A method for optimizing large-scale constellation orbital cooperative maneuvers
By optimizing the Lambert orbit change and dynamic programming algorithms using simulated annealing, the problem of high fuel consumption in space debris cleanup was solved, achieving fuel savings and improved mission efficiency.
Patent Information
- Application Number
- CN202411161519.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-23
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2044-08-23
AI Technical Summary
Existing space debris removal technologies consume a lot of fuel when the difference in orbital dihedral angle is too large or the initial positions are far apart, making it difficult to efficiently remove multiple pieces of space debris.
The simulated annealing algorithm is used to optimize the Lambert orbit change method. Combined with the dynamic programming algorithm, the matching scheme between satellites and space debris is optimized to ensure that each satellite removes only one piece of debris, thereby reducing fuel consumption.
It significantly reduced speed pulse consumption during orbit change, lowered fuel costs, improved mission execution efficiency and success rate, and achieved optimal resource allocation.
Smart Images

Figure CN119089677B_ABST
Abstract
Description
Technical Field
[0001] This relates to the field of air-ground hybrid multi-space debris removal technology, specifically to air-ground hybrid multi-space debris removal task planning based on simulated annealing algorithm. Background Technology
[0002] With the development of science and technology and the progress of the times, large-scale satellite constellations are receiving increasing attention from countries around the world. Large-scale satellite constellations in planning and in operation are constantly emerging, and more and more small satellites are being launched into space. Among the more prominent large-scale constellations are Elon Musk's Starlink and China's BeiDou. Space debris, generated by human activities, is space junk in outer space. Space debris can collide with satellites in orbit, interfering with the normal operation of satellite constellations. As more and more satellites are launched into outer space, the space debris generated by these satellites is also constantly increasing, significantly impacting their normal operation. Suitable orbital resources are therefore becoming increasingly precious. Therefore, using appropriate methods to remove space debris is extremely important.
[0003] Extensive research has been conducted on space debris removal technologies, and several relatively mature, in-orbit validated technologies exist, such as space rope-net capture technology and robotic arm capture technology. These technologies require satellites carrying removal devices to be launched into space. To conserve resources, each satellite can perform multiple orbital maneuvers and capture missions until its fuel is depleted. Therefore, conserving satellite fuel and enabling a single satellite to remove multiple pieces of space debris is paramount for clearing more space debris. Many scholars have explored the issue of fuel conservation for single satellites. However, there are optimization limits to a single satellite targeting multiple satellites. When the dihedral angles of two orbits differ significantly or their initial positions are far apart, even the best optimized result will consume a considerable amount of fuel. Summary of the Invention
[0004] To address the existing problems in space debris removal technologies, which are generally mature and verified in orbit, all of which rely on a single satellite targeting multiple satellites, this invention provides the following technical solution:
[0005] A large-scale constellation orbital cooperative maneuver optimization method, applied to air-ground hybrid space debris removal missions, includes:
[0006] The steps to optimize the acquisition algorithm;
[0007] The steps for setting the simulated annealing algorithm parameters according to the optimization objective;
[0008] The step of collecting a randomly generated initial solution as the starting point of the simulated annealing algorithm;
[0009] The step involves iteratively optimizing the initial solution until a combination of trajectory change parameters under preset conditions is obtained.
[0010] Furthermore, a preferred implementation is provided, wherein the algorithm is the Lambert algorithm.
[0011] Furthermore, a preferred implementation is provided, wherein the optimization objectives include orbit change time, rendezvous time, and number of rendezvous cycles.
[0012] Furthermore, a preferred embodiment is provided, wherein the parameters include initial temperature, termination temperature, cooling rate, and number of iterations at each temperature.
[0013] Based on the same inventive concept, this invention also provides a large-scale constellation orbital cooperative maneuver optimization device, applied to air-ground hybrid space debris removal missions, comprising:
[0014] The module for optimizing the acquisition algorithm;
[0015] A module for setting the parameters of the simulated annealing algorithm based on the optimization objective;
[0016] The module that collects randomly generated initial solutions as the starting point of the simulated annealing algorithm;
[0017] The module iteratively optimizes the initial solution until a combination of trajectory change parameters under preset conditions is obtained.
[0018] Based on the same inventive concept, this invention also provides a method for matching air-ground hybrid space debris removal schemes, including:
[0019] The steps for combining the acquired orbit change parameters;
[0020] The process involves matching individual satellites with individual space debris based on the orbit change parameter combination, after each satellite has cleared only one piece of space debris.
[0021] Based on the same inventive concept, the present invention also provides a matching device for air-ground hybrid space debris removal schemes, comprising:
[0022] A module for collecting and combining orbit change parameters;
[0023] This module, based on the principle of clearing only one piece of space debris per satellite, matches individual satellites with individual pieces of space debris according to the aforementioned combination of orbit change parameters.
[0024] Based on the same inventive concept, the present invention also provides a computer storage medium for storing a computer program, wherein when the computer program is read by a computer, the computer executes the method described thereon.
[0025] Based on the same inventive concept, the present invention also provides a computer, including a processor and a storage medium, wherein when the processor reads a computer program stored in the storage medium, the computer executes the method described thereon.
[0026] Based on the same inventive concept, the present invention also provides a computer program product, which, when executed, implements the method described.
[0027] Compared with the prior art, the advantages of the technical solution provided by the present invention are as follows:
[0028] The introduction of simulated annealing effectively solves the global optimization problem of the Lambert orbital change problem. Specifically, by gradually reducing the temperature and accepting suboptimal solutions, the simulated annealing algorithm avoids the limitation of traditional Lambert orbital change methods that are prone to getting trapped in local optima. This improvement significantly reduces velocity pulsations during the orbital change process, thereby reducing fuel consumption. Compared to directly using the Lambert method, the optimized effect is more significant.
[0029] Dynamic programming algorithms optimize the matching problem between satellites and space debris. By rationally designing the state transition equations and boundary conditions, dynamic programming ensures that the total removal cost is minimized under the constraint that each satellite can only remove one piece of debris. Compared to simple heuristic matching methods, this optimization approach is more systematic and efficient, ultimately significantly reducing the overall removal cost of the mission.
[0030] By combining simulated annealing and dynamic programming, the entire technical solution not only solves the dual problems of trajectory changing and debris removal, but also achieves optimal resource allocation. Compared with existing single optimization methods, this combined approach provides stronger global optimization capabilities, especially when facing multiple variables and complex constraints, offering higher-quality solutions and significantly improving task execution efficiency and success rate.
[0031] Suitable for use in the removal of debris in mixed open and ground spaces. Attached Figure Description
[0032] Figure 1 This is a schematic diagram of a satellite removal system;
[0033] Figure 2 This is a schematic diagram simulating a constellation of space debris.
[0034] Figure 3 This is a flowchart of the simulation algorithm;
[0035] Figure 4 This is a schematic diagram of the average velocity pulse size;
[0036] Figure 5A diagram illustrating the number of emergency satellites that need to be launched;
[0037] Figure 6 A schematic diagram showing a pre-positioned satellite approaching target space debris;
[0038] Figure 7 A schematic diagram illustrating the approach of an emergency satellite to target debris;
[0039] Figure 8 This is a schematic diagram illustrating the removal of target debris by an emergency satellite. Detailed Implementation
[0040] To make the advantages and benefits of the technical solution provided by the present invention clearer, the technical solution provided by the present invention will now be described in further detail with reference to the accompanying drawings, specifically:
[0041] Implementation Method 1: This implementation method provides a method for matching space debris removal schemes using a hybrid air-ground approach, including:
[0042] Brief
[0043] Step 1: Select the spacecraft orbit change method and determine the optimization strategy.
[0044] Step 2: Initialize the simulated annealing algorithm parameters.
[0045] Step 3: Generate an initial solution, calculate and evaluate the objective function value.
[0046] Step 4: Iteratively optimize the trajectory change parameters to find the optimal solution.
[0047] Step 5: Use dynamic programming to optimize the matching scheme between satellites and space debris.
[0048] Step 6: Integrate and optimize the results, and execute the spacecraft orbit change and debris removal mission.
[0049] Detailed description
[0050] Step 1: Select the spacecraft orbit change method and determine the optimization strategy
[0051] Input: Orbit change method options (Hohmann orbit change, three-pulse orbit change, Lambert orbit change)
[0052] Solution: Among numerous orbit change methods, the Lambert orbit change method was selected for spacecraft orbit change. Since directly using the Lambert algorithm is unlikely to yield optimal results, a simulated annealing algorithm was employed for global optimization. The simulated annealing algorithm optimizes the three key variables of the Lambert problem (orbit change time, rendezvous time, and number of rendezvous orbits) to minimize velocity impulses and achieve fuel savings.
[0053] Output: The selected optimization strategy and objective function (minimize velocity impulses) will serve as the basis for subsequent steps.
[0054] Step 2: Initialize the simulated annealing algorithm parameters
[0055] Input: Optimization strategy and objective function
[0056] Processing: Determine the initial parameters of the simulated annealing algorithm, including the initial temperature (100.0), the termination temperature (0.001), the cooling rate, and the number of iterations at each temperature (100). The settings of these parameters will directly affect the convergence speed of the algorithm and the quality of the final solution.
[0057] Output: The simulated annealing algorithm after initialization is ready for optimization.
[0058] Step 3: Generate an initial solution, calculate and evaluate the objective function value.
[0059] Input: Initialized simulated annealing algorithm
[0060] Processing: An initial solution is generated randomly or obtained through a heuristic method. This solution includes initial values for the orbit change time, rendezvous time, and number of rendezvous orbits. Then, the objective function value of the solution, i.e., the velocity pulse size, is calculated, and its quality is evaluated.
[0061] Output: The initial solution and its objective function value will serve as the starting point for the simulated annealing algorithm.
[0062] Step 4: Iteratively optimize the trajectory change parameters to find the optimal solution.
[0063] Input: Initial solution and its objective function value
[0064] Process: Iterative optimization is performed using simulated annealing. In each iteration, new candidate solutions are generated and their objective function values are calculated. Based on the Metropolis criterion, poorer solutions are accepted with a certain probability to prevent getting trapped in local optima. As the temperature gradually decreases, the optimization process gradually converges, eventually finding the optimal combination of trajectory change parameters.
[0065] Output: Optimal orbit change parameters (orbit change time, rendezvous time, number of rendezvous circles) and minimized velocity pulse value.
[0066] Step 5: Use dynamic programming to optimize the matching scheme between satellites and space debris.
[0067] Input: Optimized trajectory change parameters
[0068] Solution: The optimized orbital parameters are applied to the satellite-space debris matching problem. A dynamic programming algorithm is used to construct a two-dimensional array to represent the states and determine the state transition equations to ensure that each satellite can only remove one piece of space debris and minimize the total removal cost. Based on the state transition equations and boundary conditions, the satellite-debris matching scheme is progressively optimized.
[0069] Output: The optimal satellite and space debris matching scheme will be used to carry out the actual cleanup mission.
[0070] Step 6: Integrate and optimize the results, and execute the spacecraft orbit change and debris removal mission.
[0071] Input: Optimal orbital adjustment parameters and satellite debris matching scheme
[0072] Processing: Integrate orbit change parameters and debris matching schemes to plan the actual orbit change and debris removal missions for the spacecraft. Ensure that orbit change and debris removal are executed as planned during actual operations to achieve optimal fuel consumption and mission efficiency.
[0073] Output: Successful completion of spacecraft orbit change and space debris removal missions, ultimately achieving safe and efficient mission execution.
[0074] This summary method ensures the integrity and logical order of the technical solution, which helps in understanding and executing each step in practical applications.
[0075] Implementation Method Two: Combination Figure 1-8 This embodiment describes the technical solution provided above in further detail through specific examples. Specifically:
[0076] The technical solution provided in this embodiment uses the optimization algorithm module of SpaceSim software, developed by the School of Aerospace Engineering at Harbin Institute of Technology, to calculate the velocity pulses for the orbit change of the service satellite and directly displays the orbit change process in the software, making the optimization results of the debris removal task clear and concise. For satellite disintegration scenarios, the multi-loop Lambert algorithm is optimized using simulated annealing to obtain the velocity pulse size required for each service satellite to remove each target space debris, thus establishing a velocity pulse matrix. This matrix is then processed using a dynamic programming algorithm, and when some space debris is located in unusual positions, an emergency launch of a satellite to the target space debris location is used to remove the debris.
[0077] This implementation proposes a novel approach: establishing a constellation of space debris removal satellites and supplementing it with ad-hoc launches. It studies the problem of multiple satellites clearing space debris in a cluster setting. For each piece of space debris, dynamic programming and other methods are used to find the most suitable satellite for removal, thus saving fuel for individual satellites and clearing more debris. When some space debris is located in unusual positions, an emergency launch of satellites to the target debris location is employed to remove it, ensuring minimal fuel consumption. Much research has been conducted on space debris removal technologies, and several mature, in-orbit verified technologies exist, such as space rope net capture technology. These mechanisms can effectively capture obsolete space debris.
[0078] Simulated annealing is a heuristic algorithm that simulates the cooling process of metal in real life. As the temperature decreases, the state of the algorithm's solution gradually stabilizes, eventually ceasing calculation and outputting the optimal solution when the temperature reaches its lowest point. During the search process, the algorithm decides whether to accept a new solution based on the difference between the current solution and a new solution, as well as the current temperature. Initially, the algorithm is set at a higher temperature, allowing for the acceptance of poorer solutions. As iterations progress and the temperature gradually decreases, the acceptance probability also decreases, leading the algorithm to gradually find a better solution globally. Simulated annealing has been widely used in aerospace planning research. For example, Ding Yinan et al. from the Beijing Institute of Control Engineering established an agile imaging satellite mission planning model with multiple satellites and multiple observation targets. Further, using the imaging time as the optimization object, they proposed a multi-target variable neighborhood simulated annealing algorithm that adjusts the search range using the cooling process, achieving good results.
[0079] Dynamic programming is a method for solving optimization problems in multi-stage decision-making processes. It effectively reduces redundant computation and improves algorithm efficiency. For highly complex problems, dynamic programming can effectively reduce computational complexity through recursion and memoization, making the problem solvable. Due to its wide applicability and efficiency, dynamic programming is widely used in algorithm design and practical problem solving.
[0080] This implementation study investigates the mission planning problem of multi-satellite clearance of multiple space debris involving both air and ground. A complete mission planning module for multi-space debris clearance was established in the domestic space simulation software SpaceSim, enabling clear visualization of the results. A simulated annealing algorithm was applied to the multi-cycle Lambert algorithm to establish the velocity pulse matrix. When some space debris locations are unusual, an emergency satellite launch to the target debris location was employed for clearance. Results show that the mission planning for multi-space debris clearance yields good results, effectively reducing the total mission pulse and fuel consumption per satellite, thus saving satellite costs while eliminating more space debris.
[0081] Specifically:
[0082] Space debris strike mission planning issues
[0083] The scenario is that a satellite suddenly disintegrates, and the debris is distributed across multiple orbital planes, forming a debris cluster. It is urgent to remove this debris cluster in orbit to prevent it from colliding with other satellites and causing greater impact.
[0084] The parameters and variables are defined as follows:
[0085] Y is a pre-set set of satellites, Y = {y1, y2, ... y} Ny}where N y The number of satellites is preset.
[0086] P is a set of emergency satellites, P = {p1, p2, ... p} Np}where N p This refers to the number of emergency satellites.
[0087] A set of K spatial fragments, K = {k1, k2, ... k} Nk}where N k This represents the number of space fragments.
[0088] t1 is the time it takes for our satellite to change orbit.
[0089] The rendezvous time between the t2 satellite and space debris.
[0090] n The number of orbits our satellite made before the rendezvous.
[0091] Space debris strike mission modeling:
[0092] For space debris removal missions, an equivalent simulation model for debris removal is established as a design constraint for both pre-deployed and emergency constellations. Deployment methods for debris removal satellites include emergency deployment and pre-deployment using dedicated constellations. Co-orbital deployment is considered the primary form of emergency deployment.
[0093] Clearance Model: Clearance systems are simplified using beams, such as weapon systems... Figure 1 As shown, the beam direction represents the removal direction of our satellite. When our satellite's beam is pointed at a space debris, it means that our satellite will remove that space debris. It is assumed that the removal will be successful, and a single satellite can only remove a single piece of space debris.
[0094] Because the satellites in the Walker constellation are located in different orbital planes and are relatively evenly distributed, the display effect is good, and it can effectively play a simulation role. This implementation uses the Walker constellation as space debris to simulate the space debris group that we want to clear. The specific orbital numbers of the generated space debris are calculated by SpaceSim software and displayed within the software. Figure 2 Show.
[0095] Emergency constellations are designed for rapid deployment when the number of pre-installed satellites or the available fuel is insufficient to strike all space debris in a debris cluster. This requires careful planning of launch sites, the number of satellites to be launched, and other details. The orbital design principle for emergency satellites prioritizes the rapid achievement of strike targets.
[0096] At the initial moment, the satellite begins its erection process, reaching its designated position within Chinese territory. After erection, the satellite first travels to one of the seven aggregation points located within Chinese territory closest to its intended launch location. Once erected, the satellite maneuvers at a constant speed towards the target launch point based on its geographical location. After reaching the launch point, the satellite remains there for a period of time to prepare for launch.
[0097] After a period of simulation, our satellites will be launched successively. Satellite launches are based on the relative positions of our satellites and space debris. When the relative positions reach a minimum within a certain timeframe, the satellite will be launched immediately. If the minimum value does not appear within 3000 seconds, the satellite will not wait for the launch window and will be launched immediately. The time it takes for the satellite to reach its expected position is related to the distance between our satellite and the space debris at launch time. The shorter the distance between our satellite and the space debris at launch time, the shorter the flight time after launch.
[0098] Spacecraft orbit change model:
[0099] Common orbit-changing methods for spacecraft include Hohmann orbit change, three-pulse orbit change, and Lambert orbit change. This implementation uses the Lambert algorithm as the orbit-changing algorithm and optimizes the Lambert problem using simulated annealing. Directly using the Lambert method often yields poor results; therefore, an optimized Lambert orbit change method is typically employed to conserve fuel. This implementation uses simulated annealing to optimize the Lambert problem. Simulated annealing is a global optimization algorithm based on physics principles, initially used to solve crystal structure optimization problems during metal annealing. This algorithm finds the minimum value of a function by simulating the process of solid materials becoming more homogeneous and stable at high temperatures. For the multi-loop Lambert problem, the variables to be optimized are the orbit change time, rendezvous time, and number of rendezvous loops. The minimum value of the function refers to the minimum value of the velocity pulses.
[0100] In simulated annealing, the system is initially at a high temperature, accepting poor solutions at this state. As the temperature gradually decreases, the system becomes increasingly inclined to accept solutions better than the current one. This avoids getting trapped in local optima.
[0101] Specifically, the simulated annealing algorithm includes the following steps:
[0102] 1. Initialization: Determine parameters such as initial temperature, ending temperature, and cooling rate.
[0103] 2. Randomly generate an initial solution or obtain an initial approximate solution from a known heuristic method.
[0104] 3. In each iteration step, a small perturbation is applied to the current solution to generate a new candidate solution. This perturbation decreases as the temperature decreases, and the objective function value is calculated.
[0105] 4. Based on the Metropolis criterion (probabilistic acceptance of different solutions), choose with a certain probability whether to accept the new candidate solution as the next search point or update the current search point.
[0106] 5. Repeat steps 3-4 until the termination condition is met (such as reaching the maximum number of iterations).
[0107] For the Lambert problem, there are three variables: the orbit change time t1 of our satellite, the rendezvous time t2 between the satellite and space debris, and the number of orbits n of our satellite before rendezvous. The initial conditions are set as follows:
[0108] t1 = 0.2t, t2 = 0.8t,
[0109] Where a is the semi-major axis of our satellite's original orbit, 0.6t = t2 - t1, which is the flight time of our satellite after the orbit change, and n is the number of orbits from time t1 to time t2, rounded up, assuming our satellite does not change its orbit.
[0110] The main limitations are as follows:
[0111] The constraint for time t is 0. <t1<t2<t
[0112] The constraint for the number of revolutions n is 0. <n
[0113] The orbit cannot cross the Earth during the orbital change process.
[0114] For the annealing algorithm, the initial temperature is set to 100.0, the final temperature is set to 0.001, and the number of iterations at each temperature is set to 100. After every 100 iterations, the temperature is decreased. The simulation algorithm flowchart is as follows. Figure 3 As shown.
[0115] For the debris cluster problem of using n pre-positioned satellites to clear n pieces of space debris, assuming each satellite can only clear one piece of space debris, and each piece of space debris can only be cleared by one satellite, the correspondence between satellites clearing space debris can affect the final clearing plan result. Using the annealing algorithm, each pre-positioned satellite can obtain an optimized result for each piece of space debris. Ultimately, an n*n matrix can be obtained. The problem can be simplified to taking one value from each row, with no duplicate column values, and minimizing the sum of the final n numbers. Dynamic programming can effectively solve this problem.
[0116] The dynamic programming algorithm is used to solve this problem as follows:
[0117] 1. Define the state: Design a state representation, which is usually related to the solution to the problem. In this problem, we can design a two-dimensional array dp, a selection order p, and a final selection order q. In dp, dp[i][j] stores the minimum sum obtained after planning the array starting from column j through rows i, p[i][j] stores the selection order of each row, and q is the best selection order in p.
[0118] 2. State Transition Equation: Determine the state transition equation, i.e., how to update the current state based on the previous state. In this problem, the state transition equation is expressed as: dp[i][j]=min(dp[i-1][k]+matrix[i][j]), where k≠j, and satisfies the constraint condition (i.e., the number of columns selected in the (i-1)th row cannot be the same as the number of columns previously selected in p).
[0119] 3. Boundary Conditions: Determine the boundary conditions, i.e., the initial state. In this problem, the state of the first row can be directly determined by the values in the matrix.
[0120] Finally, the results analysis:
[0121] Classification of mixed-space debris removal tasks
[0122] For our side N y Pre-positioned satellites clear N k The problem of space debris clustering assumes that each satellite can only remove one piece of space debris, and each piece of space debris can only be removed by one satellite. The correspondence between satellites removing space debris can affect the final removal plan. Using the annealing algorithm, each pre-set satellite can obtain an optimized result for each piece of space debris. Ultimately, an N-fold space debris clustering plan can be obtained. y *N k The matrix. There are three cases here.
[0123] Case 1: N y =N k
[0124] At this point, the number of our satellites equals the number of space debris, and N y *N k The matrix is calculated using the original dynamic programming algorithm, which yields the selection order q. Let q... n The value 'a' indicates that our satellite numbered 'n' will change its orbit to clear space debris numbered 'a'. Taking a set of parameters as an example of our satellites clearing 5 pieces of space debris, the resulting matrix is shown in Table 1. 'q' is [5,1,2,3,4], the total velocity pulse size is 1189.8 m / s, and no single velocity pulse exceeds 300 m / s. The total velocity pulse is relatively small, indicating a good clearing effect.
[0125] Table 1 Case 1
[0126]
[0127] Scenario 2: N y >N k
[0128] At this point, the number of our satellites exceeds the amount of space debris, and some of our satellites will not be assigned to any removal targets. In this situation, N... y *N k Matrix flipped to N k *N yThe matrix is still calculated using the original dynamic programming algorithm. The original algorithm aims to prioritize finding the space debris with the best clearance effect from our satellites. After flipping, since the number of columns is greater than the number of rows, and the number of columns represents the number of our satellites, this can make full use of our satellite resources. Taking the example of our 5 satellites clearing 4 pieces of space debris, the matrix is shown in Table 2. q is [2,3,4,5], the total velocity pulse size is 958.4 m / s, and no single velocity pulse exceeds 300 m / s. The total velocity pulse is relatively small, indicating a good clearance effect.
[0129] Table 2, Case 2
[0130]
[0131] Scenario 3: N y <N k
[0132] At this point, the number of our satellites is less than the number of space debris, so N y *N k The matrix is calculated using the original dynamic programming algorithm, which yields the selection order q. Let q... n The value 'a' indicates that our nth satellite will change its orbit to clear space debris numbered 'a' in the plan. Taking the example of our 4 satellites clearing 5 pieces of space debris, the matrix is shown in Table 3. 'q' is [5,1,2,3], the total velocity pulse size is 918.8 m / s, and none of the individual velocity pulses exceed 300 m / s. The total velocity pulse is relatively small, indicating a good clearing effect, but one piece of space debris remains uncleared. In this case, an emergency satellite will be launched to clear the space debris not assigned to a satellite. The emergency satellite's set cost is 400 m / s. Furthermore, if the velocity pulse for a pre-assigned satellite to clear space debris exceeds 400 m / s, an emergency satellite will also be launched to clear the space debris. Therefore, the total velocity pulse in this case is 1318.8 m / s.
[0133] Table 3, Situation 3
[0134]
[0135] The effect of the number of satellites on the total velocity pulse size:
[0136] Taking scenario one as an example, we continuously increase the number of our satellites and plot the curve showing the change in the average velocity pulse magnitude as the number of our satellites increases, as shown below. Figure 4As shown in the analysis, the average velocity pulse size decreases with the increase of the number of satellites, and drops sharply at some points. This is because adding new satellites at these points significantly impacts the overall dynamic programming path selection, while adding satellites at most points has little impact, thus causing the average velocity pulse to increase. When the number of satellites is small, the average velocity pulse decreases sharply with the increase of the number of satellites. After exceeding 10, a smoother dynamic programming path can be selected, causing the decrease in velocity pulse size to slow down. Therefore, it is recommended to pre-deploy at least 10 satellites, preferably located in different orbital planes, to reduce the average velocity pulse.
[0137] The curve showing the change in the number of emergency satellites required for clearance (i.e., the amount of space debris requiring clearance with velocity pulses exceeding 400 m / s) versus the number of pre-set satellites is as follows: Figure 5 As shown, the number of emergency satellites that need to be launched is generally small, and emergency satellites are mainly used when the number of our satellites is less than the number of space debris.
[0138] The results in Spacesim show:
[0139] The calculation results are visualized in Spacesim. Taking a scenario where our satellites are fewer than space debris as an example, the Spacesim visualization window is shown below. Our pre-positioned satellites are green, space debris that has not been cleared is represented by white spheres, which turn red after being cleared. Our emergency launch satellite is represented by a blue sphere. Pre-positioned satellites approaching target space debris... Figure 6 As shown, our pre-positioned satellites have approached and overlapped with the target space debris and cleared it. Since the number of our pre-positioned satellites is less than the number of space debris, there are still small white balls in the scene. Our satellites aimed their beams at the small white balls in an attempt to clear them, but failed because the distance was too far.
[0140] After a period of time, emergency satellites are launched to intercept the remaining space debris. These emergency satellites approach the target debris, such as... Figure 7 As shown, the emergency satellite successfully cleared the target debris. Figure 8 As shown, our emergency satellite successfully approached and cleared the target debris. The scenario demonstration was successful, and Spacesim can demonstrate space debris removal scenarios in a relatively intuitive way.
[0141] This implementation study investigates the planning problem of space debris removal missions using a multi-service satellite system based on simulated annealing. Simulation results show that establishing a large-scale pre-positioned constellation can effectively remove space debris swarms. The average velocity pulse size decreases with the number of satellites, and launching emergency satellites can effectively supplement the pre-positioned satellites. Spacesim software has strong analytical and visualization capabilities, supporting simulations of large-scale scenarios.
[0142] The above description of several specific embodiments further details the technical solution provided by the present invention in order to highlight the advantages and benefits of the technical solution provided by the present invention. However, the above-described specific embodiments are not intended to limit the present invention. Any reasonable modifications and improvements to the present invention, combinations of embodiments, and equivalent substitutions based on the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A large-scale constellation orbit cooperative maneuver optimization method applied to space debris removal tasks of air-ground hybrid, characterized in that, include: The steps involved in collecting a pre-set set of satellites, a set of space debris, and the optimization objectives of the algorithm. Based on the optimization objective, the steps include setting the parameters of the simulated annealing algorithm and collecting randomly generated initial solutions as the starting point of the simulated annealing algorithm. The steps include iteratively optimizing the initial solution for each pre-set satellite in the pre-set satellite set and each space debris in the space debris set until a combination of orbit change parameters under preset conditions and a velocity pulse corresponding to the combination of orbit change parameters are obtained, and establishing a velocity pulse matrix based on each velocity pulse. Based on the premise that each satellite can only clear one piece of space debris and each piece of space debris can only be cleared by one satellite, a dynamic programming algorithm is used to process the velocity pulse matrix to obtain a pre-set matching scheme between satellites and space debris. When the number of pre-set satellites or fuel is insufficient to clear all space debris in the set of space debris, emergency satellites are deployed to clear the space debris that has not been cleared by the pre-set satellites.
2. The method of claim 1, wherein, The algorithm is the Lambert algorithm.
3. The method of claim 1, wherein, The optimization objectives include orbit change time, rendezvous time, and number of rendezvous cycles.
4. The method of claim 1, wherein, The parameters include the initial temperature, the final temperature, the cooling rate, and the number of iterations at each temperature.
5. A large-scale constellation orbit cooperative maneuver optimization device applied to a space-ground hybrid space debris removal task, characterized in that, To implement the method of claim 1, the method comprises: The module for optimizing the acquisition algorithm; A module for setting the parameters of the simulated annealing algorithm based on the optimization objective; The module that collects randomly generated initial solutions as the starting point of the simulated annealing algorithm; The module iteratively optimizes the initial solution until a combination of trajectory change parameters under preset conditions is obtained.
6. A method for matching space debris removal space hybrid schemes, characterized in that, include: The step of acquiring the orbital parameter combination obtained according to claim 1; The process involves matching individual satellites with individual space debris based on the orbit change parameter combination, after each satellite has cleared only one piece of space debris.
7. An air-space hybrid clearance space debris scheme matching apparatus, characterized by, include: A module for acquiring the combination of trajectory change parameters obtained in claim 1; This module, based on the principle of clearing only one piece of space debris per satellite, matches individual satellites with individual pieces of space debris according to the aforementioned combination of orbit change parameters.
8. Computer storage medium for storing a computer program, characterized in that When the computer program is read by the computer, the computer executes the method of claim 1.
9. A computer comprising a processor and a storage medium, characterized in that When the processor reads the computer program stored in the storage medium, the computer executes the method of claim 1.
10. Computer program product as computer program, characterized in that When the computer program is executed, it implements the method of claim 1.
Citation Information
Patent Citations
Space fragment removing system and method and space fragment removing task planning method
CN110395412A
Multi-satellite task scheduling method and system based on simulated annealing algorithm
CN113919122A
