Heterogeneous satellite group dynamic planning method oriented to multi-stage tasks

Abstract

The invention provides a heterogeneous satellite group dynamic planning method for multi-stage tasks, and belongs to the technical field of satellite task planning. The method comprises the following steps: firstly, constructing a satellite group for collaborative observation based on priori knowledge; subpackaging the observation task according to the received observation point type and observation point latitude and longitude information in the standardized task demand; secondly, quickly establishing a mapping relation model between a satellite group and a task package according to task and resource constraints; and finally, by constructing a satellite remote sensing grading task planning model, receiving earth observation tasks, and outputting a satellite task planning scheme. Satellite task planning is decomposed into a service line and a calculation line, the service line carries out satellite group construction and task subpackage, the calculation line carries out earth observation task arrangement, the two lines operate at the same time, the task planning time is effectively shortened by reducing the solution space of the satellite task planning process, and the task planning efficiency is improved. Therefore, more sufficient reaction time is reserved for subsequent tasks.

Description

Technical Field

[0001] The invention belongs to the technical field of satellite mission planning, and in particular relates to a heterogeneous star cluster dynamic planning method for multi-stage missions. Background Art

[0002] At present, my country's existing remote sensing satellite mission planning and operation system has corresponding capabilities in single-satellite or multi-satellite collaborative mission planning, but its scale and efficiency can only meet the basic support requirements of the current remote sensing satellite and mission scale.

[0003] For observation points that maintain high-speed motion for a long time, more timely observation satellite mission planning is required, which is a new challenge for earth observation mission planning. The existing system capabilities cannot meet the needs of fast planning of multi-stage motion observation points for heterogeneous satellite constellations. It is urgent to carry out intelligent planning model research for multi-stage earth observation missions, improve the efficiency of mission planning, and leave more sufficient response time for subsequent missions. Summary of the invention

[0004] In order to solve the problems of delayed task control response during the identification and confirmation of moving observation points, continuous observation, and search and discovery, the present invention proposes a dynamic planning method for heterogeneous star clusters for multi-stage tasks. This method aims at the characteristics of strong mobility and narrow observation window of moving observation points, and starting from the comprehensive resources of the whole domain, according to the characteristics of tasks at each stage and the capability characteristics of heterogeneous star clusters, establishes a matching relationship between observation tasks and observation resources, and then performs hierarchical task planning, so as to achieve the purpose of rapid planning for highly dynamic tasks.

[0005] The first aspect of the present invention discloses a heterogeneous star cluster dynamic programming method for multi-stage tasks, wherein the method comprises:

[0006] Step S1, constructing a satellite constellation for collaborative observation based on prior knowledge;

[0007] Step S2, receiving standardized heterogeneous star cluster multi-stage task requirements, and extracting observation point types and observation point longitude and latitude information from the heterogeneous star cluster multi-stage task requirements, and then performing sub-package processing on the observation tasks according to the extracted parameter information, so as to classify observation points in the same task stage, observation points that cooperate with each other to complete the same task, or observation points that travel in a team into the same task package;

[0008] Step S3: quickly establish a matching relationship between observation tasks and observation resources based on task and resource constraints, and form a mapping relationship model between satellite constellations and task packages;

[0009] Step S4, according to different task stages, the task planning is divided into three levels: periodic task planning, real-time task planning, and real-time task insertion, and a hierarchical task planning model for earth observation is constructed; wherein the hierarchical task planning model for earth observation includes a periodic task planning model, a real-time task planning model, and a real-time task insertion model;

[0010] Step S5: input the satellite constellation and task subcontracting results into the earth observation hierarchical task planning model, solve it under the constraints of the mapping relationship model, and output the earth observation plan; wherein, during the task planning process, call the corresponding task planning model for different task stages.

[0011] The step S1 specifically includes:

[0012] Obtain the deployment location and historical trajectory information of motion observation points;

[0013] Based on the deployment position and historical trajectory information of the acquired moving observation points, the potential position and detection method of the observation points are predicted, and a satellite constellation for collaborative observation is constructed.

[0014] The deployment location information includes the longitude and latitude to be observed, the type and quantity of motion observation points; the historical trajectory information includes the historical appearance time and location of the motion observation points.

[0015] In step S2, the multi-stage task requirements of the heterogeneous star cluster include the observation point type, observation start and end time, observation point longitude and latitude, payload type, working mode, and task priority information; wherein the multi-stage tasks include fixed-point observation, departure status observation, observation point search, continuous observation, identification confirmation, return confirmation status observation, or task stages with timeliness requirements greater than a set threshold.

[0016] In step S4, the periodic task planning model, the real-time task planning model and the real-time task insertion model are all constructed in parallel based on graph neural networks and deep reinforcement learning. The graph neural network is used to perceive the global resource task status, and the deep reinforcement learning network is used for task scheduling decisions; among them, the output of the periodic task planning model can be used as the input of the real-time task planning model and the real-time task insertion model.

[0017] The real-time task insertion model builds a global situation based on graph neural networks and allocates observation tasks using a first-come, first-served assignment mode.

[0018] In step S5, for fixed-point observation, departure status observation and return confirmation status observation, the periodic task planning model is called, and task planning is performed at fixed intervals to realize traversal of the observation points; for observation point search, continuous observation and identification confirmation, the real-time task planning model is called, and the real-time task insertion model is called for the task stage whose timeliness requirement is greater than the set threshold.

[0019] The second aspect of the present invention discloses a heterogeneous star cluster dynamic planning system for multi-stage tasks, the system comprising:

[0020] A first processing module is configured to construct a satellite constellation for collaborative observation based on prior knowledge;

[0021] The second processing module is configured to receive standardized heterogeneous star cluster multi-stage task requirements, and extract observation point types and observation point longitude and latitude information from the heterogeneous star cluster multi-stage task requirements, and then perform sub-package processing on the observation tasks according to the extracted parameter information, so as to classify observation points in the same task stage, observation points that cooperate with each other to complete the same task, or observation points that travel in a team into the same task package;

[0022] The third processing module is configured to quickly establish a matching relationship between observation tasks and observation resources according to task and resource constraints, and form a mapping relationship model between satellite constellations and task packages;

[0023] The fourth processing module is configured to divide the task planning into three levels: periodic task planning, real-time task planning, and real-time task insertion according to different task stages, and construct a hierarchical task planning model for earth observation; wherein the hierarchical task planning model for earth observation includes a periodic task planning model, a real-time task planning model, and a real-time task insertion model;

[0024] The fifth processing module is configured to input the satellite constellation and task subcontracting results into the earth observation hierarchical task planning model, solve it under the constraints of the mapping relationship model, and output the earth observation plan; wherein, during the task planning process, the corresponding task planning model is called for different task stages.

[0025] The third aspect of the present invention discloses an electronic device. The electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps in any one of the heterogeneous star cluster dynamic programming methods for multi-stage tasks in the first aspect of the present disclosure are implemented.

[0026] The fourth aspect of the present invention discloses a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any one of the heterogeneous star cluster dynamic programming methods for multi-stage tasks in the first aspect of the present disclosure are implemented.

[0027] In summary, the solution proposed by the present invention has the following technical effects:

[0028] 1. The present invention innovatively proposes that after receiving the standardized requirements for earth observation, the ground system performs pre-calculations for satellite constellation construction and task subcontracting, and establishes a mapping relationship model between resources and tasks. It can determine in advance the satellite constellation required to complete a small batch of tasks, i.e., the task package, thereby effectively reducing the solution space of the subsequent task planning process and reducing the time consumption of task planning.

[0029] 2. The present invention constructs a hierarchical task planning model for the characteristics of multi-stage tasks, wherein the hierarchical model is mainly reflected in two aspects: first, the task planning is divided into three levels: periodic task planning, real-time task planning, and real-time task insertion, corresponding to each task stage; second, the graph neural network used for global resource task situation awareness and the deep reinforcement learning model used for task arrangement decision-making in the model are constructed separately to form a double-layer parallel calculation. Whether it is the three-level division of task planning or the parallel calculation of the double-layer model, it is helpful to quickly and accurately implement the multi-stage tasks facing the motion observation point to the satellite resources, and realize the rapid response of task planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0031] Figure 1 It is a general framework diagram of dynamic mission planning of heterogeneous star clusters according to an embodiment of the present invention;

[0032] Figure 2 A flow chart of dynamic task planning for a heterogeneous star cluster according to an embodiment of the present invention;

[0033] Figure 3 A diagram showing the idea of ​​periodic task planning and real-time task planning according to an embodiment of the present invention;

[0034] Figure 4 Insert a flow chart for a real-time task according to an embodiment of the present invention;

[0035] Figure 5 The figure is a structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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 making creative work are within the scope of protection of the present invention.

[0037] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, a first image may be referred to as a second image, and similarly, a second image may be referred to as a first image without departing from the scope of this application. Both the first image and the second image are images, but they are not the same image.

[0038] The first aspect of the present invention proposes a dynamic planning method for heterogeneous star clusters for multi-stage tasks. The method takes into account the characteristics of strong mobility of moving observation points and narrow observation windows, and starts from the overall comprehensive resources. According to the characteristics of tasks at each stage and the capabilities of heterogeneous star clusters, a matching relationship between observation tasks and observation resources is established, and hierarchical task planning is then performed, thereby achieving the purpose of rapid planning for highly dynamic tasks, such as Figure 1 shown.

[0039] See also Figure 2 ,The present invention first builds a satellite constellation based on the prior knowledge information of the ,observation points, then receives the standardized task requirements and performs ,subcontracting processing, then establishes a mapping relationship model between the ,tasks and resources, and finally the task planning model receives the ,preprocessed tasks and resources, and outputs the satellite mission observation ,plan.

[0040] A dynamic programming method for heterogeneous star clusters for multi-stage tasks includes the following steps:

[0041] Step 1: Obtain the deployment location and historical trajectory information of the motion observation point in advance;

[0042] Specifically, the deployment location information includes information such as the longitude and latitude to be observed, the type and number of motion observation points, etc.; the historical trajectory information includes the historical appearance time and location of the motion observation points.

[0043] Step 2: Based on prior knowledge such as the distribution and activity patterns of observation points, the potential locations and detection methods of observation points are predicted, and a satellite constellation for collaborative observation is constructed in a targeted manner;

[0044] Specifically, the basis for constructing a satellite constellation includes satellite capability attributes, historical observation mission information, etc., so that satellites that can complete the same type of tasks or satellites with collaborative observation capabilities are constructed into a satellite constellation.

[0045] Step 3: Receive standardized heterogeneous constellation multi-stage mission requirements;

[0046] Specifically, the task requirement is the detection requirement of moving observation points, which includes parameter information such as observation point type, observation start and end time, observation longitude and latitude, payload type, working mode, priority, etc.; multiple stages mainly include fixed-point observation status, departure status, observation point search status, continuous observation status, identification confirmation status, return confirmation status, etc.

[0047] Step 4: Extract the observation point type and latitude and longitude information in the task requirements, and subcontract the observation task according to the extracted parameter information;

[0048] Specifically, the purpose of task subcontracting is to group observation points that are in the same stage, observation points that collaborate to complete the same task, or observation points that work in teams into the same task package.

[0049] Step 5: Quickly establish the matching relationship between observation tasks and observation resources based on task and resource constraints, and form a mapping relationship model between satellite constellations and task packages;

[0050] Step 6: Construct a hierarchical mission planning model for Earth observation;

[0051] Specifically, the Earth Observation mission planning process includes three levels: periodic mission planning, real-time mission planning, and real-time mission insertion. Different levels of mission planning correspond to different mission planning models.

[0052] The periodic task planning and real-time task planning models are built based on graph neural networks and deep reinforcement learning. The graph neural network is used to perceive the global task status, and the deep reinforcement learning network is used for task scheduling decisions. Figure 3 .

[0053] The real-time task insertion model builds a global situation based on the graph neural network and allocates observation tasks in a first-come-first-served assignment mode. Figure 4 , the specific solution process is:

[0054] Step 6.1: Input the arranged satellite task graphs of various types and ad hoc meta-tasks, and convert the arranged satellite task graphs of various types into strip tasks; insert the ad hoc meta-tasks into the arranged strip tasks, and for the remaining ad hoc meta-tasks with conflicts, construct a single task strip, and then insert the ad hoc task strip into the arranged task graph;

[0055] Step 6.2: Find the current optional task from the task graph set and select the task according to the following two basic rules: ① If the current optional task set of a satellite contains a scheduled task, select the scheduled task. ② If the current optional set of the satellite does not contain a scheduled task, select the task with the earliest occurrence time from all executable tasks;

[0056] Step 6.3: Delete the selected task and recompose;

[0057] Step 6.4: Repeat steps 6.2 and 6.3 until all satellites have no optional tasks.

[0058] In the process of task planning, the corresponding task planning model is selected for different task stages. For the characteristics of fixed observation positions and high observation frequencies in stage tasks such as fixed-point observation, departure status, and return confirmation, a periodic task planning method is adopted to carry out task planning at fixed intervals to realize the traversal of observation points; for the characteristics of high timeliness requirements for stage tasks such as observation point search, continuous observation, and identification confirmation, a real-time task planning method is adopted to quickly realize the priority arrangement of multiple tasks based on the probability of observation point distribution; for tasks with extremely high timeliness requirements, a real-time assignment method is adopted to realize the rapid insertion of tasks. At the same time, before each task planning, the task planning system needs to perceive the arrangement and completion status of each satellite task in advance, obtain its idle time window, and complete the global resource task status update.

[0059] Step 7: Input the satellite constellation in step 2 and the subcontracted tasks in step 4 into the Earth observation hierarchical task planning model, solve it under the constraints of the mapping relationship model formed in step 5, and output the Earth observation plan.

[0060] The specific solution process is:

[0061] Step 7.1: Accept the earth observation mission, conduct satellite mission planning, and output the earth observation plan;

[0062] Step 7.2: Real-time perception of the on-board mission status of each type of satellite before the next mission planning action;

[0063] Step 7.3: Jump to step 7.1.

[0064] It should be further explained that the purpose of building a satellite constellation in step 2, subcontracting tasks in step 4, and establishing a resource and task mapping relationship model in step 5 is to reduce the solution space of the satellite mission planning process, thereby reducing the time consumption of mission planning.

[0065] In summary, the present invention closely couples the research contents of task and resource matching and dynamic task planning. Compared with the traditional satellite task planning method, the hierarchical multi-stage task planning model constructed by the present invention can give full play to the capabilities of heterogeneous satellite constellations, realize the targeted observation of specific tasks by satellites with different capabilities, and has important significance for the exploration and discovery of moving observation points and continuous observation.

[0066] The third aspect of the present invention discloses an electronic device. The electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps in any one of the heterogeneous star cluster dynamic programming methods for multi-stage tasks in the first aspect of the present disclosure are implemented.

[0067] Figure 5 is a structural diagram of an electronic device according to an embodiment of the present invention, such as Figure 5 As shown, the electronic device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. Among them, the processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the electronic device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, near field communication (NFC) or other technologies. The display screen of the electronic device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the electronic device can be a touch layer covered on the display screen, or a button, a trackball or a touch pad set on the housing of the electronic device, or an external keyboard, touch pad or mouse, etc.

[0068] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a structural diagram of the part related to the technical solution of the present disclosure, and does not constitute a limitation on the electronic device to which the technical solution of the present application is applied. The specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0069] The fourth aspect of the present invention discloses a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the computer program implements the steps in any one of the steps in the heterogeneous star cluster dynamic programming method for multi-stage tasks in the first aspect of the present disclosure.

[0070] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Although the present invention has been described in detail with reference to the above embodiments, it should be understood by those skilled in the art that the technical solutions described in the above embodiments can still be modified, or some or all of the technical features therein can be replaced by equivalents, and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A dynamic programming method for heterogeneous star clusters for multi-stage tasks, characterized in that: The method comprises: Step S1, constructing a satellite constellation for collaborative observation based on prior knowledge; Step S2, receiving standardized heterogeneous star cluster multi-stage task requirements, and extracting observation point types and observation point longitude and latitude information from the heterogeneous star cluster multi-stage task requirements, and then performing sub-package processing on the observation tasks according to the extracted parameter information, so as to classify observation points in the same task stage, observation points that cooperate with each other to complete the same task, or observation points that travel in a team into the same task package; Step S3: quickly establish a matching relationship between observation tasks and observation resources based on task and resource constraints, and form a mapping relationship model between satellite constellations and task packages; Step S4, according to different task stages, the task planning is divided into three levels: periodic task planning, real-time task planning, and real-time task insertion, and a hierarchical task planning model for earth observation is constructed; wherein the hierarchical task planning model for earth observation includes a periodic task planning model, a real-time task planning model, and a real-time task insertion model; Step S5: input the satellite constellation and task subcontracting results into the earth observation hierarchical task planning model, solve it under the constraints of the mapping relationship model, and output the earth observation plan; wherein, during the task planning process, call the corresponding task planning model for different task stages.

2. The method according to claim 1, characterized in that The step S1 specifically includes: Obtain the deployment location and historical trajectory information of motion observation points; Based on the deployment position and historical trajectory information of the acquired moving observation points, the potential position and detection method of the observation points are predicted, and a satellite constellation for collaborative observation is constructed.

3. The method according to claim 2, characterized in that The deployment location information includes the longitude and latitude to be observed, the type and quantity of motion observation points; the historical trajectory information includes the historical appearance time and location of the motion observation points.

4. The method according to claim 1, characterized in that: In step S2, the multi-stage task requirements of the heterogeneous star cluster include the observation point type, observation start and end time, observation point longitude and latitude, payload type, working mode, and task priority information; wherein the multi-stage tasks include fixed-point observation, departure status observation, observation point search, continuous observation, identification confirmation, return confirmation status observation, or task stages with timeliness requirements greater than a set threshold.

5. The method according to claim 1, characterized in that In step S4, the periodic task planning model, the real-time task planning model and the real-time task insertion model are all constructed in parallel based on graph neural networks and deep reinforcement learning. The graph neural network is used to perceive the global resource task status, and the deep reinforcement learning network is used for task scheduling decisions; among them, the output of the periodic task planning model can be used as the input of the real-time task planning model and the real-time task insertion model.

6. The method according to claim 5, characterized in that The real-time task insertion model builds a global situation based on graph neural networks and allocates observation tasks using a first-come, first-served assignment model.

7. The method according to claim 4, characterized in that In step S5, for fixed-point observation, departure status observation and return confirmation status observation, the periodic task planning model is called, and task planning is performed at fixed intervals to realize traversal of the observation points; for observation point search, continuous observation and identification confirmation, the real-time task planning model is called, and the real-time task insertion model is called for the task stage whose timeliness requirement is greater than the set threshold.

8. A dynamic programming system for heterogeneous star clusters for multi-stage tasks, characterized in that: The system comprises: A first processing module is configured to construct a satellite constellation for collaborative observation based on prior knowledge; The second processing module is configured to receive standardized heterogeneous star cluster multi-stage task requirements, and extract observation point types and observation point longitude and latitude information from the heterogeneous star cluster multi-stage task requirements, and then perform sub-package processing on the observation tasks according to the extracted parameter information, so as to classify observation points in the same task stage, observation points that cooperate with each other to complete the same task, or observation points that travel in a team into the same task package; The third processing module is configured to quickly establish a matching relationship between observation tasks and observation resources according to task and resource constraints, and form a mapping relationship model between satellite constellations and task packages; The fourth processing module is configured to divide the task planning into three levels: periodic task planning, real-time task planning, and real-time task insertion according to different task stages, and construct a hierarchical task planning model for earth observation; wherein the hierarchical task planning model for earth observation includes a periodic task planning model, a real-time task planning model, and a real-time task insertion model; The fifth processing module is configured to input the satellite constellation and task subcontracting results into the earth observation hierarchical task planning model, solve it under the constraints of the mapping relationship model, and output the earth observation plan; wherein, during the task planning process, the corresponding task planning model is called for different task stages.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps in the dynamic programming method for a heterogeneous star cluster for a multi-stage task according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the heterogeneous star cluster dynamic programming method for multi-stage tasks described in any one of claims 1 to 7 are implemented.

Images