A ground station resource scheduling method and device based on an improved simulated annealing algorithm

Abstract

The application relates to the technical field of multi-ground station resource scheduling, and particularly discloses a ground station resource scheduling method and device based on an improved simulated annealing algorithm. Radar ground stations and optical ground stations are uniformly modeled, observation arcs are parameterized, state vectors of space targets are defined, and space target observation results of the radar ground stations and the optical ground stations are expressed in a unified language system; a radar and optical integrated resource scheduling model is constructed, condition constraints and task execution constraints are determined, and a radar and optical ground station resource scheduling objective function is constructed; based on the improved simulated annealing algorithm, an initial allocation scheme of the radar and optical ground station resource scheduling is acquired, optimization is carried out, and simulation results are obtained; the simulation results are analyzed, a Gantt chart is drawn, and a ground station resource scheduling scheme is output. Efficient integrated scheduling of the radar and optical ground station resources is realized, the observation task completion rate is improved, and the calculation time is shortened.

Description

Technical Field

[0001] This invention relates to the field of multi-ground station resource scheduling technology, specifically to a ground station resource scheduling method and apparatus based on an improved simulated annealing algorithm. Background Technology

[0002] With the increasing frequency of space launches, the number of categorizable space targets is growing explosively. Current ground-based space target observation equipment mainly consists of optical and radar ground stations. Existing ground station resource scheduling methods (such as traditional manual scheduling and genetic algorithms) suffer from low task completion rates and insufficient computational efficiency under complex constraints. This is especially true when simultaneously satisfying performance constraints (such as maximum detection range, elevation angle, and solar altitude angle) and task constraints (such as minimum observation time and simultaneous tracking limit) for both radar and optical ground stations. Inefficient allocation of observation resources is difficult, often leading to the consideration of only a single type of ground station. Ground stations face three main challenges in space target observation: 1) Complex constraints: Optical and radar ground stations operate on different principles, resulting in different constraints and making modeling and solution processes difficult; 2) Huge data volume: The relationship between ground stations and space targets is not a simple one-to-one relationship. An increase in the number of ground stations and space targets leads to an exponential increase in the amount of data generated in the observation arc, resulting in low computational efficiency; 3) Complex multi-target optimization problems: While solving for a single target is relatively simple, considering multiple targets and solving under multiple complex constraints often leads to getting trapped in local optima. Therefore, studying the resource scheduling problem of ground stations for space target observation is of great significance. Summary of the Invention

[0003] To address the aforementioned problems, the first objective of this invention is to provide a ground station resource scheduling method based on an improved simulated annealing algorithm. This method aims to achieve efficient integrated scheduling of radar and optical ground station resources, improving the completion rate of observation tasks and shortening computation time. It also solves the problem that the rapidly increasing number of catalogable space targets such as satellites and space debris in the current environment makes traditional manual resource scheduling methods far from meeting the ever-growing cataloging demands, leading to increased target observation conflicts and heavier equipment loads.

[0004] The second objective of this invention is to provide a ground station resource scheduling device based on an improved simulated annealing algorithm.

[0005] To achieve the first objective, the first technical solution of this invention is: a ground station resource scheduling method based on an improved simulated annealing algorithm, comprising: Step S100: Perform unified modeling of radar ground station and optical ground station, parameterize the observation arc segment, define the state vector of space target, and express the space target observation results of radar ground station and optical ground station in a unified language system; Step S200: Construct an integrated radar and optical resource scheduling model, determine conditional constraints and task execution constraints, and construct the objective function for radar and optical ground station resource scheduling; Step S300: Based on the improved simulated annealing algorithm, obtain the initial allocation scheme for radar and optical ground station resource scheduling and optimize it to obtain simulation results; Step S400: Analyze the simulation results, draw a Gantt chart, and output the ground station resource scheduling scheme.

[0006] Preferably, in step S100, the parameterization is described as follows: in, Indicates the azimuth coverage area. Indicates the range of pitch angles. Indicates the observed arc segment.

[0007] Preferably, in step S100, the space target observation results of the radar ground station and the optical ground station are expressed in a unified language system in the following manner: ; in, Indicates the observed arc segment. This indicates that the sensors at a radar ground station or optical ground station are observing space targets. Indicates space targets, Indicates the start time of the observation. Indicates the end time of observation. Indicates the duration of the arc segment.

[0008] Preferably, in step S200, the condition constraints include: in, τ represents the minimum observation time for the satellite, and τ represents the observation arc segment for the satellite; Indicates the elevation angle observed at the ground station; Indicates the azimuth angle observed from the ground station. This indicates the solar altitude angle observed by the optical ground station; This indicates the detection range of the ground station when detecting the satellite.

[0009] Preferably, in step S200, the task execution constraints include: ; in, This indicates the number of targets tracked by the sensors of the optical ground station; This indicates the number of targets tracked by the sensors at the radar ground station.

[0010] Preferably, in step S200, the objective function expression for radar and optical ground station resource scheduling is constructed as follows: in, Indicates the length of the observation arc performed by the ground station; τ represents the selection function; τ is the observation arc segment of the satellite.

[0011] Preferably, step S300 further includes establishing an observation conflict resolution method based on a penalty function to reduce observation redundancy.

[0012] Preferably, in step S400, the ground station resource scheduling scheme includes: the total number of targets with observation arcs, the number of effective observation targets, the number of unobserved targets, the observation completion rate, and a curve of scheduling scheme generation time and fitness value.

[0013] To achieve the second objective, the second technical solution of the present invention is: a ground station resource scheduling device based on an improved simulated annealing algorithm, comprising: Observation arc acquisition module: Based on the predetermined orbital dynamics model, the ground station equipment capabilities, and the position and velocity of the space target, the observation arc is acquired under the condition of satisfying the constraints. Constraint Integration Module: Integrates the constraints of the radar ground station and the optical ground station at the input end and outputs a unified constraint set; Ground station initial mission planning module: Reads the acquired observation arc segments and obtains an initial resource scheduling scheme based on the improved simulated annealing algorithm; Observation conflict reduction module: Based on the penalty function conflict resolution method, it avoids the observation redundancy of multiple ground stations for a certain target and generates an optimized resource scheduling scheme; Resource scheduling scheme visualization module: The resource scheduling scheme after conflict reduction is visualized using Gantt charts.

[0014] Preferably, the observation conflict reduction module is also used to generate a CSV file from the resource scheduling scheme and store it.

[0015] The beneficial effects of the above technical solution are as follows: This invention provides a ground station resource scheduling method and apparatus based on an improved simulated annealing algorithm. Through an integrated ground station resource model, the observation representation methods for radar and optical ground stations are clarified, fully considering the similarities and differences in their observation processes and the constraints they face, thus achieving collaborative scheduling between radar and optical ground stations and maximizing the utilization of observation resources. This invention obtains a resource scheduling scheme for collaborative observation of space targets by radar and optical ground stations through an improved simulated annealing algorithm, improving the resource scheduling efficiency of the scheduling system and shortening the scheduling scheme generation time, effectively saving human resources. This invention also establishes a method for reducing observation conflicts between multiple ground stations observing space targets, avoiding observation conflicts between ground stations. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart of a ground station resource scheduling method based on an improved simulated annealing algorithm is provided as an embodiment of the present invention; Figure 2 A schematic diagram of a ground station resource scheduling device based on an improved simulated annealing algorithm provided for an embodiment of the present invention; Figure 3 A flowchart illustrating the operation of an improved simulated annealing algorithm provided in an embodiment of the present invention. Detailed Implementation

[0018] The embodiments of this application will be described in further detail below. Obviously, the described embodiments are only a part of the embodiments of this application, and not an exhaustive list of all embodiments. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other.

[0019] The terms “first,” “second,” etc. (if applicable) in the specification and claims are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated (if applicable) or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion, such as a process, method, system, product, or apparatus that comprises a series of steps or units, not necessarily limited to those explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0020] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0021] Example 1 To address the inefficiency and low task completion rate of existing methods under complex constraints, this invention proposes a ground station resource scheduling method based on an improved simulated annealing algorithm. This method aims to achieve efficient integrated scheduling of radar and optical ground station resources, thereby improving the observation task completion rate and reducing computation time. The method includes the following steps: Step S100: Based on the modeling of radar and optical ground stations, an integrated resource scheduling description method is established, using a unified observation arc to describe the observation state, and expressing the observation results of radar and optical ground stations on space targets under a unified language system; Step S200: An integrated radar and optical resource scheduling model is established, determining the conditional constraints and task execution constraints of ground station observation of space targets, and constructing the objective function; Step S300: An improved simulated annealing algorithm is used to obtain the initial allocation scheme of ground station resource scheduling, and an observation conflict resolution method based on a penalty function is established to reduce observation redundancy; Step S400: The simulation results are analyzed, and a Gantt chart is drawn to output the resource scheduling scheme.

[0022] Step S100 specifically includes: Step S110: When observing space targets from a ground station, the main objects involved are the ground station's sensors and the space targets. First, a unified model is performed on the observation resources, namely radar and optical ground stations, defining the sensor set: (1) Each sensor belongs to both radar and optics: (2) Step S120: Parameterize the observed arc segment for any sensor: (3) in, This refers to the azimuth coverage area. For pitch angle range, For the observation arc segment, the above parameters are described in detail in S200.

[0023] Step S130: Space targets all move along certain trajectories and have position and velocity vectors. Define the state vector of the space target: (4) in, Let be the position vector of the space target in the ECI coordinate system. Let be the velocity vector of the space target in this coordinate system.

[0024] Step S140, Unify the observation equations: (5) (6) The constraints considered in radar and optical observations include elevation angle, azimuth angle, and maximum observation range. Radar ground stations can conduct observations around the clock and in all weather conditions, while optical sensors are limited by weather conditions and are represented by the solar altitude angle θ.

[0025] Coordinate unification: (7) The observation equations for radar and optical sensors can be uniformly expressed as: (8) Step S150: Express the observation results under a unified language system: While the observation processes of radar and optical ground stations differ, and the constraints they involve also differ, the observation results can be uniformly expressed as observation arcs, allowing for unified reading and retrieval. The included elements are represented as follows: (9) A certain observation arc is described as an observation of a certain space target by a certain sensor, starting from a certain time and ending at a certain time. The duration of the arc is a certain duration, and the unit is seconds.

[0026] Step S200 specifically includes: Step S210: Construct the conditional constraints of the integrated resource scheduling model for radar and optical ground stations. The expression for the conditional constraints is as follows: (10) (11) (12) (13) (14) Among them, constraint (10) means that the minimum observation time for the satellite should be greater than the value τ; constraints (11) and (12) mean that for the ground station, the required elevation and azimuth angles during observation should be higher than the minimum elevation and azimuth angles required for detection, and also lower than the maximum elevation and azimuth angles; the optical ground station should also meet the minimum and maximum solar altitude angle conditions of constraint (13); constraint (14) means that there is a maximum detection distance for the ground station when detecting the satellite, and the observation arc can only be generated when the satellite is within the maximum detection distance.

[0027] Step S220: Define task execution constraints. In addition to the conditional constraints that need to be satisfied, observation task constraints should also be considered, mainly described as follows: (15) (16) Among them, constraint (15) means that the sensor of the optical ground station can only track one target at a time; constraint (16) means that the sensor of the radar ground station can track up to three targets at a time.

[0028] Step S230: Construct the objective function for resource scheduling of radar and optical ground stations. The objective function is determined by the length of the observation arc performed by all ground stations. Selection function The composition, specifically the expression, is: (17) Equation (17) represents the possible number of space target detections completed, where, The selection function has a value of 0 or 1; a value of 1 indicates that observations are performed, and a value of 0 indicates that observations are not performed. τ represents the observation arc of the satellite, in seconds. (Total score) It's a reciprocal value, so the smaller the better.

[0029] Step S300 specifically includes: Step S310: Improve the parameter tuning of the simulated annealing algorithm. The operation steps of the classic simulated annealing algorithm are as follows: 1) Initialization: Set the initial temperature T, the initial solution F, and the number of iterations N for each temperature value; 2) A new solution F' is generated near the current solution F; 3) Calculation , here This represents the objective function, which simulates energy. 4) Judgment Criterion (Metropolis Criterion): Judgment If the value is greater than 0, accept the new current solution; otherwise, accept the new current solution with probability e(-ΔT / T). 5) Determine if the termination condition is met. If it is, output the current solution as the optimal solution and end the iteration; otherwise, repeat steps 2) to 5).

[0030] However, the classic simulated annealing algorithm suffers from problems such as easily getting trapped in local optima, long running time, and relatively low target observation rate when running the models constructed by S210 and S220. Therefore, the parameters of the classic simulated annealing algorithm were improved: the initial temperature was set to 1000, a rapid cooling strategy was adopted, and ReannealInterval was set to 50 to periodically reset the temperature to avoid getting trapped in local optima and improve the quality of the solution. The cooling strategy expression is: (18) Step S320, fitness function redesign: The fitness value is set to balance the score and penalty value, with the penalty term set within a reasonable range to constrain invalid solutions while avoiding excessive interference. The cumulative score calculation rule within a single for loop is as follows: (19) in This is the penalty weighting coefficient. It is set based on the magnitude relationship between conflict penalty items and task completion. To match the magnitude of the main term of the objective function (task completion), when the number of observations is 1000, the total observation time should be on the order of 10. 5 The penalty term P is on the order of 10. 2 ,but The order of magnitude is 10 3 .

[0031] Step S330, reduce computational complexity: convert continuous variables into binary codes to simplify problem complexity; assuming the number of tasks is n, the computational complexity of the traditional brute-force comparison strategy is n... 2 By sorting the observed arc segments according to their start time, conflict detection can be completed in just one traversal, reducing the time complexity to n*(1+lgn), which is much smaller than the traditional computational complexity.

[0032] The original collision detection complexity is: (20) The improved computational complexity includes sorting complexity and collision detection complexity, with a total complexity of: (twenty one) Taking n=1000 as an example, the original computational complexity is 10. 6 The improved computational complexity is approximately 10. 4 This can reduce the number of items by about two orders of magnitude.

[0033] Step S340: Design a penalty function to avoid observation conflicts.

[0034] (twenty two) (twenty three) Equation (22) represents the collision penalty for radar ground stations, and Equation (23) represents the collision penalty for optical ground stations. Wherein, Penalties for conflicts involving radar ground stations; For the optical ground station conflict penalty, overflow represents the value exceeding the maximum number of simultaneous observations. As can be seen from equation (17), the value of the objective function is to approach the minimum positive number as much as possible. The value of the penalty term is positive. When the penalty value increases, the value of the objective function will also increase. Therefore, in order to make the objective function approach 0 during the algorithm iteration process, conflicts will be reduced as much as possible. This also ensures the global reachability after the introduction of the penalty term and the absence of periodic loops.

[0035] Step S350: Design a penalty function for simultaneous observation of space targets to reduce observation redundancy. (twenty four) observed s This represents the number of sensors performing real-time observations of a target. When multiple sensors observe the same target simultaneously, the penalty term increases, forcing the algorithm to avoid repeated observations during iterations, thereby reducing observation redundancy.

[0036] Step S400 specifically includes: Step S410: Analyze the simulation results: Output elements include: total number of targets with observation arcs, number of effective observed targets, number of unobserved targets, observation completion rate, scheduling scheme generation time, and fitness value curve.

[0037] The total number of targets with observation arcs depends on the relative positions of the targets and ground stations. Typically, some targets will lack observation arcs, requiring an increase in the number of ground stations to mitigate this. The number of effective observation targets is the number of targets that can be effectively observed by ground stations after data processing and mission planning using the method of this invention. The number of unobserved targets is the number of targets with observation arcs but not planned for ground station observation due to potential limitations of the device itself. However, the number of unobserved targets is usually not zero, as ground station resource utilization may be excessively high, meaning that the observation demand generated by an excessive number of targets cannot be met under existing ground station resource conditions.

[0038] The goal of this invention is to minimize the number of unobserved targets, i.e., maximize the number of effective observations. The observation completion rate is the ratio of the number of effective observed targets to the total number of targets with observation arcs. If the observation completion rate of this algorithm is better than other algorithms under the same conditions, this indicator of the algorithm is considered superior, and the improvement is effective. The scheduling scheme generation time reflects the timeliness of the device in performing resource scheduling tasks. If the scheduling scheme generation time is too long, the scheduling requirements cannot be met. The fitness curve reflects whether the iteration process is basically normal and whether the fitness value iterates in the target direction. Under normal circumstances, the fitness curve of this invention decreases and tends to stabilize and gradually become a straight line when it reaches a certain value.

[0039] Step S420, output scheduling scheme: On the one hand, the method can generate a CSV file of the scheduling scheme for easy access at any time, and the data format is also displayed according to the unified data elements in formula (9); on the other hand, the invention also provides a device to support the visualization of resource scheduling schemes, and to represent the scheduling scheme in the form of a Gantt chart.

[0040] A ground station resource scheduling device based on an improved simulated annealing algorithm includes: Observation arc acquisition module: Based on the predetermined orbital dynamics model, such as J2, SGP4, etc., and considering the capabilities of ground station equipment and the position and velocity of the space target, the observation arc is acquired under the condition of satisfying the constraints. Constraint Integration Module: The observation processes of radar and optical ground stations differ, and the constraints involved also differ. The module integrates the constraints at the input end and outputs a unified constraint set. Ground station mission planning module: Reads the acquired observation arc segments and obtains an initial resource scheduling scheme based on the improved simulated annealing algorithm; Observation conflict reduction module: A penalty function conflict resolution method was established to avoid redundant observations of a certain target by multiple ground stations. After conflict resolution, an optimized resource scheduling scheme is generated. Resource scheduling scheme visualization module: The resource scheduling scheme after conflict reduction is visualized using Gantt charts.

[0041] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. For those skilled in the art, other variations or modifications can be made based on the above description. It is impossible to exhaustively list all the implementation methods here. All obvious variations or modifications derived from the technical solutions of the present invention are still within the protection scope of the present invention.

Claims

1. A ground station resource scheduling method based on an improved simulated annealing algorithm, characterized in that, include: Step S100: Perform unified modeling of radar ground station and optical ground station, parameterize the observation arc segment, define the state vector of space target, and express the space target observation results of radar ground station and optical ground station in a unified language system; Step S200: Construct an integrated radar and optical resource scheduling model, determine conditional constraints and task execution constraints, and construct the objective function for radar and optical ground station resource scheduling; Step S300: Based on the improved simulated annealing algorithm, obtain the initial allocation scheme for radar and optical ground station resource scheduling and optimize it to obtain simulation results; Step S400: Analyze the simulation results, draw a Gantt chart, and output the ground station resource scheduling scheme.

2. The ground station resource scheduling method based on the improved simulated annealing algorithm according to claim 1, characterized in that, In step S100, the parameterization is described as follows: in, Indicates the azimuth coverage area. Indicates the range of pitch angles. Indicates the observed arc segment.

3. The ground station resource scheduling method based on the improved simulated annealing algorithm according to claim 1, characterized in that, In step S100, the space target observation results of the radar ground station and the optical ground station are expressed in a unified language system in the following manner: ； in, Indicates the observed arc segment. This indicates that the sensors at a radar ground station or optical ground station are observing space targets. Indicates space targets, Indicates the start time of the observation. Indicates the end time of observation. Indicates the duration of the arc segment.

4. The ground station resource scheduling method based on the improved simulated annealing algorithm according to claim 1, characterized in that, In step S200, the condition constraints include: in, τ represents the minimum observation time for the satellite, and τ represents the observation arc segment for the satellite; Indicates the elevation angle observed at the ground station; Indicates the azimuth angle observed from the ground station. This indicates the solar altitude angle observed by the optical ground station; This indicates the detection range of the ground station when detecting the satellite.

5. The ground station resource scheduling method based on the improved simulated annealing algorithm according to claim 1, characterized in that, In step S200, the task execution constraints include: ； in, This indicates the number of targets tracked by the sensors of the optical ground station; This indicates the number of targets tracked by the sensors at the radar ground station.

6. The ground station resource scheduling method based on the improved simulated annealing algorithm according to claim 1, characterized in that, In step S200, the objective function expression for radar and optical ground station resource scheduling is constructed as follows: in, Indicates the length of the observation arc performed by the ground station; τ represents the selection function; τ is the observation arc segment of the satellite.

7. The ground station resource scheduling method based on the improved simulated annealing algorithm according to claim 1, characterized in that, Step S300 also includes establishing an observation conflict resolution method based on a penalty function to reduce observation redundancy.

8. The ground station resource scheduling method based on the improved simulated annealing algorithm according to claim 1, characterized in that, In step S400, the ground station resource scheduling scheme includes: the total number of targets with observation arcs, the number of effective observation targets, the number of unobserved targets, the observation completion rate, and a curve of scheduling scheme generation time and fitness value.

9. A ground station resource scheduling device based on an improved simulated annealing algorithm, characterized in that, include: Observation arc acquisition module: Based on the predetermined orbital dynamics model, the ground station equipment capabilities, and the position and velocity of the space target, the observation arc is acquired under the condition of satisfying the constraints. Constraint Integration Module: Integrates the constraints of the radar ground station and the optical ground station at the input end and outputs a unified constraint set; Ground station initial mission planning module: Reads the acquired observation arc segments and obtains an initial resource scheduling scheme based on the improved simulated annealing algorithm; Observation conflict reduction module: Based on the penalty function conflict resolution method, it avoids the observation redundancy of multiple ground stations for a certain target and generates an optimized resource scheduling scheme; Resource scheduling scheme visualization module: The resource scheduling scheme after conflict reduction is visualized using Gantt charts.

10. The ground station resource scheduling device based on the improved simulated annealing algorithm according to claim 9, characterized in that, The observation conflict reduction module is also used to generate a CSV file from the resource scheduling scheme and store it.

Images