Multi-target imaging mission planning method, device, and equipment based on multi-satellite collaboration

Abstract

This application relates to the field of satellite imaging technology and discloses a method, apparatus, and device for multi-target imaging mission planning based on multi-satellite collaboration. The method includes: dividing target points into several task groups and determining candidate satellites for each task group; determining a set of candidate time windows that meet preset imaging feasibility conditions for the candidate satellites corresponding to each task group, and selecting the effective time window corresponding to each task group; performing resource conflict detection on all candidate satellites and effective time windows corresponding to all task groups, and iteratively optimizing the candidate satellites and effective time windows if no resource conflict is detected, to obtain the optimal satellite allocation scheme for each target point; and controlling each candidate satellite to execute imaging tasks according to the optimal satellite allocation scheme. This application can improve the time synchronization, resource utilization, and mission execution success rate of multi-target observation, ensuring the time consistency and resource scheduling balance of multi-satellite multi-target mission planning.

Description

Technical Field

[0001] This application relates to the field of satellite imaging technology, and in particular to a multi-target imaging mission planning method, apparatus and equipment based on multi-satellite collaboration. Background Technology

[0002] With the increasing demand for Earth observation, imaging missions that simultaneously observe multiple discrete target points are gradually becoming more common. These imaging missions require multiple satellites to work together to complete the imaging within the same observation period.

[0003] However, it is difficult to guarantee the time consistency of observation data at each target point in related technologies, resulting in large differences in data timeliness, which cannot meet the requirements of high-precision imaging tasks. Furthermore, resource conflicts or satellite load imbalances are prone to occur when coordinating multiple satellites for imaging task planning, which seriously affects the execution effect of imaging tasks. Summary of the Invention

[0004] This application provides a multi-target imaging mission planning method, device, and equipment based on multi-satellite collaboration, which solves the technical problem of difficulty in ensuring time consistency and resource scheduling balance in current multi-satellite multi-target mission planning, and achieves the technical effect of improving the time synchronization of multi-target observation, resource utilization and mission execution success rate.

[0005] To achieve the above objectives, the main technical solutions adopted in this application include: Firstly, this application provides a multi-target imaging task planning method based on multi-satellite collaboration, the method comprising: Based on the spatial distribution characteristics of each target point, the target points are divided into several task groups, and candidate satellites corresponding to each task group are determined. For each of the candidate satellites corresponding to the task groups, a set of candidate time windows that meet the preset imaging feasibility conditions is determined, and the effective time window corresponding to the task group is selected from the set of candidate time windows. Resource conflict detection is performed on the candidate satellites and effective time windows corresponding to all the task groups. If no resource conflict is detected, the candidate satellites and effective time windows corresponding to each task group are iteratively optimized to obtain the optimal satellite allocation scheme for each target point. The candidate satellites are controlled to perform imaging tasks according to the optimal satellite allocation scheme.

[0006] The multi-target imaging mission planning method proposed in this application reduces the scale of the multi-target mission planning problem by dividing discretized target points into multiple task groups according to their spatial distribution. It also identifies corresponding candidate satellites for each task group, achieving preliminary planning for multi-satellite collaborative observation missions. Through candidate time window screening and effective time window optimization under pre-defined imaging feasibility conditions, it ensures imaging quality and temporal consistency. Simultaneously, it performs resource conflict detection on candidate satellites and effective time windows and iterative optimization when there are no conflicts, achieving globally optimal satellite allocation while ensuring mission feasibility. This helps to balance mission performance gains and satellite load balancing. Finally, it controls satellites to execute imaging missions based on the optimal allocation scheme. This effectively solves the temporal consistency problem in multi-satellite collaborative scenarios, avoids resource conflicts and satellite load imbalances, and improves the execution effect of imaging missions.

[0007] Secondly, this application provides a multi-target point imaging mission planning device based on multi-satellite cooperation, the device comprising: The satellite allocation module is used to divide the target points into several task groups according to the spatial distribution characteristics of each target point, and to determine the candidate satellites corresponding to each task group. The window filtering module is used to determine a set of candidate time windows that meet the preset imaging feasibility conditions for the candidate satellites corresponding to each task group, and to select the effective time window corresponding to the task group from the set of candidate time windows. The planning module is used to perform resource conflict detection on the candidate satellites and effective time windows corresponding to all the task groups, and if no resource conflict is detected, iteratively optimize the allocation relationship between each task group and the candidate satellites and the effective time window to obtain the optimal satellite allocation scheme for each target point. The execution module is used to control each of the candidate satellites to perform imaging tasks according to the optimal satellite allocation scheme.

[0008] Thirdly, this application provides a computer device, comprising: The system includes a memory and a processor, which are interconnected. The memory stores computer instructions, and the processor executes these computer instructions to perform the aforementioned multi-target imaging task planning method based on multi-satellite collaboration.

[0009] Fourthly, this application provides a computer-readable storage medium storing computer instructions, which are used to cause a computer to execute the above-described multi-target point imaging task planning method based on multi-satellite collaboration. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0011] Figure 1 This is one of the flowcharts illustrating a multi-target imaging task planning method based on multi-satellite collaboration, provided in an embodiment of this application. Figure 2 A structural block diagram of a multi-target imaging task planning device based on multi-satellite collaboration provided in this application embodiment; Figure 3 This is a structural block diagram of a computer device provided in an embodiment of this application. Detailed Implementation

[0012] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0013] In application scenarios such as multi-regional disaster comparative analysis and urban traffic congestion monitoring, multiple satellites are usually required to work together to observe and image multiple discrete target points. Such multi-target imaging tasks usually need to ensure that the observation data of each target point has temporal consistency, that is, multiple satellites need to work together to complete the imaging within the same observation period.

[0014] However, current satellite mission planning technologies primarily focus on revisit planning for single satellites targeting multiple targets, lacking solutions for concurrent scenarios involving multiple satellites and multiple target points. Specifically, these technologies struggle to coordinate multiple satellites to effectively observe targets at different geographical locations within the same time window, resulting in significant differences in data timeliness and low data availability. This leads to poor imaging quality and an inability to support high-precision imaging analysis.

[0015] Meanwhile, the related technologies often ignore the actual operating conditions of the satellites when planning satellite imaging missions, which can easily lead to mission failure due to insufficient power or storage overflow during the execution process. In addition, the related technologies cannot perform global optimization for mission planning of multiple satellites and multiple target points, which can easily cause some satellites to be overloaded while others are idle, resulting in low overall mission benefits and seriously affecting the execution effect of imaging missions.

[0016] Therefore, there is an urgent need for a multi-satellite collaborative multi-target imaging mission planning method that can take into account time consistency, resource constraints, and imaging quality.

[0017] The multi-target imaging mission planning method provided in this specification can be applied to multi-target imaging mission planning systems with satellite mission planning and scheduling capabilities. This system may include ground mission planning servers, onboard computers, communication terminals, and other electronic equipment. Of course, the multi-target imaging mission planning method provided in this specification can also be applied to applications running on the aforementioned electronic equipment.

[0018] According to an embodiment of this application, a multi-target point imaging mission planning method based on multi-satellite collaboration is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0019] This embodiment provides a multi-target imaging task planning method based on multi-satellite collaboration. Figure 1 This is a flowchart of a multi-target imaging task planning method based on multi-satellite collaboration according to an embodiment of this application, such as... Figure 1 As shown, the process includes the following steps: Step S1: Divide the target points into several task groups according to the spatial distribution characteristics of each target point, and determine the candidate satellites corresponding to each task group.

[0020] Specifically, a target point refers to an object to be observed that has a clear geographical location, parsed from an observation request. In this embodiment, the system receives observation requests from multiple sources, including ground control terminal APIs and natural language commands. It not only extracts the latitude and longitude coordinates corresponding to the target point from the observation request but also deeply analyzes the semantic constraints of the observation request. For example, when a user inputs "Please take a picture of Hangzhou's urban congestion during the evening rush hour," the system's built-in semantic understanding engine first accurately extracts the geographical entity "Hangzhou," calls a high-precision geocoding service to convert it into a polygon coordinate set representing the main urban area, and adaptively meshes the area according to the needs of urban congestion monitoring, generating hundreds of discrete latitude and longitude observation points covering major traffic arteries as the aforementioned target point. Simultaneously, the semantic understanding engine identifies the time entity "evening rush hour," and, combining the current UTC time and traffic pattern knowledge base of the target time zone, automatically maps it to a specific standard UTC time window interval (e.g., 17:00 to 19:00 on the same day). Based on the implicit requirements of the scenario, it automatically derives key constraint parameters, mapping the "urban congestion" monitoring accuracy requirement to a ground sampling distance (GSD) ≤ 0.5 meters, defaulting to using an optical imaging effectiveness strategy to set a cloud cover threshold < 5%, and marking the task as high priority (coefficient 1.0). Finally, the system encapsulates the above parsing results into a standardized list of task objects with complete attributes (time window, resolution, meteorological limitations, and priority). During the encapsulation process, strict legality checks are performed simultaneously. If any illegal situations are detected, such as coordinates exceeding the Earth's ellipsoid range, logical inversion of the time interval, or the requested resolution exceeding the physical limits of the on-orbit satellite, the processing flow is immediately interrupted, a specific exception code is thrown, and a natural language prompt containing the specific error reason is returned to the user to guide them in correcting the request.

[0021] Based on this, the geographical location information of all target points is obtained, and target points with similar geographical locations are grouped into the same task group according to their relative distance or degree of clustering. Subsequently, in combination with the satellite's orbital patterns, satellites with the potential to observe all target points within that group are pre-selected as candidate satellites for each task group.

[0022] Step S3: For each task group, determine the set of candidate time windows that meet the preset imaging feasibility conditions for the candidate satellites, and select the effective time window corresponding to the task group from the set of candidate time windows.

[0023] Specifically, for each mission group and its corresponding candidate satellite, the common-view time window in which the satellite can simultaneously cover all target points in the mission group is calculated within the satellite's operating cycle. Common-view time windows that meet the preset imaging feasibility conditions are selected as candidate time windows. These candidate time windows form a set of candidate time windows. The preset imaging feasibility conditions can be determined based on the satellite's attitude adjustment range, illumination requirements, weather conditions, etc., to ensure that the selected candidate time windows meet the imaging requirements in terms of geometric visibility and environmental adaptability.

[0024] Subsequently, the imaging quality of the candidate time windows is evaluated based on their geometric visibility (e.g., side-swing angle) and external meteorological conditions (e.g., illumination, cloud cover). Based on the evaluated imaging quality parameters, a valid time window is selected from the candidate time window set. This valid time window represents the feasible observation period on the corresponding candidate satellite where the mission group can achieve relatively good imaging results. It should be noted that candidate time windows that are not selected are used as backup time windows, serving as alternative solutions for adjusting mission execution time in subsequent resource conflict detection or iterative optimization.

[0025] Step S5: Perform resource conflict detection on the candidate satellites and effective time windows corresponding to all task groups. If no resource conflict is detected, iteratively optimize the candidate satellites and effective time windows corresponding to each task group to obtain the optimal satellite allocation scheme for each target point.

[0026] Specifically, based on the candidate satellites currently allocated to each mission group and the selected effective observation windows, the energy, storage space, and other satellite resource consumption required to execute these tasks are analyzed and compared with the actual available resources of each satellite to determine whether there is a resource overrun. If the resources required by all mission groups under their allocated candidate satellites and effective time windows do not exceed the upper limit of the available resources of the corresponding candidate satellites, then it is a feasible mission planning scheme for the candidate satellites to execute the imaging tasks of the mission group within the effective time window.

[0027] Based on this, an iterative optimization approach is adopted. Building upon the obtained feasible task planning schemes, new allocation combinations are continuously generated by iteratively adjusting the correspondence between task groups and candidate satellites, as well as the selected effective time windows. Optionally, in some embodiments of this application, a multi-objective genetic algorithm is employed. An initial population is generated based on feasible task planning schemes that have passed resource conflict detection. This ensures that the initial population contains at least one feasible solution, avoiding the algorithm from starting from a search space of completely infeasible solutions, reducing the exploration of a large number of invalid feasible solutions in the early stages, and thus improving the algorithm's convergence speed. Simultaneously, with multiple objective functions—maximizing task benefits, minimizing resource consumption variance, and maximizing constellation load balancing—and combining selection, crossover, and mutation operations in the genetic algorithm, a parallel computing architecture is used to decompose a large-scale task set into several sub-problems that evolve concurrently on different computing nodes, quickly converging to the globally optimal scheduling scheme.

[0028] Step S7: Control each candidate satellite to perform imaging tasks according to the optimal satellite allocation scheme.

[0029] Specifically, candidate satellites in the optimal satellite allocation scheme are used as execution satellites for each task group. The effective observation windows and necessary imaging parameters in the optimal satellite allocation scheme are converted into control commands that the execution satellites can recognize. These commands are then sent to the execution satellites via a communication link, and the execution satellites perform imaging and capturing of each target point in the task group according to the command content.

[0030] The multi-target imaging mission planning method provided in this application reduces the scale of the multi-target mission planning problem by dividing discretized target points into multiple task groups according to their spatial distribution. It also determines corresponding candidate satellites for each task group, achieving preliminary planning for multi-satellite collaborative observation missions. Through candidate time window screening and effective time window optimization under preset imaging feasibility conditions, it ensures imaging quality and time consistency. Simultaneously, it performs resource conflict detection on candidate satellites and effective time windows and iterative optimization when there are no conflicts. This enables globally optimal satellite allocation while ensuring mission feasibility, facilitating a balance between mission performance benefits and satellite load balancing. Finally, it controls satellites to execute imaging missions based on the optimal allocation scheme. This effectively solves the time consistency problem in multi-satellite collaborative scenarios, avoids resource conflicts and satellite load imbalances, and improves the execution effect of imaging missions.

[0031] In some embodiments of this application, the step S1 described above, which involves dividing the target points into several task groups based on the spatial distribution characteristics of each target point, may include the following steps: Step S111: Determine the preset task priority corresponding to each target point.

[0032] Specifically, a priority value is assigned to each target point based on the source of the observation task or the importance set by the user. i For example, if the observation task for the i-th target point comes from an emergency disaster relief need, then a higher preset task priority is assigned to that target point. i =1.0, if the observation task for the target point comes from routine monitoring needs, then a lower preset task priority will be assigned to the target point. i =0.5~0.8, if the observation task of the target point comes from experimental verification needs, then a lower preset task priority is assigned to the target point. i =0.3~0.5. It can be understood that a higher priority indicates a more urgent need for observation of the target point or a greater value, and it will receive a higher weight in subsequent clustering and resource allocation.

[0033] Step S113: Determine the spatial distribution characteristics based on the relative distance and local density of each target point, and perform weighted processing on the spatial distribution characteristics according to the preset task priority, so as to determine the cluster center based on the weighted spatial distribution characteristics.

[0034] Specifically, the local density of the i-th target point can be determined according to the following formula (1): In the formula, This represents the local density of the i-th target point. This represents the spherical distance between the i-th and j-th target points, calculated using the Haversine formula based on latitude and longitude coordinates. This represents the cutoff distance, which is a pre-defined distance threshold used to determine whether two target points are close to each other. The cutoff distance is calculated using all spherical distances. The kth value after sorting from smallest to largest, in one example of the embodiments of this application, k takes 1% to 2% of the total number of target points. Indicates an indicator function, <0 o'clock The value is 1, otherwise The value is 0.

[0035] The relative distance to the i-th target point can be determined according to the following formula (2): In the formula, This represents the relative distance to the i-th target point. It can be understood that the relative distance to a target point represents the minimum distance between that target point and all target points with a local density greater than its own.

[0036] Furthermore, the product of relative distance and local density is used as the spatial distribution feature of each target point, that is, the spatial distribution feature of the i-th target point. The cluster centers are selected by weighting the data based on the preset task priorities and the highest weighted values.

[0037] It should be noted that in some embodiments of this application, target points with preset task priorities higher than a set level threshold can be directly set as independent cluster centers to prevent them from being overwhelmed by target points with lower task priorities.

[0038] Step S115: Determine the cluster radius based on the maximum ground coverage width of all satellites.

[0039] Specifically, the real-time orbital elements of all available satellites are iterated to calculate the maximum ground coverage swath of each satellite within the next 24 hours, taking into account its side-swing capability; that is, the ground width that can be observed during a single transit. The average swath width of all satellites is then taken as the cluster radius R. init .

[0040] Step S117: Divide the target points into several task groups based on the cluster center and cluster radius.

[0041] Specifically, starting from a defined cluster center, neighboring target points within the cluster radius are progressively assigned to various task groups. Furthermore, this embodiment introduces a common-view constraint penalty during the clustering process. If adding a target point to the current group results in any two target points within the group being unable to be simultaneously observed by the same satellite in the same overpass, a significant distance penalty is imposed on that target point. The distance from that point to the cluster center is multiplied by a penalty coefficient P much greater than 1, such as P=1000, causing it to be excluded from the current task group and reassigned to another suitable task group.

[0042] After clustering is completed, check whether the spatial span of each task group exceeds the maximum side-swing coverage range of a single satellite. If it does, recursively split the group until all task groups meet the single-satellite coverage condition. Finally, output several spatially compact task groups that can be synchronously observed by a single satellite.

[0043] This application's embodiments, by introducing task priority weighting, can ensure that high-priority target points become cluster centers first, avoiding being overwhelmed by low-priority targets, thereby improving the rationality of task group division and the targeting of high-level task execution.

[0044] In some embodiments of this application, the step S1 described above, which involves determining the candidate satellites for each mission group, may further include the following steps: Step S121: Calculate the geometric visibility relationship and observation quality between each mission group and each candidate satellite based on the satellite orbit prediction information.

[0045] Satellite orbit prediction information refers to the position vector of a satellite over a future period of time. Specifically, by loading the latest two-line elements (TLE) file, the Simplified General Perturbations model 4 (SGP4) is used to extrapolate the ephemeris of all satellites over the next 72 hours in 1-second increments to determine the corresponding position vector. For each mission group, the coordinates of its spatial boundary vertices are extracted, and the elevation angle of each satellite over these vertices is calculated. Specifically, this elevation angle can be determined using the following formula (3): In the formula, Indicates the angle of elevation. This represents the Earth's average radius, which is taken as 6,371,000 meters in this formula. This represents the satellite's position vector in the ECEF coordinate system. This represents the position vector of the target point in the ECEF coordinate system.

[0046] When the elevation angle of a satellite to all target points within the mission group is greater than a preset minimum elevation angle threshold at a certain moment, the satellite is determined to be visible to the mission group at that moment. Optionally, the minimum elevation angle threshold is set to 5°. This constructs a three-dimensional sparse tensor V with dimensions of mission group, satellite, and time slice. ijk The geometric visibility relationship described above is represented by a three-dimensional sparse matrix. Specifically, i represents the mission group, j represents the candidate satellite, k represents the time slice, and an element value of 1 indicates visibility, while 0 indicates invisibility.

[0047] Simultaneously, the side-slip angle is calculated based on the geometric relationship between the satellite and the target point. The magnitude of the side-slip angle is used as the basis for evaluating the geometric quality of the observation; the smaller the side-slip angle, the higher the observation quality. Specifically, the side-slip angle can be determined using the following formula (4): In the formula, Indicates the lateral sway angle. This represents the Earth's average radius, which is taken as 6,371,000 meters in this formula. Indicates the satellite's orbital altitude. The geocentric angle between the satellite's nadir and the target point is determined by the following formula (5): In the formula, Indicates the latitude of the satellite's nadir point. Indicates the longitude of the satellite's nadir point. Indicates the latitude of the target point. Indicates the longitude of the target point.

[0048] Step S123: Construct a weighted bipartite graph with the task group as the left node and the satellite as the right node based on the geometric visibility relationship. The weight of each edge in the weighted bipartite graph is determined according to the observation quality.

[0049] Specifically, let the set of task groups be C = {c1, c2}. 2, ..., c m The set of satellites is S = {s1, s2, ..., s}. n Each task group is considered a left-hand node in a bipartite graph, and each satellite is considered a right-hand node. If the three-dimensional sparse tensor V representing the geometric visibility relationship... ijk There exists at least one time slice in the process that makes the satellite visible to the mission group, i.e., V ijk =1, then in the corresponding task group c i The left node and the corresponding satellite s j An edge is established between the right-hand nodes, and a list of available satellites is formed by connecting the satellites with the left-hand nodes. The weight of the edge is determined by the observation geometry quality and the effective co-view duration, and can be specifically determined by the following formula (6): In the formula, Indicates task group c i Corresponding node and satellite s j The weights of the edges between corresponding nodes. Indicates task group c i By satellite s j The observation quality parameters during observation can be understood as follows: It is determined by the lateral angle between the satellite and the target point. This represents the effective co-view duration, which in this embodiment is the total number of visible time slices, i.e., V. ijk The total number of time slices equal to 1.

[0050] It is understandable to utilize the edge weights. It can comprehensively reflect the observation quality and the length of time that the satellite can be used for observation.

[0051] Step S125: Solve the weighted bipartite graph based on the maximum weight matching algorithm to determine the candidate satellites corresponding to each task group based on the solution results.

[0052] Specifically, the Kuhn-Munkres (KM) algorithm is used to solve for the maximum weight matching of the weighted bipartite graph described above. First, feasible vertex labels are initialized, where c is the left vertex. i The top label l(c) i ) is max j (w ij ), right vertex sj The top label l(s) j The first step is to set the leftmost vertex to 0. Then, for each leftmost vertex, search for an augmenting path in the equality subgraph. If found, update the matching and adjust the vertex labels until all leftmost nodes are matched or no augmenting path can be found. It can be understood that an equality subgraph is defined as a graph consisting of all nodes that satisfy l(c) = 0. i )+l(s j )=w ij An augmenting path is a subgraph formed by the edges of a given set of unmatched vertices. It starts from an unmatched left vertex, alternately traverses unmatched and matched edges, and finally reaches an unmatched right vertex. The optimal matching pair {(c} is then obtained. i, s j )}, in this optimal matching pair s j That is, for task group c i, The first-choice satellite is assigned. For mission groups that are not matched, a satellite is selected from the list of available satellites in descending order of edge weight as the second-choice satellite. These first-choice and second-choice satellites are the candidate satellites for each mission group.

[0053] Simultaneously, for each mission group, satellites other than the selected candidate satellites in the available satellite list are recorded as candidate satellites for use in subsequent resource conflict detection or iterative optimization. Furthermore, the matching process also considers matching constraints based on satellite payload type. For example, the synthetic aperture radar mission group only matches satellites with SAR payloads; satellites that do not meet the payload type requirements will not appear in the available satellite list.

[0054] This application embodiment uses weighted bipartite graph modeling and maximum weight matching algorithm to solve the problem, which can quickly allocate the most suitable candidate satellites to each task group, while ensuring the comprehensive optimization of observation quality and available window duration. This improves the global optimization of satellite allocation and provides an accurate solution basis for subsequent mission planning.

[0055] In some embodiments of this application, step S3 may include the following steps: Step S31: Obtain the common viewing time window that the candidate satellite can cover all target points in the corresponding task group.

[0056] Specifically, for each mission group and its assigned candidate satellites, within the orbital period of the candidate satellites, the operational segments of the candidate satellites are traversed to search for the time interval during which the satellite is simultaneously visible to all target points within the mission group, thus obtaining the common visibility time window. It can be understood that the visibility of the satellite to the target point is determined by calculating the elevation angle of the satellite to each target point and comparing it with a preset minimum elevation angle threshold.

[0057] Step S33: Correct the start time of the common-view time window based on the satellite attitude maneuver time of the candidate satellite to obtain the corrected visible time window.

[0058] Specifically, using the attitude dynamics model, the attitude maneuvering time of the satellite is determined based on the angular difference between the side-slip angle of the candidate satellite at the end of the previous mission and the side-slip angle of the initial attitude required for the current mission, combined with the satellite's attitude maneuvering capability parameters. T slew Then, the satellite attitude maneuver time... T slew Accumulated to the start time of the common-view time window, shifting the start time backward, thus obtaining the result after deducting the satellite attitude maneuver time. T slew The subsequent correction is visible in the time window.

[0059] Step S35: Obtain the imaging quality parameters of the candidate satellite pairs for the mission group under the modified visible time window, and select the modified visible time window that meets the preset imaging feasibility conditions as the candidate time window based on the imaging quality parameters, so as to form a set of candidate time windows.

[0060] Optionally, the side-slip angle of the candidate satellites relative to the mission group is obtained within the corrected visible time window. Solar altitude angle and average cloud cover As an image quality parameter, the side sway angle is... For specific calculations, please refer to formula (4) above. The solar altitude angle can be determined by the following formula (7): In the formula, Indicates the solar altitude angle. Indicates the solar declination angle. Indicates the latitude of the target point. It represents the solar hour angle and has , The local time of the target point. Indicates the longitude of the target point. This indicates the longitude of the central meridian of the survey area.

[0061] Simultaneously, the system connects to the global numerical weather prediction data interface to obtain cloud cover prediction grid data for the next 72 hours over the target point, and calculates the average cloud cover in the area where the task group is located through bilinear interpolation. .

[0062] In some embodiments of this application, the preset imaging feasibility conditions are determined based on preset side-swing angle constraints, preset solar altitude angle constraints, and preset cloud cover constraints for the imaging mission. Specifically, the preset side-swing angle constraint is a preset maximum side-swing angle threshold, the preset solar altitude angle constraint is a preset minimum solar altitude angle threshold, and the preset cloud cover constraint is a preset maximum cloud cover threshold. That is, for each corrected visibility time window, it is checked whether the side-swing angle of the candidate satellite to each target point in the mission group does not exceed the preset maximum side-swing angle threshold during the window, whether the solar altitude angle of the region where the mission group is located is not lower than the preset minimum solar altitude angle threshold during the window, and whether the average cloud cover of the region where the mission group is located does not exceed the preset maximum cloud cover threshold during the window.

[0063] For example, the side-swing angle of candidate satellites during the correction visibility window is obtained, and correction visibility windows with an absolute side-swing angle less than or equal to 20° are eliminated to ensure that imaging geometric distortion is within a controllable range.

[0064] The solar altitude angle of the mission group's area was determined by combining the solar ephemeris, and corrected visible time windows with a solar altitude angle of less than 15° were eliminated to ensure sufficient light.

[0065] Meanwhile, the system connects to the global numerical weather prediction data interface to obtain cloud cover prediction grid data for the next 72 hours over the target point. It calculates the average cloud cover in the area where the task group is located through bilinear interpolation, and removes the corrected visibility time window with cloud cover exceeding 30% to ensure the clarity of optical imaging.

[0066] It is understandable that the modified visual time window that simultaneously satisfies the above three constraints is retained as a candidate time window, and the candidate time window set is composed of all candidate time windows.

[0067] Step S37: Sort the candidate time windows according to the imaging quality parameters, and determine the effective time window based on the sorting results.

[0068] Specifically, based on the lateral sway angle Solar altitude angle and average cloud cover The quality score of the candidate time window can be expressed by the following formula (8): In the formula, This represents the imaging quality parameter corresponding to a candidate time window. This indicates the aforementioned lateral sway angle. This indicates the aforementioned preset maximum lateral sway angle threshold. This indicates the solar altitude angle mentioned above. This indicates the aforementioned preset minimum solar altitude angle threshold. This represents the average cloud cover mentioned above. This indicates the preset maximum cloud cover threshold. , , These represent the weighting coefficients for the three factors: lateral tilt angle, solar altitude angle, and average cloud cover, respectively, with values ​​ranging from [0,1].

[0069] Furthermore, the candidate time windows are sorted in descending order of quality score, and the window with the highest quality score is selected as the effective time window. It can be understood that the effective time window represents the observation period during which the candidate satellite can perform imaging tasks with better imaging quality for a certain task group. Other windows serve as backup time windows for adjusting task execution time during subsequent resource conflict detection or iterative optimization.

[0070] This application embodiment uses attitude maneuver time correction and candidate observation window scoring and ranking, combined with preset imaging feasibility conditions and imaging quality parameters, to select observation windows that meet multi-dimensional constraints such as attitude, illumination, and weather for each mission group and its candidate satellites, and selects the window with the best imaging quality as the effective time window, thereby ensuring imaging quality and improving the mission execution success rate.

[0071] In some embodiments of this application, the resource conflict detection of candidate satellites and effective time windows corresponding to all task groups described in step S5 above may include the following steps: Step S511: Obtain the remaining energy and remaining storage of the candidate satellite at the start of the effective time window.

[0072] Specifically, remaining energy refers to the current state of charge (SoC) of the candidate satellite at the beginning of a certain effective time window, and remaining storage refers to the remaining storage space of the candidate satellite at the beginning of a certain effective time window.

[0073] In this embodiment, the remaining energy and storage capacity of each candidate satellite are first initialized. Then, simulations are performed according to the time sequence of the effective time windows corresponding to each task group to estimate the total energy consumption and data volume of each candidate satellite during the effective time window. See step S513 below for details. Then, the remaining energy and storage capacity at the start of the next effective time window are updated based on the estimated total energy consumption.

[0074] Step S513: Determine the estimated total energy consumption based on the attitude maneuver energy consumption and payload imaging energy consumption of the candidate satellite when performing the corresponding task group, and determine the estimated data volume of the candidate satellite when performing the imaging task of the corresponding task group.

[0075] Specifically, for each task group and its allocated effective time window, firstly, based on the change in lateral angle of the candidate satellite from the end attitude of the previous task to the start attitude of the current task, then look up the "lateral angle change - current integral" lookup table calibrated by the attitude control system. It can be understood that this lookup table is calibrated by the satellite development unit through ground testing. By looking up the table, the current time integral value corresponding to the lateral angle change Δθ is obtained, and then the current time integral value is multiplied by the bus voltage of the satellite attitude control system to obtain the attitude maneuver energy consumption. That is, the attitude maneuver energy consumption can be expressed by the following formula (9): In the formula, Indicates energy consumption for attitude maneuvering. This represents the bus voltage of the satellite attitude control system. This represents the integral value of the current over time. This represents the time integral value of the current obtained from the table lookup.

[0076] Simultaneously, the satellite's operating power consumption is determined based on the imaging mode of the candidate satellites, such as the power consumption P in pushbroom mode. push Gaze mode power consumption P stare The payload energy consumption is given by the payload specifications of the candidate satellite. Then, the satellite's operating power consumption is multiplied by the imaging duration to obtain the payload imaging energy consumption, which can be expressed by the following formula (10): In the formula, Indicates the energy consumption of payload imaging. Indicates the satellite's operating power consumption. The duration of imaging is determined by dividing the required ground coverage length by the satellite ground velocity and the sensor swath width.

[0077] Furthermore, based on attitude maneuver energy consumption and payload imaging energy consumption The sum determines the estimated total energy consumption. .

[0078] Furthermore, in this embodiment of the application, the estimated data volume of a candidate satellite performing the imaging task of the corresponding task group is determined based on the sensor data output rate, imaging duration, and quantization bit number of the candidate satellite. This estimated data volume can be expressed by the following formula (11): In the formula, This indicates the estimated amount of data. This indicates the sensor data output rate, specifically the amount of data the sensor outputs per second during imaging. Indicates the duration of imaging. This indicates the number of quantization bits, specifically the number of bits used for each data item.

[0079] Step S515: The estimated total energy consumption is compared with the remaining energy to obtain a first comparison result, and the estimated data volume is compared with the remaining storage volume to obtain a second comparison result, so as to determine whether there is a resource conflict based on the first comparison result and the second comparison result.

[0080] Specifically, the first comparison result indicates whether the estimated total energy consumption exceeds a preset safety threshold for the remaining energy, and the second comparison result indicates whether the estimated data volume exceeds a preset safety threshold for the remaining storage. For example, in this embodiment of the application, an energy consumption constraint as shown in formula (12) is set as the first preset condition for resource conflict detection, and a storage constraint as shown in formula (13) is set as the second preset condition: In the formula, Indicates the amount of remaining energy. This indicates the remaining storage capacity.

[0081] In this embodiment, 80% of the remaining energy is used as the preset safety threshold for the remaining energy, and 80% of the remaining storage is used as the preset safety threshold for the remaining storage.

[0082] Furthermore, if the first comparison result does not meet the first preset condition or the second comparison result does not meet the second preset condition, that is, if the first comparison result does not meet the above formula (12) or the second comparison result does not meet the above formula (13), then it is determined that a resource conflict has been detected.

[0083] If no resource conflict is detected, the candidate satellites and effective time windows corresponding to each task group will be used as a preliminary feasible scheme for imaging mission planning.

[0084] Therefore, this application embodiment, by comprehensively judging the energy consumption and data volume of candidate satellites during mission execution, can predict the resource status after mission execution in advance, effectively identify potential energy and storage conflicts, and ensure the feasibility of mission planning schemes.

[0085] In the event of a detected resource conflict, the method further includes adjusting the resource conflict based on a preset conflict resolution strategy, which includes: The candidate time windows are traversed according to their sorting results, and the time windows of the task groups are adjusted based on the traversed candidate time windows. If there are no resource conflicts after the time window adjustment, the preset conflict resolution strategy is terminated.

[0086] If resource conflicts still exist after traversing all candidate time windows and adjusting the time windows, then reduce the imaging parameters of the task group. If there are no resource conflicts after reducing the imaging parameters, then end the preset conflict resolution strategy.

[0087] If resource conflicts persist after reducing imaging parameters, the task group will be moved to another satellite. If no resource conflicts exist after the move, the preset conflict resolution strategy will be terminated. If resource conflicts still exist after the move, the task group will be temporarily removed.

[0088] Specifically, the aforementioned pre-defined conflict resolution strategy employs a hierarchical heuristic algorithm. First, the first-level resolution strategy is executed, attempting to adjust the execution time window of the task group to its backup time window, or making minor adjustments on the timeline, such as delaying the start time to utilize the charging gap. After adjusting the time window, steps S511 to S515 are executed again to determine if the resource conflict has been resolved. If the resource conflict cannot be resolved after traversing all candidate time windows, it indicates that the time window adjustment method is ineffective in resolving resource conflicts, and then the second-level resolution strategy is executed.

[0089] The second-level resolution strategy attempts to reduce imaging parameters such as the imaging resolution of the task group and shorten the imaging duration to reduce energy consumption and data volume. After reducing the imaging parameters, steps S511 to S515 are executed again to determine whether the resource conflict has been resolved. If the resource conflict still cannot be resolved, it indicates that reducing the imaging parameters is ineffective in adjusting the resource conflict, and then the third-level resolution strategy is executed.

[0090] The third-level resolution strategy attempts to migrate the mission group to other candidate satellites with sufficient resources, i.e., changing the candidate satellites assigned to the mission group. It is understood that these candidate satellites can be selected from the list of available satellites obtained after step S125 above; see the explanation of step S125 above for details, which will not be repeated here. After migration, steps S511 to S515 are executed again to determine whether the resource conflict has been resolved. If the resource conflict still cannot be resolved, it indicates that the satellite migration method is ineffective in resolving the resource conflict.

[0091] If all the above multi-level resolution strategies fail, the task group will be temporarily removed and will be dealt with again in a subsequent planning cycle.

[0092] The embodiments of this application adopt a hierarchical resolution strategy, which is conducive to flexibly resolving resource conflicts and improving the success rate of task planning and resource utilization efficiency.

[0093] In some embodiments of this application, the iterative optimization of candidate satellites and effective time windows corresponding to each task group described in step S5 above to obtain the optimal satellite allocation scheme for each target point may further include the following steps: Step S521: Generate an initial population based on the candidate satellites currently allocated to each task group and the effective time window.

[0094] Specifically, in the embodiments of this application, the candidate satellites and effective time windows currently allocated to each task group are used as seed individuals, and then combined with random perturbation, such as randomly changing the candidate satellites or effective time windows of some task groups, an initial population containing multiple chromosomes is generated.

[0095] It is understood that, in this application's embodiments, the triple sequence consisting of "task group ID - candidate satellite ID - effective time window ID" is encoded as a chromosome, and each chromosome corresponds to a complete task planning scheme. For example, the chromosome encoding of an individual in the initial population is... Each chromosome represents a task group. Assigned to satellite and within the time window Execute internally.

[0096] Furthermore, when the task scale is large, the search space of the genetic algorithm grows exponentially, and single-machine computing is difficult to meet the real-time requirements of task planning. Therefore, in this embodiment, the initial population is divided into P subpopulations according to geographical regions and distributed to P computing nodes to execute the following steps S523 to S529 in parallel. That is, each computing node is independently responsible for the optimization solution search of a corresponding subpopulation. For example, computing node 1 is responsible for optimizing the satellite imaging task planning scheme corresponding to the northern region task group, and computing node 2 is responsible for optimizing the satellite imaging task planning scheme corresponding to the southern region task group. At the same time, the local Pareto optimal solution set of each computing node is synchronized periodically. Optionally, every preset number of generations, each computing node sends the Pareto optimal solution set of its respective subpopulation to the master coordinator. The master coordinator merges all solution sets and re-executes the non-dominated sorting, and then broadcasts the obtained global Pareto optimal solution set back to each computing node to guide the subsequent search direction.

[0097] Step S523: Determine the first objective function based on the task priority and imaging quality of each task group, and determine the second objective function based on the load of each candidate satellite.

[0098] The first objective function is used to maximize the total task benefit of all task groups, which represents the number of high-priority tasks completed and the imaging quality. Specifically, the first objective function can be constructed using the following formula (14): In the formula, This represents maximizing the total task reward. This indicates the task priority corresponding to the cluster center of the i-th task group. This represents the quality score corresponding to the effective time window allocated to the i-th task group, as detailed in formula (8) above.

[0099] The second objective function is used to minimize the load differences among the candidate satellites in order to balance satellite resources. Specifically, the second objective function can be constructed using the following formula (15): In the formula, This represents the minimum satellite payload difference. Indicates satellite load rate, The standard deviation of the load factor of all satellites is used to characterize the degree of load balancing. The satellite load factor is defined as the proportion of the total energy consumption of a single satellite's assigned tasks to its remaining available energy, i.e., the satellite load factor is determined by the following formula (16): In the formula, This represents the total energy consumption of a satellite that has been assigned tasks. It is determined by summing the attitude maneuver energy consumption and payload imaging energy consumption of each task group corresponding to the satellite. For details, please refer to the explanation of step S513 above. This indicates the satellite's remaining available energy, which is determined based on the difference between the satellite's rated total energy and the total energy consumption of the assigned missions.

[0100] Step S525: Iteratively optimize the initial population based on the first objective function and the second objective function. In each iteration, the parent individuals in the current population are sorted in a non-dominated manner to determine the Pareto front level and the crowding degree of the parent individuals. The parent individuals are selected according to the Pareto front level and the crowding degree of the individual, and crossover and mutation operations are performed on the selected parent individuals to obtain offspring individuals.

[0101] A fast non-dominated sort is performed on the current individuals in the current population. Based on the first and second objective functions, the current individuals are assigned to different Pareto front ranks F1, F2, ..., Fn. The Pareto front is a set of non-dominated solutions representing the optimal trade-offs among multiple objectives. A solution is superior to another solution if it is better on one objective without harming others; this relationship is called dominance. The corresponding Pareto front rank is then determined based on the dominance relationships between individuals. Specifically, for each current individual p, the number n of solutions that dominate p is determined. p And the solution set S dominated by p p Find all n p Individuals with a value of 0 are assigned to the Pareto frontier rank F1. Then, for each individual p in F1, its S... pFor each solution q, n p Decrease by 1 if n p If the value becomes 0, the individual q is added to the next Pareto frontier level F2, and this process is repeated until all individuals are stratified.

[0102] Subsequently, for each individual within each Pareto front level, its individual crowding degree in the target space is determined. Individual crowding degree measures the density of individuals in the target space and is typically obtained by calculating the sum of the distances between the individual and its neighboring individuals in each objective function direction; the greater the distance, the greater the crowding degree. Specifically, the crowding degree can be determined using the following formula (17): In the formula, This represents the crowding degree of the i-th individual. The number of objective functions is indicated in the embodiments of this application. The value is 2. and Let represent the values ​​of the (i+1)th current individual and the ith current individual on the m-th objective function, respectively. and Let represent the maximum and minimum values ​​of all current individuals in the current Pareto frontier on the m-th objective function, respectively.

[0103] It should be noted that the genetic algorithm in this application embodiment favors individuals with large crowding distances in order to maintain the diversity of solutions.

[0104] Next, a binary tournament selection is adopted, prioritizing the current individual with a high Pareto level and high crowding as the parent individual. Among the selected parent individuals, the chromosome gene segments of the two parent individuals are exchanged with a 90% crossover probability to generate offspring individuals. Then, the "satellite ID" or "effective time window ID" gene position of the offspring individual is randomly mutated with a 5% mutation probability to update the offspring individual.

[0105] Step S527: Remove the offspring individuals whose effective time windows are not in the candidate time window set, and perform resource conflict detection on the remaining offspring individuals after removal. Generate the next generation population based on the offspring individuals for which no resource conflict was detected, and perform the next iteration optimization based on the next generation population until the preset iteration stop condition is met.

[0106] Specifically, it checks whether the effective time window selected for the task group in each offspring individual belongs to the pre-determined candidate time window set for that task group. If it does not, the individual is removed. Then, resource conflict detection is performed on the remaining offspring individuals. For details, please refer to the explanation of steps S511 to S515 above, which will not be repeated here. In other words, the offspring individuals after crossover and mutation must satisfy the time window constraint and resource constraint in order to pass the feasibility check.

[0107] Subsequently, the offspring individuals form the next generation population, and the above steps S525 are repeated for iterative optimization until the preset maximum number of iterations is reached or the global Pareto front shows no significant change for a preset number of generations. At this point, the preset iteration stopping condition is met, and the iteration stops.

[0108] Step S529: Determine the optimal satellite allocation scheme based on the individual with the highest Pareto front rank in the final population of the last iteration.

[0109] Specifically, the highest-ranking Pareto front is selected from the Pareto fronts of the final population. If there are multiple individuals in the front, the first objective function and the second objective function are weighted and summed according to the weights preset in the mission scenario, and the individual with the largest weighted value is selected as the optimal satellite allocation scheme.

[0110] This application embodiment uses a non-dominated sorting genetic algorithm to search the Pareto optimal solution set in parallel, using the feasible scheduling scheme after conflict resolution as the initial population. This ensures that the initial population contains at least one feasible solution, avoiding blindly searching for the optimal solution in the infeasible solution space, effectively improving search efficiency. As a result, it can quickly converge to the globally optimal satellite imaging mission planning scheme while ensuring the maximization of mission benefits and the balance of satellite load, effectively improving the quality and efficiency of multi-satellite collaborative mission planning.

[0111] In some embodiments of this application, step S7 may include the following steps: Step S71: The optimal satellite allocation scheme is parsed into a directed acyclic graph consisting of multiple task nodes.

[0112] Specifically, the optimal scheduling scheme output in step S6 is parsed into individual observation tasks, with each observation task serving as a task node. Each observation task includes attributes such as target ID, planned execution time window, satellite ID, and side-slip angle parameters. Based on the temporal dependencies and resource constraints between tasks, directed edges are constructed according to the order of tasks on the same satellite and attitude maneuver times, thus forming a Directed Acyclic Graph (DAG). This graph is stored in the onboard computer as the baseline data structure for online evaluation and replanning.

[0113] Step S73: Acquire on-board real-time sensing data within a preset protection time before each task node begins execution, and conduct an execution risk assessment of the task node based on the on-board real-time sensing data, so as to adjust the directed acyclic graph according to the result of the execution risk assessment.

[0114] Specifically, within the protection time window before each task node is scheduled to begin execution, the system performs the following online risk assessment process: First, using real-time image data acquired by the onboard payload, the real-time cloud cover in the mission group's area is calculated using cloud detection algorithms such as infrared cloud detection based on Otsu adaptive thresholding or cloud segmentation networks based on deep learning. If the cloud cover exceeds a preset maximum cloud cover threshold, it is marked as an unmet observation condition. Simultaneously, real-time measurement data from the gyroscope and star sensor are read to calculate the current deviation angle Δψ between the current attitude and the target attitude in the offline planning scheme. If the current deviation angle Δψ exceeds a preset maximum allowable value... ,like If the remaining power is below the safe power threshold (e.g., 20%), or the remaining storage is below the reserved capacity (e.g., 10% of the total capacity), it is marked as an abnormal posture. In addition, the system reads the real-time remaining power of the power management system and the remaining space of the storage management unit. If the remaining power is below the safe power threshold (e.g., 20%), or the remaining storage is below the reserved capacity (e.g., 10% of the total capacity), it is marked as insufficient resources.

[0115] Based on the above three evaluation results, the system constructs an executability scoring function as shown in the following formula (18): In the formula, Indicates the feasibility score. This indicates the real-time cloud cover in the area where the task group is located. This indicates the preset maximum cloud cover threshold. This indicates the satellite's real-time remaining battery power. Indicates the safe power threshold. , , These represent the weighting coefficients corresponding to real-time cloud cover, current deviation angle, and real-time remaining power, respectively, with values ​​ranging from [0,1].

[0116] When the executability score is lower than the preset executability score threshold, the current task is determined to be unexecutable, thereby triggering the risk assessment process. Specifically, based on the above assessment results, a three-element coupled risk assessment model including environmental risk, task risk, and resource risk is constructed. Among them, environmental risk is comprehensively assessed by real-time cloud cover, changes in solar altitude angle, and space weather events, and can be characterized by the following formula (19): In the formula, Indicates the environmental risk value. This indicates the real-time solar altitude angle in the area where the mission group is located. This represents the flux of high-energy particles in space. It represents a nonlinear function that comprehensively assesses environmental risk values ​​based on real-time cloud cover, real-time solar altitude angle, and space high-energy particle flux.

[0117] The mission risk is comprehensively assessed by attitude deviation and data transmission link status, and can be characterized by the following formula (20): In the formula, Indicates the task risk value. Indicates the communication quality of a data transmission link, such as signal-to-noise ratio or bit error rate. This represents a nonlinear function that comprehensively assesses task risk based on attitude deviation angle and link quality.

[0118] Resource risk is comprehensively assessed by considering the remaining power, the satellite's current remaining storage capacity, and the momentum margin of the attitude control wheels. The resource risk value can be characterized by the following formula (21): In the formula, Indicates the resource risk value. This indicates the satellite's current remaining storage capacity. This represents the momentum margin of the attitude control wheel. It represents a nonlinear function that comprehensively assesses resource risk based on remaining power, current remaining storage capacity of the satellite, and momentum margin of the attitude control wheels.

[0119] Finally, the weighted sum of the three factors yields the comprehensive risk value, which can be represented by the following formula (22): In the formula, This represents the overall risk value. , , These represent the weighting coefficients corresponding to the environmental risk value, task risk value, and resource risk value, respectively.

[0120] When the overall risk value exceeds the preset risk threshold, the adjustment planning process for the directed acyclic graph is initiated: When it is determined that the planning process for a directed acyclic graph needs to be adjusted, the system performs a partial adjustment according to the following procedure: If only environmental risks such as excessive cloud cover exist, and a backup time window exists for the current task, the current time window is skipped, and the task is postponed to the next available window. If no available time window exists for the current task, a backup target that meets the current observation conditions and has sufficient resources is selected from the pre-stored list of backup targets to replace the task. If the task must be executed but conditions are partially degraded, payload parameters are dynamically adjusted, such as reducing resolution to reduce data volume and energy consumption, or shortening integration time to reduce power consumption. If the above adjustments are feasible, the directed acyclic graph is updated accordingly. If none of the above adjustments are feasible, the current task node and its associated edges are removed from the directed acyclic graph, the remaining subgraph is re-topologically sorted to update the execution schedule of subsequent tasks, and the changes are synchronized to the ground control center via inter-satellite links.

[0121] Step S75: Generate control commands for each satellite based on the adjusted directed acyclic graph, and distribute the control commands to each satellite to perform the imaging task.

[0122] Specifically, the system maintains a database containing protocol templates for various satellite models through an instruction generation engine. Based on the satellite identifier corresponding to each task node in the adjusted directed acyclic graph, it automatically loads the corresponding interface control document definition. Then, it maps high-level scheduling parameters such as the task execution time window and yaw angle to low-level instruction fields, automatically calculates time stamps and fills reserved bits, and assembles them into binary control instruction packets according to the prescribed frame structure, such as header synchronization word, length, data area, checksum, and tail synchronization word. Finally, it distributes the control instructions to the corresponding satellites for execution via satellite-to-ground links or inter-satellite communication networks.

[0123] This application embodiment, through directed acyclic graph analysis and online risk assessment during the execution process, can dynamically adjust the mission planning scheme and generate control commands adapted to each satellite, thereby realizing online autonomous planning and closed-loop management of multi-satellite collaborative imaging missions.

[0124] Accordingly, please refer to Figure 2 This application provides a multi-target point imaging task planning device, which includes: The satellite allocation module 100 is used to divide the target points into several task groups according to the spatial distribution characteristics of each target point, and to determine the candidate satellites corresponding to each task group. For details, please refer to step S1. The window filtering module 200 is used to determine the set of candidate time windows that meet the preset imaging feasibility conditions for the candidate satellites corresponding to each task group, and select the effective time window corresponding to the task group from the set of candidate time windows. For details, please refer to step S3. The planning module 300 is used to detect resource conflicts between candidate satellites and effective time windows for all task groups. If no resource conflicts are detected, it iteratively optimizes the allocation relationship between each task group and candidate satellites and effective time windows to obtain the optimal satellite allocation scheme for each target point. For details, please refer to step S5. The execution module 400 is used to control each candidate satellite to perform imaging tasks according to the optimal satellite allocation scheme. For details, please refer to step S7.

[0125] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0126] In this embodiment, the multi-target imaging task planning device is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0127] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application, such as... Figure 3 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 3 Take a processor 10 as an example.

[0128] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0129] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0130] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0131] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0132] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0133] This application also provides a computer-readable storage medium. The methods described in this application can be implemented in hardware or firmware, or implemented as recordable on a storage medium, or implemented as computer code downloaded over a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and subsequently stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the methods shown in the above embodiments are implemented.

[0134] This application provides a computer program product including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the method of any embodiment of this application.

[0135] Although embodiments of this application have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of this application, and all such modifications and variations fall within the scope defined by the appended claims.

[0136] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0137] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0138] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0139] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0140] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1The steps of the function specified in one or more boxes.

[0141] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0142] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the apparatus embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0143] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

[0144] Although embodiments of this application have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of this application, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A multi-target imaging task planning method based on multi-satellite collaboration, characterized in that, The method includes: Based on the spatial distribution characteristics of each target point, the target points are divided into several task groups, and candidate satellites corresponding to each task group are determined. For each of the candidate satellites corresponding to the task groups, a set of candidate time windows that meet the preset imaging feasibility conditions is determined, and the effective time window corresponding to the task group is selected from the set of candidate time windows. Resource conflict detection is performed on the candidate satellites and effective time windows corresponding to all the task groups. If no resource conflict is detected, the candidate satellites and effective time windows corresponding to each task group are iteratively optimized to obtain the optimal satellite allocation scheme for each target point. The candidate satellites are controlled to perform imaging tasks according to the optimal satellite allocation scheme.

2. The method according to claim 1, characterized in that, The step of dividing the target points into several task groups based on the spatial distribution characteristics of each target point includes: Determine the preset task priority corresponding to each of the target points; The spatial distribution features are determined based on the relative distance and local density of each target point, and the spatial distribution features are weighted according to the preset task priority, so as to determine the cluster center based on the weighted spatial distribution features. The cluster radius is determined based on the maximum ground coverage width of all satellites; The target point is divided into several task groups based on the cluster center and the cluster radius.

3. The method according to claim 1, characterized in that, The process of determining the candidate satellites corresponding to each of the mission groups includes: Calculate the geometric visibility relationship and observation quality between each task group and each candidate satellite based on satellite orbit prediction information; Based on the geometric visibility relationship, a weighted bipartite graph is constructed with the task group as the left node and the satellite as the right node, wherein the weight of each edge in the weighted bipartite graph is determined according to the observation quality; The weighted bipartite graph is solved using the maximum weight matching algorithm to determine the candidate satellites corresponding to each task group based on the solution results.

4. The method according to claim 1, characterized in that, The step of determining a set of candidate time windows that meet preset imaging feasibility conditions for each candidate satellite corresponding to each task group, and selecting the effective time window corresponding to the task group from the set of candidate time windows, includes: Obtain the common-view time window in which the candidate satellites can cover all target points within the corresponding task group; The start time of the common-view time window is corrected based on the satellite attitude maneuver time of the candidate satellites to obtain a corrected visible time window. The imaging quality parameters of the candidate satellites for the mission group are obtained under the modified visible time window, and the modified visible time window that meets the preset imaging feasibility conditions is selected as a candidate time window based on the imaging quality parameters to form the candidate time window set. The candidate time windows are sorted according to the imaging quality parameters, and the effective time window is determined based on the sorting results.

5. The method according to claim 4, characterized in that, The step of performing resource conflict detection on all candidate satellites and effective time windows corresponding to all the aforementioned task groups includes: Obtain the remaining energy and remaining storage of the candidate satellite at the start of the effective time window; The estimated total energy consumption is determined based on the attitude maneuver energy consumption and payload imaging energy consumption of the candidate satellite when performing the corresponding task group, and the estimated data volume is determined when the candidate satellite performs the imaging task of the corresponding task group. The estimated total energy consumption is compared with the remaining energy to obtain a first comparison result, and the estimated data volume is compared with the remaining storage volume to obtain a second comparison result, so as to determine whether there is a resource conflict based on the first comparison result and the second comparison result.

6. The method according to claim 5, characterized in that, If the first comparison result does not meet the first preset condition or the second comparison result does not meet the second preset condition, it is determined that a resource conflict has been detected. In the event of a detected resource conflict, the method further includes: Resource conflicts are adjusted based on a preset conflict resolution strategy, which includes: The candidate time windows are traversed according to the sorting result, and the time windows of the task group are adjusted according to the traversed candidate time windows. If there are no resource conflicts after the time window adjustment, the preset conflict resolution strategy is terminated. If resource conflicts still exist after traversing all the candidate time windows and adjusting the time windows, then reduce the imaging parameters of the task group. If there are no resource conflicts after reducing the imaging parameters, then end the preset conflict resolution strategy. If resource conflicts still exist after reducing imaging parameters, the task group will be moved to other satellites. If there are no resource conflicts after the move, the preset conflict resolution strategy will be terminated. If resource conflicts persist after migration, the task group will be temporarily removed.

7. The method according to claim 1, characterized in that, The iterative optimization of the candidate satellites and effective time windows corresponding to each of the task groups to obtain the optimal satellite allocation scheme for each of the target points includes: An initial population is generated based on the candidate satellites currently assigned to each of the aforementioned task groups and the effective time window; A first objective function is determined based on the task priority and imaging quality of each of the aforementioned task groups, and a second objective function is determined based on the load of each of the aforementioned candidate satellites; The initial population is iteratively optimized based on the first objective function and the second objective function. In each iteration, the parent individuals in the current population are sorted in a non-dominated manner to determine the Pareto front level and the crowding degree of the parent individuals. The parent individuals are selected according to the Pareto front level and the crowding degree of the individuals, and crossover and mutation operations are performed on the selected parent individuals to obtain offspring individuals. The offspring individuals whose effective time windows are not in the candidate time window set are removed, and resource conflict detection is performed on the remaining offspring individuals after removal. The next generation population is generated based on the offspring individuals for which no resource conflict is detected, and the next iteration optimization is performed based on the next generation population until the preset iteration stop condition is met. The optimal satellite allocation scheme is determined based on the individual with the highest Pareto front rank in the final population of the last iteration.

8. The method according to claim 1, characterized in that, The step of controlling each of the candidate satellites to perform imaging tasks according to the optimal satellite allocation scheme includes: The optimal satellite allocation scheme is parsed as a directed acyclic graph consisting of multiple task nodes; Before each of the task nodes begins execution, real-time on-board sensing data is acquired within a preset protection time, and the execution risk assessment of the task node is performed based on the real-time on-board sensing data, so as to adjust the directed acyclic graph according to the result of the execution risk assessment. Control commands for each of the satellites are generated based on the adjusted directed acyclic graph, and the control commands are distributed to each of the satellites to perform the imaging task.

9. A multi-target imaging task planning device based on multi-satellite collaboration, characterized in that, The device includes: The satellite allocation module is used to divide the target points into several task groups according to the spatial distribution characteristics of each target point, and to determine the candidate satellites corresponding to each task group. The window filtering module is used to determine a set of candidate time windows that meet the preset imaging feasibility conditions for the candidate satellites corresponding to each task group, and to select the effective time window corresponding to the task group from the set of candidate time windows. The planning module is used to perform resource conflict detection on the candidate satellites and effective time windows corresponding to all the task groups, and if no resource conflict is detected, iteratively optimize the allocation relationship between each task group and the candidate satellites and the effective time window to obtain the optimal satellite allocation scheme for each target point. The execution module is used to control each of the candidate satellites to perform imaging tasks according to the optimal satellite allocation scheme.

10. A computer device, characterized in that, include: The system includes a memory and a processor, which are interconnected. The memory stores computer instructions, and the processor executes the computer instructions to perform the multi-target imaging task planning method based on multi-satellite collaboration as described in any one of claims 1 to 8.

Images