ground base station description features and the multi-user task description features, performing associated feature aggregation on the target task description data, and obtaining global task aggregation features corresponding to the target task description data; and performing task planning processing on the global task aggregation features corresponding to the target task description data, and outputting the target satellite mission planning data; thereby improving the reliability of satellite mission planning.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a satellite mission planning method, device, equipment and medium. Background Art

[0002] With the continuous advancement of satellite technology, the planning problem of multi-satellite network collaborative observation missions has become increasingly complex. Traditional mission planning methods often have problems such as low planning efficiency and insufficient optimization when facing large-scale satellite networks. In the existing technology, tasks are generally simply allocated to ground base stations and satellites based on task priorities, which makes the reliability of allocation relatively low.

[0003] In view of this, there is an urgent need to design a more effective satellite mission planning method to solve the problem of low reliability of satellite mission planning in the existing technology. Summary of the invention

[0004] The purpose of the present invention is to provide a satellite mission planning method, device, equipment and medium, which utilizes the feature processing model included in the target satellite mission planning network to perform at least feature mining and feature aggregation processing on multiple multi-source data related to satellite mission planning to obtain global task aggregation features corresponding to target task description data, and further utilizes the task planning model included in the target satellite mission planning network to perform task planning processing on the global task aggregation features and output target satellite mission planning data; thereby improving the reliability of satellite mission planning and solving the problem of relatively low reliability of satellite mission planning in the prior art.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] In a first aspect, the present invention provides a satellite mission planning method, which may include:

[0007] Acquire multiple multi-source data related to satellite mission planning; the multi-source data at least includes multiple satellite description data corresponding to the satellite, multiple ground base station description data corresponding to the ground base station, and multiple user task description data corresponding to the user task;

[0008] Performing feature mining on the plurality of multi-source data using a target satellite mission planning network model to obtain multi-satellite description features corresponding to the multi-satellite description data, multi-ground base station description features corresponding to the multi-ground base stations, and multi-user task description features corresponding to the multi-user task description data;

[0009] Based on the multi-satellite description features, the multi-ground base station description features and the multi-user task description features, the target task description data is aggregated with associated features to obtain a global task aggregation feature corresponding to the target task description data;

[0010] Performing task planning processing on the global task aggregation features corresponding to the target task description data, and outputting target satellite task planning data.

[0011] Preferably, the performing associated feature aggregation on the target task description data based on the multi-satellite description feature, the multi-ground base station description feature and the multi-user task description feature to obtain the global task aggregation feature corresponding to the target task description data may include:

[0012] For the target task description data, the first feature aggregation unit in the feature processing model included in the target satellite mission planning network model is used to associate and aggregate the task description features corresponding to the target task description data in combination with the task description features corresponding to the non-target task description data, and the target task aggregation features corresponding to the target task description data are output; wherein the semantic information of the target task aggregation features includes the semantic information of the target task description features and the semantic information of the non-target task description features;

[0013] For the target task description data, the second feature aggregation unit in the feature processing model included in the target satellite task planning network model is used to combine the multi-satellite description features and the multi-ground base station description features to associate and aggregate the target task aggregation features corresponding to the target task description data, and output the global task aggregation features corresponding to the target task description data.

[0014] Preferably, the step of utilizing the first feature aggregation unit in the feature processing model included in the target satellite mission planning network model, combining the task description features corresponding to the non-target task description data, associating and aggregating the task description features corresponding to the target task description data, and outputting the target task aggregation features corresponding to the target task description data may include:

[0015] Using the first association aggregation branch included in the first feature aggregation unit, in combination with the task description features corresponding to the non-target task description data, the task description features corresponding to the target task description data are associated and aggregated, and the first task aggregation features corresponding to the target task description data are output;

[0016] Utilizing the second association aggregation branch included in the first feature aggregation unit, in combination with the task description features corresponding to the non-target task description data, the task description features corresponding to the target task description data are associated and aggregated, and the second task aggregation features corresponding to the target task description data are output; wherein the association aggregation method corresponding to the first association aggregation branch is different from the association aggregation method corresponding to the second association aggregation branch;

[0017] The first task aggregation feature and the second task aggregation feature are merged by using the branch feature merging subunit included in the first feature aggregation unit, and the target task aggregation feature corresponding to the target task description data is output.

[0018] Preferably, the utilizing the first associative aggregation branch included in the first feature aggregation unit, combining the task description features corresponding to the non-target task description data, associating and aggregating the task description features corresponding to the target task description data, and outputting the first task aggregation features corresponding to the target task description data may include:

[0019] The target task description data and the non-target task description data are loaded by using the first association aggregation branch, and a task description feature corresponding to each non-target task description data is centrally determined in the first association aggregation branch to output a central task description feature corresponding to each non-target task description data;

[0020] Multiplying the transposed result of the central task description feature corresponding to each non-target task description data and the task description feature corresponding to the target task description data, and outputting a parameter representing the correlation relationship between the target task description feature and the central task description feature;

[0021] Based on the correlation relationship representation parameter between the target task description feature and the central task description feature, feature adjustment is performed on the task description feature corresponding to the target task description data, and a first task aggregation feature corresponding to the target task description data is output.

[0022] Preferably, the utilizing the second associative aggregation branch included in the first feature aggregation unit, combining the task description features corresponding to the non-target task description data, associating and aggregating the task description features corresponding to the target task description data, and outputting the second task aggregation features corresponding to the target task description data may include:

[0023] For the target task description data, the target task description data and the non-target task description data are loaded by using the second associated aggregation branch, and in the second associated aggregation branch, the target distance between the task description feature corresponding to each non-target task description data and the center is determined respectively; and based on the target distance, the task description feature corresponding to each non-target task description data is serialized to form a task description feature sequence corresponding to the non-target task description data;

[0024] For a first non-target task description feature in the task description feature sequence, self-attention processing is performed on the first non-target task description feature, and a task description attention feature corresponding to the first non-target task description feature is output; and based on a correlation relationship representation parameter between the task description attention feature corresponding to the first non-target task description feature and the transposed result of the task description feature corresponding to the target task description data, feature adjustment is performed on the task description feature corresponding to the target task description data, and a task description adjustment feature corresponding to the first task description feature is output;

[0025] For the Nth task description feature in the task description feature sequence, perform self-attention processing on the Nth task description feature, and output the task description attention feature corresponding to the Nth task description feature, where N is greater than 1 and less than the number corresponding to the non-target task description data; based on the correlation relationship representation parameter between the Nth task description attention feature and the transposed result of the task description adjustment feature corresponding to the N+1th task description feature, perform feature adjustment on the task description adjustment feature corresponding to the N+1th task description feature, and output the task description adjustment feature corresponding to the non-target task description feature;

[0026] A second task aggregation feature corresponding to the target task description data is determined based on at least the task description adjustment feature corresponding to the last task description sub-feature in the task description feature sequence.

[0027] Preferably, utilizing the second feature aggregation unit in the feature processing model included in the target satellite mission planning network model, combining the multi-satellite description features and the multi-ground base station description features, associating and aggregating the target task aggregation features corresponding to the target task description data, and outputting the global task aggregation features corresponding to the target task description data, may include:

[0028] Using the first feature aggregation subunit in the second feature aggregation unit, multiply the multi-satellite description feature by a transposed result of the target task aggregation feature corresponding to the target task description data, and output a first association relationship representation parameter corresponding to the target task description data;

[0029] Using the second feature aggregation subunit in the second feature aggregation unit, multiply the multiple ground base station description features by the transposed result of the target task aggregation features corresponding to the target task description data, and output a second association relationship representation parameter corresponding to the target task description data;

[0030] And using a third feature aggregation subunit in the second feature aggregation unit, multiplying the mean feature of the multi-satellite description feature and the multi-ground base station description feature by the transposed result of the target task aggregation feature corresponding to the target task description data, and outputting a third association relationship representation parameter corresponding to the target task description data;

[0031] The first association relationship representation parameter, the second association relationship representation parameter and the third association relationship representation parameter are fused to output the fused association relationship representation parameter corresponding to the target task description data;

[0032] Based on the fusion association relationship representation parameter, feature adjustment is performed on the target task aggregation feature corresponding to the target task description data, and the global task aggregation feature corresponding to the target task description data is output.

[0033] Preferably, the method of performing feature mining on the plurality of multi-source data using the target satellite mission planning network model may include:

[0034] Establishing the target satellite mission planning network model;

[0035] The establishing of the target satellite mission planning network model may include:

[0036] Acquire sample multi-satellite description data corresponding to a plurality of satellites, sample multi-ground base station description data corresponding to a plurality of ground base stations, and a plurality of sample task description data corresponding to a plurality of sample user tasks;

[0037] Using a feature processing model included in the candidate satellite mission planning network, feature mining and feature aggregation are performed on the sample multi-satellite description data, the sample multi-ground base station description data, and each of the sample task description data, and a sample global task aggregation feature corresponding to each of the sample task description data is output, wherein the semantic information of each of the sample global task aggregation features includes the semantic information of the corresponding sample task description data, the semantic information of other sample task description data, the semantic information of the sample multi-satellite description data, and the semantic information of the sample multi-ground base station description data;

[0038] Using the task planning model included in the candidate satellite task planning network, the sample global task aggregation feature corresponding to each of the sample task description data is processed to output sample satellite task planning data, wherein the sample satellite task planning data is used to reflect the allocation of each satellite and each ground base station to the plurality of sample user tasks;

[0039] Based on the difference between the sample satellite mission planning data and the actual satellite mission planning data corresponding to the sample satellite mission planning data, the network parameters of the candidate satellite mission planning network model are optimized to obtain the target satellite mission planning network model.

[0040] In a second aspect, the present invention provides a satellite mission planning device, which may include:

[0041] An acquisition module, the acquisition module is used to acquire multiple multi-source data related to satellite mission planning; the multi-source data at least includes multiple satellite description data corresponding to the satellite, multiple ground base station description data corresponding to the ground base station, and multiple user task description data corresponding to the user task;

[0042] A satellite mission feature mining module, wherein the satellite mission feature mining module is used to perform feature mining on the plurality of multi-source data using a target satellite mission planning network model to obtain multi-satellite description features corresponding to the multi-satellite description data, multi-ground base station description features corresponding to the multi-ground base stations, and multi-user task description features corresponding to the multi-user task description data;

[0043] A satellite mission feature aggregation module, wherein the satellite mission feature aggregation module is used to perform associated feature aggregation on the target mission description data based on the multi-satellite description features, the multi-ground base station description features and the multi-user mission description features to obtain a global mission aggregation feature corresponding to the target mission description data;

[0044] A satellite mission planning module is used to perform mission planning processing on the global mission aggregation features corresponding to the target mission description data and output target satellite mission planning data.

[0045] In a third aspect, the present invention provides a satellite mission planning device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the satellite mission planning method described in the first aspect when executing the computer program.

[0046] In a fourth aspect, the present invention provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed, the satellite mission planning method described in the first aspect is implemented.

[0047] Compared with the prior art, the present invention provides a satellite mission planning method, which obtains multiple multi-source data related to satellite mission planning; the multi-source data at least includes multi-satellite description data corresponding to satellites, multi-ground base station description data corresponding to ground base stations, and multi-user task description data corresponding to user tasks; uses a target satellite mission planning network model to perform feature mining on the multiple multi-source data to obtain multi-satellite description features corresponding to the multi-satellite description data, multi-ground base station description features corresponding to multiple ground base stations, and multi-user task description features corresponding to multi-user task description data; based on the multi-satellite description features, the multi-ground base station description features, and the multi-user task description features, the target task description data is aggregated with associated features to obtain global task aggregation features corresponding to the target task description data; The global task aggregation features corresponding to the target task description data are used for task planning processing, and target satellite task planning data is output; based on this, since feature mining and feature aggregation are performed separately for each user task when analyzing and mining multiple user tasks, the granularity of feature processing is finer and the accuracy is higher, and in the process of separate processing, the features of other user tasks, the features of the satellite, and the features of the ground base station are also aggregated, so that the semantic representation ability of the obtained global task aggregation features is higher; further, when task planning is performed based on the global task aggregation features, the reliability of the obtained target satellite task planning data is higher, thereby solving the problem of relatively low reliability of satellite task planning in the prior art, and can ensure the reliable operation of multiple satellites, multiple ground base stations and multiple user tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0049] Figure 1 A schematic diagram of the main flow of a satellite mission planning method provided by the present invention;

[0050] Figure 2 A schematic diagram of feature processing of a second associated aggregation branch in a satellite mission planning method provided by the present invention;

[0051] Figure 3 A schematic diagram of the structure of a satellite mission planning device provided by the present invention;

[0052] Figure 4 A schematic diagram of the hardware structure of a satellite mission planning device provided by the present invention. DETAILED DESCRIPTION

[0053] In order to clearly describe the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, words such as "first" and "second" are used to distinguish the same items or similar items with basically the same functions and effects. For example, the first threshold and the second threshold are only used to distinguish different thresholds, and their order of precedence is not limited. Those skilled in the art can understand that words such as "first" and "second" do not limit the quantity and execution order, and words such as "first" and "second" do not necessarily limit them to be different.

[0054] It should be noted that, in the present invention, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0055] In the present invention, "at least one" means one or more, and "plurality" means two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist, for example: A and / or B, which can mean: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The character " / " generally indicates that the associated objects in the previous time are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example: at least one of a, b or c, can mean: a, b, c, the combination of a and b, the combination of a and c, the combination of b and c, or the combination of a, b and c, where a, b, c can be single or multiple.

[0056] At present, when facing large-scale satellite networking, there are problems such as low planning efficiency and insufficient optimization. The commonly used method in the existing technology is to allocate tasks to ground base stations and satellites based on the priority of the tasks. Practical applications show that this allocation method is not reliable for satellite mission planning.

[0057] In view of this, the present invention provides a satellite mission planning method, device, equipment and medium. On the basis of artificial intelligence, a feature processing model and a task planning model included in a target satellite mission planning network are used to perform at least feature mining and feature aggregation processing on multiple multi-source data obtained related to satellite mission planning to output global task aggregation features, and then output target satellite mission planning data based on the global task aggregation features; thereby improving the reliability of satellite mission planning and solving the problem of relatively low reliability of satellite mission planning in the prior art.

[0058] Next, the technical solution of the present invention is described in detail with reference to the accompanying drawings:

[0059] In a first aspect, the present invention provides a satellite mission planning method, see Figure 1 , Figure 1 A schematic diagram of the main flow of a satellite mission planning method provided by the present invention; the executor of the present invention is a service platform or electronic device equipped with the method provided by the present invention, such as a platform server or a handheld device.

[0060] exist Figure 1 In, the method may include:

[0061] Step 110: Acquire multiple multi-source data related to satellite mission planning; the multi-source data at least includes multiple satellite description data corresponding to the satellite, multiple ground base station description data corresponding to the ground base station, and multiple user task description data corresponding to the user task.

[0062] In step 110, multiple multi-source data including multiple satellite description data corresponding to multiple satellites, multiple ground base station description data corresponding to multiple ground base stations, and multiple task description data corresponding to multiple user tasks may be obtained. Each of the task description data at least includes the data size, bandwidth requirement, and task priority of the user task; for example, each of the task description data at least reflects the data size of the user task (for example, a high-resolution image of a satellite may require tens of GB of data, while a low-resolution image or sensor data may only require a few MB to GB of data), bandwidth requirement (for example, live video, high-definition images, or real-time monitoring tasks require a higher bandwidth to ensure smooth data flow; while low-speed meteorological data or remote control signals may not require too high a bandwidth) and task priority (for example, military tasks, disaster monitoring, and other tasks may have a higher priority, while some scientific research tasks with low timeliness may be ranked behind).

[0063] Step 120: Use the target satellite mission planning network model to perform feature mining on the multiple multi-source data to obtain multi-satellite description features corresponding to the multi-satellite description data, multi-ground base station description features corresponding to the multiple ground base stations, and multi-user task description features corresponding to the multi-user task description data.

[0064] Step 130: Based on the multi-satellite description features, the multi-ground base station description features and the multi-user task description features, the target task description data is aggregated with associated features to obtain a global task aggregation feature corresponding to the target task description data.

[0065] In step 120 to step 130, after obtaining various description data, the feature processing model included in the target satellite mission planning network can be used to perform feature mining and feature aggregation on the multi-satellite description data, the multi-ground base station description data and each of the task description data, and output the global task aggregation feature corresponding to each of the task description data. Wherein, the target satellite mission planning network belongs to a neural network model formed by network optimization in advance, and the semantic information of the global task aggregation feature corresponding to the target task description data includes the semantic information of the target task description data, the semantic information of the non-target task description data, the semantic information of the multi-satellite description data and the semantic information of the multi-ground base station description data; that is, the semantic information of each global task aggregation feature includes the semantic information of the task description data corresponding to the task, the semantic information of other task description data, the semantic information of the multi-satellite description data and the semantic information of the multi-ground base station description data, so that the global task aggregation feature can be guaranteed to have richer semantic information. In addition, in the embodiment of the present application, the expression form of each feature can be a vector or a matrix.

[0066] It should be noted that the target task description data is any one of the multiple task description data corresponding to the multiple user tasks. The satellite task planning method provided by the present invention needs to perform data processing on each of the multiple task description data.

[0067] Step 140: Performing task planning processing on the global task aggregation features corresponding to the target task description data, and outputting target satellite task planning data.

[0068] In step 130, based on the global task aggregation features output in step 120, the task planning model included in the target satellite task planning network can be used to process the global task aggregation features corresponding to each task description data, and output the target satellite task planning data. Among them, the target satellite task planning data is used to reflect the allocation of each satellite and each ground base station to multiple user tasks. For example: the task planning model may include a fully connected unit (used to fully connect the global task aggregation features) and an output unit, and the output unit may include a classification function (such as softmax function) for classifying and outputting the fully connected features output by the fully connected unit, and obtaining the probability of each satellite being assigned to each user task and the probability of each ground base station being assigned to each user task. In this way, the target satellite task planning data can be determined. It should be noted that for each user task, based on the fully connected features corresponding to the user task, the probability of each satellite being assigned to the user task and the probability of each ground base station being assigned to the user task can be output.

[0069] Based on this, the present invention provides a satellite mission planning method, which, when analyzing and mining multiple user tasks, performs feature mining and feature aggregation for each user task separately, so that the granularity of feature processing is finer and the accuracy is higher, and, in the process of separate processing, the features of other user tasks, the features of the satellite, and the features of the ground base station are aggregated, so that the semantic representation ability of the obtained global task aggregation feature is higher. Therefore, when performing task planning based on the global task aggregation feature, its reliability can be guaranteed, so that the reliability of the obtained target satellite mission planning data is higher, thereby improving the problem of relatively low reliability of satellite mission planning in the prior art, thereby ensuring the reliable operation of multiple satellites, multiple ground base stations and multiple user tasks.

[0070] In step 120 to step 130, the specific method of performing feature mining and feature aggregation on the multi-satellite description data, the multi-ground base station description data and each mission description data is not limited and can be selected according to actual needs.

[0071] Preferably, in step 120, the feature space mapping unit in the feature processing model can be used to perform feature processing on the multi-satellite description data, the multi-ground base station description data and the target task description data, and output the multi-satellite description features corresponding to the multi-satellite description data, the multi-ground base station description features corresponding to the multi-ground base stations and the task description features corresponding to the target task description data, respectively; wherein the multi-satellite description data is at least used to reflect the orbit information, visibility information and health status information of each satellite, and the multi-ground base station description data is at least used to reflect the bandwidth information of the communication link between each ground base station and the satellite, and the communication window for effective communication with the satellite.

[0072] Specifically, the feature space mapping unit in the feature processing model included in the target satellite mission planning network can be used to perform feature processing on the multi-satellite description data and output the multi-satellite description features corresponding to the multi-satellite description data; and perform feature processing on the multi-ground base station description data and output the multi-ground base station description features corresponding to the multi-ground base stations, and perform feature processing on each target task description data and output the task description features corresponding to each target task description data. Wherein, the multi-satellite description data is at least used to reflect the orbit information, visibility information and health status information of each satellite, and the multi-ground base station description data is at least used to reflect the bandwidth information of the communication link between each ground base station and the satellite and the communication window for effective communication with the satellite. In a specific implementation, other or more relevant information can also be included as long as it is related to the task allocation of the satellite and the ground base station. In addition, for the feature space mapping unit, in some implementations, various description data can be text data, and the feature space mapping unit can be a word embedding model, which is used to perform word embedding processing on the description data to obtain corresponding embedding features, that is, as multi-satellite description features, multi-ground base station description features and task description features respectively.

[0073] It should be noted that the orbital information of the satellite can be orbital altitude, orbital inclination, orbital period, orbital relative position, and orbital relative position refers to the specific position of the satellite in orbit and other information; visibility information can be whether the satellite is within the field of view of the ground base station; health status information can be whether the equipment is operating normally, whether there is a fault, whether there is sufficient power supply, etc.; the bandwidth information of the communication link between each ground base station and the satellite represents the transmission capacity of the communication link between the ground base station and the satellite, among which the bandwidth determines the amount of data that can be transmitted per unit time. The ground base station usually needs to allocate bandwidth according to the different requirements of the task. For example, some tasks may require higher bandwidth to transmit high-definition remote sensing images, while other tasks may only require lower bandwidth voice or simple data transmission. The communication window refers to the time period during which effective communication can be carried out between the ground base station and the satellite. The task scheduling of the ground base station and the scheduling of the satellite need to ensure that the communication windows of both parties are connected to effectively exchange data.

[0074] For example, some of the multi-satellite description data can be "orbital altitude is 400 kilometers, orbital inclination is 51.6 degrees, and orbital period is 92 minutes". Based on this, the embedded features corresponding to each word can be obtained:

[0075] "track": [0.21, -0.35, 0.57, 0.23, -0.41, 0.30, 0.12, -0.45, 0.62, -0.18, 0.50, 0.29, ...]; "height": [0.29, -0.51, 0.43, 0.19, -0.32, 0.23, 0.19, -0.50, 0.55, -0.12, 0.45, 0.34, ...]; "for": [0.03, 0.14, -0.29, 0.11, -0.06, -0.02, 0.08, 0.19, 0.05, -0.07, 0.14, -0.03, ...]; "40 "0": [0.01, 0.04, 0.07, -0.02, 0.03, 0.09, -0.05, 0.02, 0.06, -0.04, 0.00, 0.11, ...]; "km": [0.18, -0.27, 0.35, 0.19, -0.22, 0.28, 0.10, -0.34, 0.42, -0.16, 0.25, 0.18, ...]; "inclination": [0.33, -0.21, 0.41, 0.28, -0.18, 0.31, 0.14, -0.47, 0.57, -0.13, 0.29, 0.20, ...]; "51.6": "degree": [0.11, -0.05, 0.14, 0.03, -0.01, 0.07, 0.01, -0.03, 0.10, 0.00, 0.05, 0.089, ...]; "period": [0.29, -0.33, 0.49, 0.20, -0.25, 0.22, 0.14, -0.51, 0.60, -0.21, 0.37, 0.23, ...]; "92": [0.09, -0.02, 0.11, 0.02, -0.01, 0.06, 0.04, 0.01, 0.08, 0.04, 0.03, 0.07, ...]; "minutes": [0.17, -0.19, 0.27, 0.12, -0.13, 0.21, 0.10, -0.38, 0.45, -0.16, 0.31, 0.20, ...].

[0076] Preferably, in step 130, the performing associated feature aggregation on the target task description data based on the multi-satellite description feature, the multi-ground base station description feature and the multi-user task description feature to obtain the global task aggregation feature corresponding to the target task description data may include:

[0077] (1) For the target task description data, using the first feature aggregation unit in the feature processing model included in the target satellite task planning network model, combined with the task description features corresponding to the non-target task description data, the task description features corresponding to the target task description data are associated and aggregated, and the target task aggregation features corresponding to the target task description data are output; wherein the semantic information of the target task aggregation features includes the semantic information of the target task description features and the semantic information of the non-target task description features.

[0078] Specifically, after obtaining each description feature, for each task (target task) description data, the first feature aggregation unit in the feature processing model can be used to associate and aggregate the task description features corresponding to the task (target task) description data based on the task description features corresponding to other task (non-target task) description data, and output the target task aggregation features corresponding to the task description data. The semantic information of the target task aggregation features includes the semantic information of the corresponding task description features and the semantic information of other task description features.

[0079] For example, when there are three task description data, firstly, task description data 1 is taken as the target task description data and task description data 2 and 3 are taken as non-target task description data. Based on the task description features corresponding to task description data 2 and the task description features corresponding to task description data 3, the task description features corresponding to task description data 1 can be associated and aggregated, and the target task aggregated features 1 corresponding to task description data 1 can be output. Further, task description data 2 is taken as the target task description data and task description data 1 and 3 are taken as non-target task description data. Based on the task description features corresponding to task description data 1 and the task description features corresponding to task description data 3, the task description features corresponding to task description data 2 can be associated and aggregated, and the target task aggregated features 2 corresponding to task description data 2 can be output. Further, task description data 3 is taken as the target task description data and task description data 2 and 3 are taken as non-target task description data. Based on the task description features corresponding to task description data 1 and the task description features corresponding to task description data 2, the task description features corresponding to task description data 3 can be associated and aggregated, and the target task aggregated features 3 corresponding to task description data 3 can be output.

[0080] (2) For the target task description data, using the second feature aggregation unit in the feature processing model included in the target satellite task planning network model, combined with the multi-satellite description feature and the multi-ground base station description feature, the target task aggregation feature corresponding to the target task description data is associated and aggregated, and the global task aggregation feature corresponding to the target task description data is output.

[0081] Specifically, after obtaining the corresponding target task aggregation feature, for each target task description data, the target task aggregation feature corresponding to the target task description data can be associated and aggregated using the second feature aggregation unit in the feature processing model based on the multi-satellite description feature and the multi-ground base station description feature, and the global task aggregation feature corresponding to the target task description data can be output. Exemplarily, the target task aggregation feature 1 can be associated and aggregated based on the multi-satellite description feature and the multi-ground base station description feature, and the global task aggregation feature 1 corresponding to the task description data 1 can be output. And the target task aggregation feature 2 can be associated and aggregated based on the multi-satellite description feature and the multi-ground base station description feature, and the global task aggregation feature 2 corresponding to the task description data 2 can be output. And the target task aggregation feature 3 can be associated and aggregated based on the multi-satellite description feature and the multi-ground base station description feature, and the global task aggregation feature 3 corresponding to the task description data 3 can be output.

[0082] Preferably, for the target mission description data, the first feature aggregation unit in the feature processing model included in the target satellite mission planning network model is used to associate and aggregate the task description features corresponding to the target mission description data in combination with the task description features corresponding to the non-target mission description data, and output the target mission aggregation features corresponding to the target mission description data, which may include:

[0083] First, for the target task description data, the first association aggregation branch included in the first feature aggregation unit is used to associate and aggregate the task description features corresponding to the target task description data in combination with the task description features corresponding to the non-target task description data, and the first task aggregation features corresponding to the target task description data are output.

[0084] Specifically, for each target task description data, the first associated aggregation branch included in the first feature aggregation unit in the feature processing model can be used to associate and aggregate the task description features corresponding to the target task description data based on the task description features corresponding to other task description data, and output the first task aggregation features corresponding to the target task description data. It should be noted that the target task can represent all tasks corresponding to the user task or any of them.

[0085] Then, for the target task description data, the second association aggregation branch included in the first feature aggregation unit is used to associate and aggregate the task description features corresponding to the target task description data in combination with the task description features corresponding to the non-target task description data, and the second task aggregation features corresponding to the target task description data are output; wherein the association aggregation method corresponding to the first association aggregation branch is different from the association aggregation method corresponding to the second association aggregation branch.

[0086] Specifically, for each target task description data, the second association aggregation branch included in the first feature aggregation unit in the feature processing model can be used to associate and aggregate the task description features corresponding to the target task description data based on the task description features corresponding to the non-target task description data, and output the second task aggregation features corresponding to the target task description data. Among them, the association aggregation method corresponding to the first association aggregation branch is different from the association aggregation method corresponding to the second association aggregation branch. That is to say, in an embodiment of the present invention, association aggregation can be performed based on two different association aggregation methods, so that the semantics of the aggregated features are richer.

[0087] Finally, for the target task description data, the branch feature merging subunit included in the first feature aggregation unit is used to merge the first task aggregation feature and the second task aggregation feature, and the target task aggregation feature corresponding to the target task description data is output.

[0088] Specifically, after obtaining the first task aggregation feature and the second task aggregation feature, for each task description data, the first task aggregation feature and the second task aggregation feature corresponding to the target task description data can be merged using the branch feature merging subunit included in the first feature aggregation unit in the feature processing model, and the target task aggregation feature corresponding to the task description data can be output; illustratively, the first task aggregation feature and the second task aggregation feature can be spliced, superimposed, or weightedly superimposed to obtain the target task aggregation feature.

[0089] Preferably, for the target task description data, the first associated aggregation branch included in the first feature aggregation unit is used to associate and aggregate the task description features corresponding to the target task description data in combination with the task description features corresponding to the non-target task description data, and output the first task aggregation features corresponding to the target task description data, which may include: for each target task description data, the target task description data and the non-target task description data are loaded by the first associated aggregation branch, and data processing is performed in the first associated aggregation branch, which may include the following steps:

[0090] (1) Performing center determination processing on the task description features corresponding to each non-target task description data, and outputting the center task description features corresponding to each non-target task description data.

[0091] (2) Multiply the transposed result of the central task description feature corresponding to each non-target task description data and the task description feature corresponding to the target task description data, and output a parameter representing the correlation relationship between the target task description feature and the central task description feature.

[0092] (3) Based on the correlation relationship representation parameter between the target task description feature and the central task description feature, feature adjustment is performed on the task description feature corresponding to the target task description data, and the first task aggregation feature corresponding to the target task description data is output. For example, the task description feature may be weighted based on the correlation relationship representation parameter.

[0093] It should be noted that the specific manner of performing association aggregation using the first association aggregation branch included in the first feature aggregation unit in the feature processing model is not limited and can be selected according to actual needs.

[0094] Preferably, for the target task description data, the second associated aggregation branch included in the first feature aggregation unit is used to associate and aggregate the task description features corresponding to the target task description data in combination with the task description features corresponding to the non-target task description data, and the second task aggregation features corresponding to the target task description data are output, including: for the target task description data, the target task description data and the non-target task description data are loaded by the second associated aggregation branch, and data processing is performed in the second associated aggregation branch. Please refer to Figure 2 , Figure 2 A schematic diagram of feature processing of a second associated aggregation branch in a satellite mission planning method provided by the present invention.

[0095] exist Figure 2 In the method, for target task description data (each target task description data), using the second associated aggregation branch to load the target task description data and non-target task description data, and performing data processing in the first associated aggregation branch, the following steps may be included:

[0096] (1) Taking the task description feature corresponding to the target task description data as the center, determine the target distance between the task description feature corresponding to each non-target task description data and the center, such as the cosine distance between the features; and based on the target distance, perform serialization processing on the task description features corresponding to each non-target task description data to form a task description feature sequence corresponding to the non-target task description data; for example, the task description features can be sorted in ascending order according to the corresponding distance.

[0097] (2) For the first non-target task description feature in the task description feature sequence, perform self-attention processing on the first non-target task description feature, and output the task description attention feature corresponding to the first non-target task description feature; and characterize parameters based on the correlation relationship between the task description attention feature corresponding to the first non-target task description feature and the transposed result of the task description feature corresponding to the target task description data, such as: the result of multiplication; perform feature adjustment (such as weighting) on ​​the task description feature corresponding to the target task description data, and output the task description adjustment feature corresponding to the first task description feature.

[0098] (3) For the Nth task description feature in the task description feature sequence, perform self-attention processing on the Nth task description feature, and output the task description attention feature corresponding to the Nth task description feature, where N is greater than 1 and less than the number of non-target task description data; based on the correlation relationship representation parameter between the Nth task description attention feature and the transposed result of the task description adjustment feature corresponding to the N+1th task description feature, perform feature adjustment on the task description adjustment feature corresponding to the N+1th task description feature, and output the task description adjustment feature corresponding to the non-target task description feature; where N represents each other task description sub-feature other than the first task description feature; for example, when the number of non-target task description data is 3, N is 2; when the number of non-target task description data is 10, N is any number among 2, 3, 4, 5, 6, 7, 8, and 9.

[0099] (4) Determine a second task aggregation feature corresponding to the target task description data based on at least the task description adjustment feature corresponding to the last task description sub-feature in the task description feature sequence.

[0100] Exemplarily, in one embodiment, the task description adjustment feature corresponding to the last task description sub-feature can be used as the second task aggregation feature corresponding to the task description data; in another embodiment, the task description adjustment feature corresponding to each task description sub-feature in the task description feature sequence can be weighted and superimposed to obtain the second task aggregation feature corresponding to the task description data, wherein the weighting coefficient is the inverse of the aforementioned distance.

[0101] It should be noted that the specific manner of performing association aggregation using the second association aggregation branch included in the first feature aggregation unit in the feature processing model is not limited and can be selected according to actual needs.

[0102] Preferably, for the target task description data, using the second feature aggregation unit in the feature processing model included in the target satellite task planning network model, combining the multi-satellite description features and the multi-ground base station description features, the target task aggregation features corresponding to the target task description data are associated and aggregated, and the global task aggregation features corresponding to the target task description data are output, which may include the following steps:

[0103] (1) For the target task description data (each target task description data), the first feature aggregation subunit in the second feature aggregation unit is used to multiply the multi-satellite description feature with the transposed result of the target task aggregation feature corresponding to the target task description data, and output the first association relationship representation parameter corresponding to the target task description data; and the second feature aggregation subunit in the second feature aggregation unit is used to multiply the multi-ground base station description feature with the transposed result of the target task aggregation feature corresponding to the target task description data, and output the second association relationship representation parameter corresponding to the target task description data; and the third feature aggregation subunit in the second feature aggregation unit is used to multiply the mean feature of the multi-satellite description feature and the multi-ground base station description feature with the transposed result of the target task aggregation feature corresponding to the target task description data, and output the third association relationship representation parameter corresponding to the target task description data.

[0104] (2) The first association relationship representation parameter, the second association relationship representation parameter and the third association relationship representation parameter are fused to output the fused association relationship representation parameter corresponding to the target task description data; for example, the first association relationship representation parameter, the second association relationship representation parameter and the third association relationship representation parameter may be averaged to obtain the fused association relationship representation parameter corresponding to the task description data.

[0105] (3) Based on the fusion association relationship representation parameters, feature adjustment is performed on the target task aggregation features corresponding to the target task description data, such as weighted processing, to output the global task aggregation features corresponding to the target task description data.

[0106] It should be noted that the specific method of associating and aggregating the target task aggregation features corresponding to the task description data based on the multi-satellite description features and the multi-ground base station description features is not limited and can be selected according to actual needs. For example, in an alternative implementation, in order to ensure that the global task aggregation features output by the associated aggregation have a more reliable semantic representation capability.

[0107] Preferably, before step 120, feature processing is performed on multiple multi-source data using a feature processing model included in the target satellite mission planning network to obtain multi-satellite description features corresponding to the multi-satellite description data, multi-ground base station description features corresponding to the multi-ground base stations, and multi-user task description features corresponding to the multi-user task description data. This may include training a neural network to establish a target satellite mission planning network model with higher data processing capabilities.

[0108] Specifically, establishing a target satellite mission planning network model may include the following steps:

[0109] First, obtain sample multi-satellite description data corresponding to multiple satellites, sample multi-ground base station description data corresponding to multiple ground base stations, and multiple sample task description data corresponding to multiple sample user tasks, wherein each sample task description data is used to reflect the data size, bandwidth requirement and task priority of the corresponding sample user task; as described in the relevant description of step 110.

[0110] Secondly, using the feature processing model included in the candidate satellite mission planning network, feature mining and feature aggregation are performed on the sample multi-satellite description data, the sample multi-ground base station description data and each sample task description data, and the sample global task aggregation feature corresponding to each sample task description data is output, wherein the semantic information of each sample global task aggregation feature includes the semantic information of the corresponding sample task description data, the semantic information of other sample task description data, the semantic information of the sample multi-satellite description data and the semantic information of the sample multi-ground base station description data; as described in the relevant description of steps 120 to 130.

[0111] Then, the task planning model included in the candidate satellite task planning network is used to process the sample global task aggregation features corresponding to each sample task description data, and output sample satellite task planning data, wherein the sample satellite task planning data is used to reflect the allocation of multiple sample user tasks to each satellite and each ground base station; as described in the relevant description of step 140.

[0112] Finally, based on the difference between the sample satellite mission planning data and the actual satellite mission planning data corresponding to the sample satellite mission planning data, the network parameters of the candidate satellite mission planning network can be optimized to obtain the target satellite mission planning network. That is to say, the error between the sample satellite mission planning data and the actual satellite mission planning data can be calculated first (the specific calculation method of the error is not specifically limited here and can be configured according to actual needs), and then the network parameters of the candidate satellite mission planning network can be adjusted and optimized in the direction of reducing the error until convergence, thereby forming the target satellite mission planning network. Among them, the actual satellite mission planning data can be data formed based on analysis by corresponding experts.

[0113] Based on this, the present invention provides a method for satellite mission planning. When analyzing and mining multiple user tasks, feature mining and feature aggregation are performed separately for each user task, so that the granularity of feature processing is finer and the accuracy is higher; and in the process of separate processing, the features of other user tasks, the features of the satellite, and the features of the ground base station are aggregated, so that the semantic representation ability of the obtained global task aggregation feature is higher; further, when task planning is performed based on the global task aggregation feature, the reliability of the obtained target satellite mission planning data is higher, thereby solving the problem of relatively low reliability of satellite mission planning in the prior art.

[0114] In a second aspect, the present invention provides a satellite mission planning device, see Figure 3 , Figure 3 A schematic structural diagram of a satellite mission planning device provided by the present invention.

[0115] exist Figure 3 In the embodiment, the device may include:

[0116] The acquisition module 310 is used to acquire multiple multi-source data related to satellite mission planning; the multi-source data at least includes multiple satellite description data corresponding to the satellite, multiple ground base station description data corresponding to the ground base station, and multiple user task description data corresponding to the user task.

[0117] The satellite mission feature mining module 320 is used to perform feature mining on multiple multi-source data using the target satellite mission planning network model to obtain multi-satellite description features corresponding to the multi-satellite description data, multi-ground base station description features corresponding to the multi-ground base stations, and multi-user task description features corresponding to the multi-user task description data;

[0118] The satellite mission feature aggregation module 330 is used to perform associated feature aggregation on the target mission description data based on multi-satellite description features, multi-ground base station description features and multi-user mission description features to obtain global mission aggregation features corresponding to the target mission description data;

[0119] The satellite mission planning module 340 is used to perform mission planning processing on the global mission aggregation features corresponding to the target mission description data and output the target satellite mission planning data.

[0120] Based on this, the present invention provides a satellite mission planning device, which uses an acquisition module 310 to acquire multiple multi-source data related to satellite mission planning; the multi-source data at least includes multi-satellite description data corresponding to the satellite, multi-ground base station description data corresponding to the ground base station, and multi-user task description data corresponding to the user task; further, the satellite mission feature mining module 320 uses the target satellite mission planning network model to perform feature mining on the multiple multi-source data to obtain multi-satellite description features corresponding to the multi-satellite description data, multi-ground base station description features corresponding to the multi-ground base stations, and multi-user task description features corresponding to the multi-user task description data; and the satellite mission feature aggregation module 330 is used to perform associated feature aggregation on the target task description data based on the multi-satellite description features, the multi-ground base station description features and the multi-user task description features to obtain the global task aggregation features corresponding to the target task description data; finally, the satellite mission planning module 340 is used to perform task planning processing on the global task aggregation features corresponding to the target task description data, and output the target satellite mission planning data; thereby obtaining the target satellite mission planning data with higher reliability, solving the problem of relatively low reliability of satellite mission planning existing in the prior art, and ensuring the reliable operation of multi-satellites, multi-ground base stations and multi-user tasks.

[0121] In a third aspect, the present invention provides a device for satellite mission planning, see Figure 4 , Figure 4 A schematic diagram of the hardware structure of a satellite mission planning device provided by the present invention.

[0122] exist Figure 4 In the present invention, the device 400 is a satellite mission planning device, and the device 400 includes a memory 420, a first processor 410, a second processor 450, and a computer program stored in the memory 420 and executable on the first processor 410 and / or the second processor 450. When the first processor 410 and / or the second processor 450 executes the computer program, the method described in the first aspect, a satellite mission planning method, is implemented.

[0123] Optionally, the device 400 may include a communication interface 430. There may be one or more communication interfaces 430. The communication interface 430 may use any transceiver or other device for communicating with other devices or communication networks.

[0124] Optionally, the device 400 may further include a communication line 440. The communication line 440 may include a path to transmit information between the above components.

[0125] It should be noted that the memory may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory may exist independently and be connected to the processor through a communication line. The memory may also be integrated with the processor.

[0126] Optionally, the computer-executable instructions in the embodiment of the present invention may also be referred to as application program codes, which is not specifically limited in the embodiment of the present invention.

[0127] In a specific implementation, as an example, Figure 4 As shown, the first processor 410 and the second processor 450 may include one or more CPUs, such as Figure 4 CPU0 and CPU1 in.

[0128] In a fourth aspect, the present invention provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed, the satellite mission planning method described in the first aspect is implemented.

[0129] Among them, the steps executed when the aforementioned computer program is running will not be described one by one here, and reference may be made to the explanation of the satellite mission planning method described above.

[0130] Although the present invention is described herein in conjunction with various embodiments, in the process of implementing the claimed invention, those skilled in the art may understand and implement other variations of the disclosed embodiments by viewing the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other components or steps, and "one" or "an" does not exclude multiple situations. A single processor or other unit may implement several functions listed in a claim. Certain measures are recorded in different dependent claims, but this does not mean that these measures cannot be combined to produce good results.

[0131] Although the present invention has been described in conjunction with specific features and embodiments thereof, it is apparent that various modifications and combinations may be made thereto without departing from the spirit and scope of the present invention. Accordingly, this specification and the accompanying drawings are merely exemplary illustrations of the present invention as defined by the appended claims and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present invention. Obviously, those skilled in the art may make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, the present invention is intended to include such modifications and variations if they fall within the scope of the claims of the present invention and their equivalents.

Claims

1. A satellite mission planning method, characterized in that: include: Acquire multiple multi-source data relevant to satellite mission planning; The multi-source data at least includes multi-satellite description data corresponding to satellites, multi-ground base station description data corresponding to ground base stations, and multi-user task description data corresponding to user tasks; The target satellite mission planning network model is used to perform feature mining on the plurality of multi-source data to obtain multi-satellite description features corresponding to the multi-satellite description data, multi-ground base station description features corresponding to the multiple ground base stations, and multi-user task description features corresponding to the multi-user task description data; wherein, before the target satellite mission planning network model is used to perform feature mining on the plurality of multi-source data, the method includes: establishing the target satellite mission planning network model; the establishing of the target satellite mission planning network model includes: obtaining sample multi-satellite description data corresponding to multiple satellites, sample multi-ground base station description data corresponding to multiple ground base stations, and multiple sample task description data corresponding to multiple sample user tasks; using the feature processing model included in the candidate satellite mission planning network, the sample multi-satellite description data, the sample multi-ground base station description data, and each of the sample task description data are feature mined and feature aggregated, and each of the sample task description data is output. The sample global task aggregation features corresponding to the data are obtained, wherein the semantic information of each of the sample global task aggregation features includes the semantic information of the corresponding sample task description data, the semantic information of other sample task description data, the semantic information of the sample multi-satellite description data, and the semantic information of the sample multi-ground base station description data; using the task planning model included in the candidate satellite task planning network, the sample global task aggregation features corresponding to each of the sample task description data are processed, and sample satellite task planning data are output, wherein the sample satellite task planning data is used to reflect the allocation of each satellite and each ground base station to multiple sample user tasks; based on the difference between the sample satellite task planning data and the actual satellite task planning data corresponding to the sample satellite task planning data, the network parameters of the candidate satellite task planning network model are optimized to obtain the target satellite task planning network model; Based on the multi-satellite description features, the multi-ground base station description features and the multi-user task description features, the target task description data is aggregated with associated features to obtain a global task aggregation feature corresponding to the target task description data; Performing task planning processing on the global task aggregation features corresponding to the target task description data, and outputting target satellite task planning data.

2. The satellite mission planning method according to claim 1, characterized in that: The step of performing associated feature aggregation on the target task description data based on the multi-satellite description features, the multi-ground base station description features, and the multi-user task description features to obtain a global task aggregation feature corresponding to the target task description data includes: For the target task description data, the first feature aggregation unit in the feature processing model included in the target satellite mission planning network model is used to associate and aggregate the task description features corresponding to the target task description data in combination with the task description features corresponding to the non-target task description data, and the target task aggregation features corresponding to the target task description data are output; wherein the semantic information of the target task aggregation features includes the semantic information of the target task description features and the semantic information of the non-target task description features; For the target task description data, the second feature aggregation unit in the feature processing model included in the target satellite task planning network model is used to combine the multi-satellite description features and the multi-ground base station description features to associate and aggregate the target task aggregation features corresponding to the target task description data, and output the global task aggregation features corresponding to the target task description data.

3. The satellite mission planning method according to claim 2, characterized in that: The first feature aggregation unit in the feature processing model included in the target satellite mission planning network model is used to associate and aggregate the task description features corresponding to the target task description data in combination with the task description features corresponding to the non-target task description data, and output the target task aggregation features corresponding to the target task description data, including: Using the first association aggregation branch included in the first feature aggregation unit, in combination with the task description features corresponding to the non-target task description data, the task description features corresponding to the target task description data are associated and aggregated, and the first task aggregation features corresponding to the target task description data are output; Utilizing the second association aggregation branch included in the first feature aggregation unit, in combination with the task description features corresponding to the non-target task description data, the task description features corresponding to the target task description data are associated and aggregated, and the second task aggregation features corresponding to the target task description data are output; wherein the association aggregation method corresponding to the first association aggregation branch is different from the association aggregation method corresponding to the second association aggregation branch; The first task aggregation feature and the second task aggregation feature are merged by using the branch feature merging subunit included in the first feature aggregation unit, and the target task aggregation feature corresponding to the target task description data is output.

4. The satellite mission planning method according to claim 3, characterized in that: The first associative aggregation branch included in the first feature aggregation unit is used to associate and aggregate the task description features corresponding to the target task description data in combination with the task description features corresponding to the non-target task description data, and output the first task aggregation features corresponding to the target task description data, including: The target task description data and the non-target task description data are loaded by using the first association aggregation branch, and a task description feature corresponding to each non-target task description data is centrally determined in the first association aggregation branch to output a central task description feature corresponding to each non-target task description data; Multiplying the transposed result of the central task description feature corresponding to each non-target task description data and the task description feature corresponding to the target task description data, and outputting a parameter representing the correlation relationship between the target task description feature and the central task description feature; Based on the correlation relationship representation parameter between the target task description feature and the central task description feature, feature adjustment is performed on the task description feature corresponding to the target task description data, and a first task aggregation feature corresponding to the target task description data is output.

5. The satellite mission planning method according to claim 3, characterized in that: The step of utilizing the second association aggregation branch included in the first feature aggregation unit to associate and aggregate the task description features corresponding to the target task description data in combination with the task description features corresponding to the non-target task description data, and outputting the second task aggregation features corresponding to the target task description data, includes: For the target task description data, the target task description data and the non-target task description data are loaded by using the second associated aggregation branch, and in the second associated aggregation branch, the target distance between the task description feature corresponding to each non-target task description data and the center is determined respectively; and based on the target distance, the task description feature corresponding to each non-target task description data is serialized to form a task description feature sequence corresponding to the non-target task description data; For a first non-target task description feature in the task description feature sequence, self-attention processing is performed on the first non-target task description feature, and a task description attention feature corresponding to the first non-target task description feature is output; and based on a correlation relationship representation parameter between the task description attention feature corresponding to the first non-target task description feature and the transposed result of the task description feature corresponding to the target task description data, feature adjustment is performed on the task description feature corresponding to the target task description data, and a task description adjustment feature corresponding to the first task description feature is output; For the Nth task description feature in the task description feature sequence, perform self-attention processing on the Nth task description feature, and output the task description attention feature corresponding to the Nth task description feature, where N is greater than 1 and less than the number corresponding to the non-target task description data; based on the correlation relationship representation parameter between the Nth task description attention feature and the transposed result of the task description adjustment feature corresponding to the N+1th task description feature, perform feature adjustment on the task description adjustment feature corresponding to the N+1th task description feature, and output the task description adjustment feature corresponding to the non-target task description feature; A second task aggregation feature corresponding to the target task description data is determined based on at least the task description adjustment feature corresponding to the last task description sub-feature in the task description feature sequence.

6. The satellite mission planning method according to claim 2, characterized in that: Utilizing the second feature aggregation unit in the feature processing model included in the target satellite mission planning network model, combining the multi-satellite description features and the multi-ground base station description features, associating and aggregating the target task aggregation features corresponding to the target task description data, and outputting the global task aggregation features corresponding to the target task description data, including: Using the first feature aggregation subunit in the second feature aggregation unit, multiply the multi-satellite description feature by a transposed result of the target task aggregation feature corresponding to the target task description data, and output a first association relationship representation parameter corresponding to the target task description data; Using the second feature aggregation subunit in the second feature aggregation unit, multiply the multiple ground base station description features by the transposed result of the target task aggregation features corresponding to the target task description data, and output a second association relationship representation parameter corresponding to the target task description data; And using a third feature aggregation subunit in the second feature aggregation unit, multiplying the mean feature of the multi-satellite description feature and the multi-ground base station description feature by the transposed result of the target task aggregation feature corresponding to the target task description data, and outputting a third association relationship representation parameter corresponding to the target task description data; The first association relationship representation parameter, the second association relationship representation parameter and the third association relationship representation parameter are fused to output the fused association relationship representation parameter corresponding to the target task description data; Based on the fusion association relationship representation parameter, feature adjustment is performed on the target task aggregation feature corresponding to the target task description data, and the global task aggregation feature corresponding to the target task description data is output.

7. A satellite mission planning device, characterized in that: When the device is running, the satellite mission planning method according to claim 1 is implemented, including: An acquisition module, the acquisition module is used to acquire multiple multi-source data related to satellite mission planning; the multi-source data at least includes multiple satellite description data corresponding to the satellite, multiple ground base station description data corresponding to the ground base station, and multiple user task description data corresponding to the user task; A satellite mission feature mining module, wherein the satellite mission feature mining module is used to perform feature mining on the plurality of multi-source data using a target satellite mission planning network model to obtain multi-satellite description features corresponding to the multi-satellite description data, multi-ground base station description features corresponding to the multi-ground base stations, and multi-user task description features corresponding to the multi-user task description data; A satellite mission feature aggregation module, wherein the satellite mission feature aggregation module is used to perform associated feature aggregation on the target mission description data based on the multi-satellite description features, the multi-ground base station description features and the multi-user mission description features to obtain a global mission aggregation feature corresponding to the target mission description data; A satellite mission planning module is used to perform mission planning processing on the global mission aggregation features corresponding to the target mission description data and output target satellite mission planning data.

8. A satellite mission planning device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the satellite mission planning method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed, the satellite mission planning method according to any one of claims 1 to 6 is implemented.

Images