技术领域Technical field
本发明涉及卫星遥感技术领域，尤其是涉及顾及动中成像影像MTF退化的超敏捷卫星多星区域成像任务规划方法及系统。The present invention relates to the field of satellite remote sensing technology, and in particular to a super-agile satellite multi-satellite regional imaging mission planning method and system that takes into account MTF degradation of moving imaging images.
背景技术Background technique
超敏捷卫星在“动中成像”的过程中，会存在满足不同姿态机动约束前提下的若干不同的成像方案。这些不同的成像任务方案不仅在覆盖率、成像分辨率上存在差异，其所获影像的图像调制传递函数（MTF）也有所不同。调制传递函数MTF表征空间相机对目标对比度的传输能力，主要影响遥感影像的清晰度。In the process of "imaging in motion" of ultra-agile satellites, there will be several different imaging solutions that meet the constraints of different attitude maneuvers. These different imaging mission solutions not only differ in coverage and imaging resolution, but also in the image modulation transfer function (MTF) of the images obtained. The modulation transfer function MTF characterizes the transmission ability of the space camera to the target contrast, and mainly affects the clarity of remote sensing images.
“动中成像”对影像MTF的影响，主要体现如下：超敏捷卫星的“动中成像”与传统敏捷卫星的“稳态推扫”不同，“动中成像”过程中姿态的连续变化，使得传感器每个扫描行的积分时间不同，扫描方向也随姿态时变，导致积分时间、积分级数选择、偏流角修正的匹配关系具有高动态性，引起图像调制传递函数MTF下降变化，造成图像的模糊现象。The impact of "moving imaging" on image MTF is mainly reflected as follows: the "moving imaging" of super-agile satellites is different from the "steady-state push sweep" of traditional agile satellites. The continuous changes in attitude during the "moving imaging" process make The integration time of each scanning line of the sensor is different, and the scanning direction also changes with the attitude. As a result, the matching relationship between the integration time, integration series selection, and deflection angle correction is highly dynamic, causing the image modulation transfer function MTF to decrease and change, resulting in image distortion. blurring phenomenon.
敏捷卫星在成像时执行“稳态推扫”，其成像任务规划通常以优化成像覆盖收益为核心目标构建成像任务规划模型。超敏捷卫星“动中成像”的成像任务规划，若直接采用传统方法仅以最大化成像覆盖收益和最小化任务执行时间为目标，将直接导致获取的影像MTF不受控，严重制约动中成像应用效能的发挥。因此，在“动中成像”任务规划过程中，需要定量分析卫星姿态机动与影像MTF之间的映射关系，将影像MTF纳入动中成像任务规划中，从任务规划这一源头开始，对超敏捷卫星成像时的影像MTF实施全流程控制。Agile satellites perform "steady-state push sweep" during imaging, and their imaging mission planning usually constructs an imaging mission planning model with the core goal of optimizing imaging coverage benefits. If the imaging mission planning of ultra-agile satellite "imaging in motion" directly adopts traditional methods with the goal of maximizing imaging coverage income and minimizing mission execution time, it will directly lead to the MTF of the acquired images being uncontrolled, seriously restricting the imaging in motion. application performance. Therefore, in the process of "imaging in motion" mission planning, it is necessary to quantitatively analyze the mapping relationship between satellite attitude maneuvers and image MTF, and incorporate image MTF into the imaging in motion mission planning. Starting from the source of mission planning, we must analyze the mapping relationship between satellite attitude maneuvers and image MTF. The image MTF during satellite imaging implements full process control.
针对顾及动中成像影像MTF退化的超敏捷卫星多星区域成像任务规划需要和现有方法的不足，本领域亟待提出新的技术方案。In view of the needs of ultra-agile satellite multi-satellite regional imaging mission planning that takes into account the MTF degradation of moving imaging images and the shortcomings of existing methods, new technical solutions are urgently needed in this field.
发明人的研究团队之前做出的相关领域技术成果如下，但针对本发明要解决的技术问题还存在缺陷：The inventor's research team has previously made the following technical achievements in related fields, but there are still deficiencies in the technical problems to be solved by the present invention:
专利文献CN109948852A提供了一种敏捷卫星的同轨多点目标成像任务规划方法，方法包括：步骤S1，确定需要成像的n个点目标，以及与n个点目标对应的成像时刻；步骤S2，按照n个点目标对应的成像时刻的先后顺序，获得成像时刻集合x，以及与成像时刻集合x对应的n个点目标的集合p；步骤S3，依次取i从1至n-1，取j从2至n，且i小于j，确定集合p对应的公有顶点的值的集合a；步骤S4，根据公有顶点值集合a，获得姿态约束无向图型结构；步骤S5，对姿态约束无向图型结构分块定向，获得姿态约束有向图型结构，并根据姿态约束有向图型结构，获得多组成像顺序；步骤S6，在多组成像顺序中，选择至少一组成像顺序作为最优成像顺序。Patent document CN109948852A provides a co-orbital multi-point target imaging mission planning method for agile satellites. The method includes: Step S1, determine n point targets that need to be imaged, and the imaging time corresponding to the n point targets; Step S2, according to The sequence of imaging moments corresponding to n point targets is obtained to obtain the imaging moment set x, and the set p of n point targets corresponding to the imaging moment set 2 to n, and i is less than j, determine the set a of the public vertex values corresponding to the set p; Step S4, obtain the attitude-constrained undirected graph structure according to the public vertex value set a; Step S5, obtain the attitude-constrained undirected graph The structure is oriented in blocks to obtain an attitude-constrained directed graph structure, and according to the attitude-constrained directed graph structure, multiple sets of imaging sequences are obtained; step S6, among the multiple sets of imaging sequences, select at least one set of imaging sequences as the optimal Imaging sequence.
专利文献CN111666661A提供了一种敏捷卫星单轨动中成像多条带拼接任务规划方法及系统，包括针对成像任务区域，基于旋转卡壳的原理建立区域的外接矩形，分割得到若干条带；求出每一个条带对应覆盖率，条带起始边和终止边的中点坐标，计算每个条带起始、终止端点的成像时间窗口；对将卫星姿态运动约化为相机指向点的平面运动，构建多条带拼接成像过程中卫星相机指向点的平面运动约束，确定约束条件；对成像时间窗口进行裁剪和成像时刻归一化操作，确定决策变量；构建敏捷卫星单轨动中成像多条带拼接任务规划数学模型，确定模型决策变量与目标函数、约束条件的定量关系；采用PSO优化算法进行求解，得到成像多条带拼接任务规划方案，实现对成像任务区域的最大覆盖。Patent document CN111666661A provides an agile satellite monoorbit in-motion imaging multi-strip splicing mission planning method and system, which includes establishing a circumscribed rectangle of the area based on the principle of rotation jamming for the imaging task area, and dividing it into several strips; finding each The corresponding coverage of the strip, the midpoint coordinates of the starting edge and the ending edge of the strip, calculate the imaging time window of the starting and ending end points of each strip; to reduce the satellite attitude motion to the plane motion of the camera pointing point, construct During the multi-strip splicing imaging process, the plane motion constraints of the satellite camera pointing point are determined, and the constraint conditions are determined; the imaging time window is clipped and the imaging time is normalized to determine the decision variables; the multi-strip splicing task of agile satellite single-orbit in-motion imaging is constructed Plan the mathematical model and determine the quantitative relationship between the model decision variables, the objective function, and the constraints; use the PSO optimization algorithm to solve the problem, and obtain the imaging multi-strip splicing task planning plan to achieve maximum coverage of the imaging task area.
专利文献CN112149911A提出一种超敏捷卫星同轨多点目标动中成像任务规划方法。本发明进行多点目标聚类、非沿迹条带分解的预处理，并求出所有条带端点的成像时间窗口；构建条带成像编号序列、条带成像方向序列得到条带端点成像序列，并利用成像时刻归一化系数计算条带端点成像时刻序列；根据条带成像编号序列、条带成像方向序列、条带端点成像时刻序列构建决策变量，成像覆盖收益最大化、任务完成时间最小化构建目标函数，成像时间窗口、姿态转换时间构建约束条件，进一步构建同轨多点目标动中成像任务规划数学模型；通过改进粒子群算法优化得到最优的成像任务规划方案。本方法可充分利用敏捷卫星的姿态机动能力，实现对同轨多点目标动中成像任务方案的优化。Patent document CN112149911A proposes a super-agile satellite co-orbital multi-point target in-motion imaging mission planning method. The present invention performs preprocessing of multi-point target clustering and non-along-track strip decomposition, and obtains the imaging time windows of all strip endpoints; constructs a strip imaging number sequence and a strip imaging direction sequence to obtain a strip endpoint imaging sequence. And use the imaging time normalization coefficient to calculate the strip endpoint imaging time sequence; construct decision variables based on the strip imaging number sequence, strip imaging direction sequence, and strip endpoint imaging time sequence to maximize imaging coverage benefits and minimize task completion time. Construct the objective function, constraint conditions for the imaging time window and attitude conversion time, and further construct a mathematical model for the imaging mission planning of the co-orbital multi-point target in motion; the optimal imaging mission planning solution is obtained through improved particle swarm algorithm optimization. This method can make full use of the attitude maneuverability of agile satellites to optimize the moving imaging mission plan for multi-point targets in the same orbit.
专利文献CN115688568A提供了一种超敏捷卫星多星区域成像任务的调度方法，包括如下步骤：步骤1，分析面向超敏捷卫星多星区域成像任务需求，对任务调度过程进行假设化简；步骤2，以条带长度总和最短为目标，对待成像区域进行条带分解，获得任务调度的原子任务集；步骤3，根据步骤2的任务集，选取相应的多类决策变量，以最大化成像覆盖收益、最小化成像任务执行时间为目标，构建多星成像任务调度模型；步骤4，求解所构建的多星成像任务调度模型；步骤5，将步骤4求解得到的最优决策变量组合还原为任务调度方案。本发明有效适应了存多类决策变量问题的优化求解需求，能够有效地适用于超敏捷卫星多星区域成像任务调度。Patent document CN115688568A provides a scheduling method for ultra-agile satellite multi-satellite regional imaging tasks, including the following steps: Step 1, analyze the requirements for ultra-agile satellite multi-satellite regional imaging tasks, and make assumptions about the task scheduling process; Step 2, With the shortest sum of strip lengths as the goal, decompose the strips of the area to be imaged to obtain the atomic task set for task scheduling; Step 3, according to the task set in step 2, select the corresponding multi-category decision variables to maximize imaging coverage benefits. Minimize the imaging task execution time as the goal, and build a multi-star imaging task scheduling model; Step 4, solve the constructed multi-star imaging task scheduling model; Step 5, restore the optimal decision variable combination obtained in Step 4 to a task scheduling plan . The invention effectively adapts to the optimization and solution requirements of problems with multiple types of decision variables, and can be effectively applied to ultra-agile satellite multi-satellite regional imaging task scheduling.
然而，专利文献CN109948852A和CN112149911A所提出的方法主要面向点目标的成像任务规划。专利文献CN111666661A所提出的方法主要面向单星的成像任务规划，且该方法无法优化条带的成像动作序列。专利文献CN115688568A所提出的方法能够实现对卫星资源的配置和单星成像动作序列的一体化优化，然而，该方法能够在保证成像覆盖收益最优的前提下，提升超敏捷卫星成像时的观测效率，但无法同步实现影像MTF和观测效率的均衡优化。However, the methods proposed in patent documents CN109948852A and CN112149911A are mainly oriented to imaging task planning of point targets. The method proposed in the patent document CN111666661A is mainly oriented to single-star imaging mission planning, and this method cannot optimize the imaging action sequence of the strip. The method proposed in the patent document CN115688568A can realize the integrated optimization of the configuration of satellite resources and the single-star imaging action sequence. However, this method can improve the observation efficiency of ultra-agile satellite imaging while ensuring optimal imaging coverage benefits. , but it is impossible to achieve balanced optimization of image MTF and observation efficiency simultaneously.
发明内容Contents of the invention
本发明针对现有技术问题，提出了一种顾及动中成像影像MTF退化的超敏捷卫星多星区域成像任务规划方法及系统。In view of the existing technical problems, the present invention proposes a super-agile satellite multi-satellite regional imaging mission planning method and system that takes into account the MTF degradation of moving imaging images.
本发明提供的技术方案提供一种顾及影像MTF退化的超敏捷卫星任务规划方法，包括以下处理：The technical solution provided by the present invention provides an ultra-agile satellite mission planning method that takes into account image MTF degradation, including the following processing:
对待成像区域以条带长度总和最短为优化目标，进行条带分解生成成像条带集合；Taking the shortest sum of strip lengths as the optimization goal in the area to be imaged, perform strip decomposition to generate an imaging strip set;
构建姿态机动角速度与影像MTF之间的映射关系；Construct a mapping relationship between attitude maneuver angular velocity and image MTF;
构建以最大化成像覆盖收益、最大化影像MTF和最小化成像任务执行时间为目标函数，条带选星编号、条带成像动作序列编号、条带成像方向编号、条带端点成像时刻的归一化系数为决策变量的基于偏序的Pareto多目标成像任务规划模型；Construct the objective function of maximizing imaging coverage income, maximizing image MTF and minimizing imaging task execution time, and normalizing the stripe star selection number, stripe imaging action sequence number, stripe imaging direction number, and stripe endpoint imaging time Pareto multi-objective imaging task planning model based on partial ordering with coefficients as decision variables;
针对所构建基于偏序的Pareto多目标成像任务规划模型，利用基于偏序的NSGA-II方法求解模型；所述利用基于偏序的NSGA-II方法求解模型，是在求解过程中优先比较种群中的两个个体之间的覆盖收益进行非支配排序，当且仅当两个个体之间的覆盖收益相等时，再对影像MTF和成像任务执行时间进行非支配排序；For the Pareto multi-objective imaging mission planning model based on partial ordering, the NSGA-II method based on partial ordering is used to solve the model; the NSGA-II method based on partial ordering is used to solve the model, which is to give priority to comparison among the populations during the solving process. Perform a non-dominated sorting on the coverage benefits between the two individuals. If and only if the coverage benefits between the two individuals are equal, then perform a non-dominated sorting on the imaging MTF and imaging task execution time;
获得顾及动中成像影像MTF退化的超敏捷卫星多星区域成像任务规划方案并输出。Obtain and output a super-agile satellite multi-satellite regional imaging mission planning plan that takes into account the MTF degradation of moving imaging images.
而且，所述构建姿态机动角速度与影像MTF之间的映射关系，实现方式为，进行TDI积分级数与影像MTF的映射关系构建，进行姿态机动角速度与TDI积分级数的映射关系构建，通过对上述两节映射关系的结合，最终确定不同信噪比下的姿态机动角速度与影像MTF之间的映射关系。Moreover, the method of constructing the mapping relationship between the attitude maneuvering angular velocity and the image MTF is to construct a mapping relationship between the TDI integral series and the image MTF, and construct a mapping relationship between the attitude maneuvering angular velocity and the TDI integral series. The combination of the mapping relationships in the above two sections finally determines the mapping relationship between attitude maneuvering angular velocity and image MTF under different signal-to-noise ratios.
而且，所构建的基于偏序的Pareto多目标成像任务规划模型中，最大化成像覆盖收益、最大化影像MTF和最小化成像任务执行时间这三个目标函数之间具备偏序关系，包括以最大化成像覆盖收益为首要优化目标，第二优化目标为最大化影像MTF和最小任务执行时间，其中，首要优化目标与第二优化目标构成优先级排序关系，优先确保首要优化目标基础上，再实现两个第二优化目标的均衡。Moreover, in the Pareto multi-objective imaging task planning model based on partial ordering, the three objective functions of maximizing imaging coverage revenue, maximizing image MTF and minimizing imaging task execution time have a partial ordering relationship, including maximizing The primary optimization goal is to maximize imaging coverage income, and the second optimization goal is to maximize image MTF and minimize task execution time. Among them, the primary optimization goal and the second optimization goal form a priority ranking relationship. On the basis of ensuring the first optimization goal first, then achieving Equilibrium of two secondary optimization objectives.
而且，针对所构建基于偏序的Pareto多目标成像任务规划模型，利用基于偏序的NSGA-II方法求解模型，实现过程如下，Moreover, for the Pareto multi-objective imaging mission planning model based on partial ordering, the NSGA-II method based on partial ordering is used to solve the model. The implementation process is as follows:
步骤4.1，对父代种群中的个体进行随机初始化：Step 4.1, randomly initialize the individuals in the parent population:
设置种群规模为N，生成N个初始父代种群，每个个体代表一种顾及动中成像影像MTF退化的超敏捷卫星多星区域成像任务规划方案，最大进化次数设置为Evos；Set the population size toN and generateN initial parent populations. Each individual represents a super-agile satellite multi-satellite regional imaging mission planning scheme that takes into account the MTF degradation of moving imaging images. The maximum number of evolutions is set toEvos ;
步骤4.2，计算当前父代种群中每个个体的规划方案，并得到当前种群中每个个体的适应度值，所述适应度值为成像覆盖收益、影像MTF和成像任务执行时间；Step 4.2: Calculate the planning plan of each individual in the current parent population, and obtain the fitness value of each individual in the current population. The fitness value is the imaging coverage income, imaging MTF and imaging task execution time;
步骤4.3，进行基于偏序的非支配排序，然后通过拥挤度距离计算对具有相同非支配排序层的个体进行选择性的排序；Step 4.3, perform non-dominated sorting based on partial ordering, and then selectively sort individuals with the same non-dominated sorting layer through crowding distance calculation;
所述基于偏序的非支配排序实现方式为，优先比较种群中的两个个体之间的覆盖收益，当个体l的覆盖收益Fitevo,l,cov小于个体r的覆盖收益Fitevo,r,cov时，个体r支配个体l；当个体l的覆盖收益Fitevo,l,cov大于个体r的覆盖收益Fitevo,r,cov时，个体l支配个体r；当个体l和个体r的覆盖收益相等时，即Fitevo,l,cov=Fitevo,r,cov，再对第二目优化目标，即最大化影像MTF与最小化任务执行时间这两者的支配关系进行判断，从而实现个体的非支配排序；The non-dominated sorting based on partial order is implemented by first comparing the coverage benefits between two individuals in the population. When the coverage benefitFitevo,l,cov of individuall is less than the coverage benefitFit evo,r ofindividualr, cov , individualr dominates individuall ; when individuall' s covering incomeFitevo,l,cov is greater than individualr 's covering incomeFitevo,r,cov , individuall dominates individualr ; when the covering income of individuall and individualr When equal, that is,Fitevo,l,cov =Fitevo,r,cov , and then judge the dominance relationship between the second optimization goal, that is, maximizing image MTF and minimizing task execution time, so as to achieve individual non-dominated sorting;
其中，Fitevo,l,cov和Fitevo,r,cov分别表示基于偏序的NSGA-II方法在evo次进化过程中第l个个体和第r个个体对应的适应度值中的覆盖收益；Among them,Fitevo,l,cov andFitevo,r,cov respectively represent the coverage benefits of the partial order-based NSGA-II method in the fitness values corresponding to thel- th individual and ther- th individual in theevo evolution process;
步骤4.4，进行选择、交叉和变异操作；Step 4.4, perform selection, crossover and mutation operations;
步骤4.5，对父代种群和子代种群进行合并，包括对父代种群和子代种群进行合并形成大小为2N的种群Nt；Step 4.5, merge the parent population and the offspring population, including merging the parent population and the offspring population to form a populationNt of size2N ;
步骤4.6，对合并形成的规模为2N的种群Nt利用步骤4.3的方式，进行基于偏序的非支配排序和拥挤度距离计算，生成新的种群；Step 4.6: Use the method of step 4.3 to perform non-dominated sorting and crowding distance calculation based on partial ordering for the2N populationNt formed by the merger, and generate a new population;
步骤4.7，进行优化迭代，直至达到所设置的最大进化次数Evos，获取在确保最大化成像覆盖收益前提下，影像MTF和成像任务执行时间均衡优化的成像任务规划方案。Step 4.7: Perform optimization iterations until the set maximum number of evolutionsEvos is reached, and obtain an imaging task planning solution that balances and optimizes the image MTF and imaging task execution time while ensuring maximum imaging coverage benefits.
而且，粒子的位置表示所构建的基于偏序的Pareto多目标成像任务规划模型所述的决策变量，记为Sj,evo,l、Pi,evo,l、Qi,k,evo,l、ti,m,evo,l，各粒子的初始值在其各自的取值范围内进行随机取值；Moreover, the position of the particle represents the decision variable described in the constructed Pareto multi-objective imaging mission planning model based on partial order, which is recorded asSj,evo,l ,Pi,evo,l ,Qi,k,evo,l ,ti,m,evo,l , the initial value of each particle is randomly selected within its respective value range;
其中，evo表示采用基于偏序的NSGA-II方法进行的第evo次进化，evo∈{1,2,…,evo,…,Evos}，l表示第evo次进化时N个父代种群中的第l个个体，l∈{1,2,…,l,…N}；Sj,evo,l表示第evo次进化时父代种群中第l个个体的第j个条带的条带选星编号；Pi,evo,l表示第evo次进化时父代种群中第l个个体的第i颗卫星所包含条带的条带成像动作序列编号；Qi,k,evo,l表示第evo次进化时父代种群中第l个个体的第i颗卫星中的第j个条带的条带成像方向编号；ti,m,evo,l表示第evo次进化时父代种群中第l个个体的第i颗卫星中的第k个条带端点成像时刻的归一化系数。Among them,evo represents theevo- th evolution using the NSGA-II method based on partial ordering,evo ∈ {1,2,…,evo ,…,Evos },l representsthe N parent populations in the evo-th evolution Thel- th individual,l ∈ {1,2,…,l ,…N };Sj,evo,l represents the band selection of thej -th band of thel -th individual in the parent population at theevo- th evolution. Star number;Pi,evo,l represents the strip imaging action sequence number of the strip contained in thei- th satellite of thel-th individual in the parent population at theevo-th evolution;Qi,k,evo,l represents the Thestrip imaging direction number of thej- th strip in the i-th satellite of thel -th individual in the parent population at the time of evo evolution;ti,m,evo,l represents the j-th band in the parent population at the time ofevo evolution. The normalization coefficient of the imaging time of thek -th strip endpoint in the i-th satellite ofl individuals.
而且，所述计算当前父代种群每个个体的规划方案，并得到当前种群中每个个体的适应度值实现如下，Moreover, the calculation of the planning plan for each individual in the current parent population and obtaining the fitness value of each individual in the current population is implemented as follows:
根据决策变量Sj,evo,l、Pi,evo,l、Qi,k,evo,l、ti,m,evo,l，利用所构建基于偏序的Pareto多目标成像任务规划模型和所建姿态机动角速度与影像MTF之间的映射关系，确定在evo次进化过程中的规划方案，得到当前种群中每个个体的适应度值Fitevo,l；According to the decision variablesSj,evo,l ,Pi,evo,l ,Qi,k,evo,l ,ti,m,evo,l , the Pareto multi-objective imaging task planning model based on partial order and The mapping relationship between the established attitude maneuvering angular velocity and the image MTF determines the planning scheme inthe evo evolution process, and obtains the fitness valueFitevo,l of each individual in the current population;
其中，Fitevo,l表示基于偏序的NSGA-II方法在evo次进化过程中第l个个体对应的适应度值。Among them,Fitevo,l represents the fitness value corresponding to thel- th individual in theevo evolution process based on the partial ordering NSGA-II method.
而且，所述进行选择、交叉和变异操作，实现方式为，Moreover, the selection, crossover and mutation operations described above are implemented as follows:
对父代种群中的决策变量Sj,evo,l、Pi,evo,l、Qi,k,evo,l、ti,m,evo,l采用选择、交叉和变异生成子代种群中的决策变量S’j,evo,l、P’i,evo,l、Q’i,k,evo,l、t’i,m,evo,l；The decision variablesSj,evo,l ,Pi,evo,l ,Qi,k,evo,l ,ti,m,evo,l in the parent population are selected, crossed and mutated to generate the offspring population. The decision variablesS'j,evo,l ,P'i,evo,l ,Q'i,k,evo,l ,t'i,m,evo,l ;
其中，S’j,evo,l表示第evo次进化时子代种群中第l个个体的第j个条带的条带选星编号；P’i,evo,l表示第evo次进化时子代种群中第l个个体的第i颗卫星所包含条带的条带成像动作序列编号；Q’i,k,evo,l表示第evo次进化时子代种群中第l个个体的第i颗卫星中的第j个条带的条带成像方向编号；t’i,m,evo,l表示第evo次进化时子代种群中第l个个体的第i颗卫星中的第k个条带端点成像时刻的归一化系数。Among them,S'j,evo,l represents the strip star selection number of thej -th strip of thel-th individual in the progeny population at theevo -th evolution;P'i,evo,l represents the child at theevo -th evolution. The strip imaging action sequence number of the strip contained in the i-th satellite of thel -th individual in the generation population;Q'i,k,evo,l represents thei -th of thel- th individual in the offspring population during theevo- th evolution The strip imaging direction number of thej -th strip in the satellite;t'i,m,evo,l represents thek -th strip in thei- th satellite of thel -th individual in the progeny population at theevo- th evolution Normalization coefficient with endpoint imaging time.
另一方面，本发明还提供一种顾及影像MTF退化的超敏捷卫星任务规划系统，包括处理器和存储器，存储器用于存储程序指令，处理器用于调用存储器中的存储指令执行如上所述的一种顾及影像MTF退化的超敏捷卫星任务规划方法。On the other hand, the present invention also provides an ultra-agile satellite mission planning system that takes into account image MTF degradation, including a processor and a memory. The memory is used to store program instructions, and the processor is used to call the storage instructions in the memory to execute the above-mentioned one. An ultra-agile satellite mission planning method that takes into account image MTF degradation.
另一方面，本发明还提供一种顾及影像MTF退化的超敏捷卫星任务规划系统，包括可读存储介质，所述可读存储介质上存储有计算机程序，所述计算机程序执行时，实现如上所述的一种顾及影像MTF退化的超敏捷卫星任务规划方法。On the other hand, the present invention also provides an ultra-agile satellite mission planning system that takes into account image MTF degradation, including a readable storage medium. A computer program is stored on the readable storage medium. When the computer program is executed, the above-mentioned steps are implemented. An ultra-agile satellite mission planning method that takes into account image MTF degradation is described.
与先前提出的专利文献相比，本发明专利的区别在于：1.本发明专利面向区域目标成像任务；2.本发明专利具有三个目标函数：最大化成像覆盖收益、最大化影像MTF和最小化成像任务执行时间；3.先前所提出的专利文献所构建的模型为优先级排序多目标成像任务规划模型，而本发明专利所构建的模型为基于偏序的Pareto多目标成像任务规划模型。4.先前所提出的专利文献所采用的模型求解算法为改进的粒子群优化算法，而本发明是在NSGA-II方法的基础上，对NSGA-II方法的非支配排序策略进行了重新设计，提出了基于偏序的NSGA-II方法对本发明专利所构建的特殊模型进行求解。Compared with previously proposed patent documents, the differences of the patent of the present invention are: 1. The patent of the present invention is oriented to the regional target imaging task; 2. The patent of the present invention has three objective functions: maximizing imaging coverage income, maximizing image MTF and minimizing optimization of the imaging task execution time; 3. The model constructed by the previously proposed patent document is a priority sorting multi-objective imaging task planning model, while the model constructed by the patent of the present invention is a Pareto multi-objective imaging task planning model based on partial ordering. 4. The model solving algorithm used in the previously proposed patent documents is an improved particle swarm optimization algorithm, and the present invention redesigns the non-dominated sorting strategy of the NSGA-II method based on the NSGA-II method. The NSGA-II method based on partial ordering is proposed to solve the special model constructed by the patent of this invention.
与现有技术相比，本发明包括以下优点和有益效果：Compared with the prior art, the present invention includes the following advantages and beneficial effects:
通过构建姿态机动角速度与影像MTF之间的映射关系，并将影像MTF纳入动中成像任务规划建模中，避免了超敏捷卫星动中成像导致所获取的影像MTF不受控问题。构建最大化成像覆盖收益、最大化影像MTF和最小化成像任务执行时间为目标函数的基于偏序的Pareto多目标成像任务规划模型，实现了在确保最大化成像覆盖收益前提下，平衡影像MTF和观测效率的一体化优化；进一步，针对所构建基于偏序的Pareto多目标成像任务规划模型的特殊性，本发明采用一种基于偏序的NSGA-II方法，在进行偏序非支配排序时优先比较种群中两个个体间的覆盖收益，确定支配关系，当且仅当两个个体间的覆盖收益相等时，在对影像MTF和任务执行时间进行支配关系的判断，有效实现了对本发明构建的规划模型的求解需求。本发明能够有效地适用于顾及动中成像影像MTF退化的超敏捷卫星多星区域成像任务规划。By constructing a mapping relationship between attitude maneuvering angular velocity and image MTF, and incorporating the image MTF into the in-motion imaging mission planning modeling, the problem of uncontrolled MTF in the acquired images caused by ultra-agile satellite in-motion imaging is avoided. A partial order-based Pareto multi-objective imaging task planning model is constructed with the objective functions of maximizing imaging coverage revenue, maximizing image MTF and minimizing imaging task execution time, achieving a balance between image MTF and imaging coverage revenue while ensuring maximum imaging coverage revenue. Integrated optimization of observation efficiency; further, in view of the particularity of the constructed Pareto multi-objective imaging task planning model based on partial ordering, the present invention adopts an NSGA-II method based on partial ordering, giving priority to partial ordering non-dominated sorting Compare the coverage benefits between two individuals in the population to determine the dominance relationship. If and only if the coverage benefits between the two individuals are equal, the dominance relationship is judged on the image MTF and task execution time, effectively realizing the construction of the present invention. Solving requirements for planning models. The present invention can be effectively applied to ultra-agile satellite multi-satellite regional imaging mission planning that takes into account MTF degradation of moving imaging images.
附图说明Description of drawings
图1为本发明实施例的步骤流程图；Figure 1 is a step flow chart of an embodiment of the present invention;
图2为本发明实施例的区域目标成像任务地理位置信息图；Figure 2 is a geographical location information diagram of the regional target imaging task according to the embodiment of the present invention;
图3为本发明实施例的成像任务条带分解结果图；Figure 3 is a diagram of the imaging task strip decomposition results according to the embodiment of the present invention;
图4为本发明实施例的成像任务规划方案的Pareto前沿面结果图。Figure 4 is a Pareto front result diagram of the imaging task planning solution according to the embodiment of the present invention.
具体实施方式Detailed ways
下面结合图1至图4和本发明实施例介绍本发明的技术方案。The technical solution of the present invention will be introduced below with reference to Figures 1 to 4 and embodiments of the present invention.
参见图1，本发明实施例提供的一种顾及动中成像影像MTF退化的超敏捷卫星多星区域成像任务规划方法，包括以下步骤：Referring to Figure 1, an embodiment of the present invention provides an ultra-agile satellite multi-satellite regional imaging mission planning method that takes into account the MTF degradation of moving imaging images, including the following steps:
步骤1，对待成像区域以条带长度总和最短为优化目标进行条带分解生成成像条带集合；Step 1: Perform strip decomposition on the area to be imaged with the shortest sum of strip lengths as the optimization goal to generate an imaging strip set;
本发明实施例所构建的顾及动中成像影像MTF退化的超敏捷卫星多星区域成像任务规划模型是在成像条带作为原子任务集基础上构建的，因此，需将区域成像任务进行预处理。实施例以条带长度总和最短为优化目标实现对区域成像任务的条带分解，为规划模型的构建提供带时间戳的成像条带集合。The ultra-agile satellite multi-satellite regional imaging mission planning model constructed by the embodiment of the present invention taking into account MTF degradation of moving imaging images is constructed on the basis of imaging strips as atomic task sets. Therefore, the regional imaging tasks need to be preprocessed. The embodiment uses the shortest sum of strip lengths as the optimization goal to implement strip decomposition of regional imaging tasks, and provides a time-stamped set of imaging strips for the construction of planning models.
步骤2，构建姿态机动角速度与影像MTF之间的映射关系；Step 2: Construct a mapping relationship between attitude maneuver angular velocity and image MTF;
实施例的步骤2优选采用的具体实现方式如下：The preferred specific implementation method of step 2 of the embodiment is as follows:
现有的“动中成像”下姿态机动角速度对影像MTF的影响研究表明，成像过程中姿态机动角速度是影响影像MTF的最重要因素。在成像时不同的姿态机动角速度下，超敏捷卫星“动中成像”具有不同的偏流角、姿态、积分时间、时延积分级数，导致所获取图像的MTF不同，一般认为，姿态机动角速度越小对应的影像MTF越高。Existing research on the impact of attitude maneuvering angular velocity on image MTF under "moving imaging" shows that attitude maneuvering angular velocity during the imaging process is the most important factor affecting image MTF. Under different attitude maneuvering angular velocities during imaging, the "moving imaging" of super-agile satellites has different deflection angles, attitude, integration time, and delay integration series, resulting in different MTFs of the acquired images. It is generally believed that the higher the attitude maneuvering angular velocity, the higher the deflection angle, attitude, integration time, and delay integration series. The smaller the corresponding image, the higher the MTF.
要确保超敏捷卫星“动中成像”时，获得高清晰度空间特性的影像，需满足一定的空间分辨率（调制传递函数MTF）和辐射分辨率（信噪比SNR）。考虑到“动中成像”时，姿态机动角速度的剧烈变化是导致所获影像MTF下降的重要因素，因此，本发明实施例优选构建姿态机动角速度与影像MTF之间的映射关系如下：To ensure that ultra-agile satellites can obtain high-definition images with spatial characteristics during "imaging in motion", certain spatial resolution (modulation transfer function MTF) and radiation resolution (signal-to-noise ratio SNR) must be met. Considering that during "imaging in motion", drastic changes in attitude maneuvering angular velocity are an important factor leading to a decrease in the MTF of the acquired image. Therefore, the embodiment of the present invention preferably constructs a mapping relationship between attitude maneuvering angular velocity and image MTF as follows:
1）TDI积分级数与影像MTF的映射关系构建1) Construction of the mapping relationship between TDI integral series and image MTF
超敏捷卫星“动中成像”时，要确保其姿态机动过程中的影像MTF需对卫星的TDI-CCD相机参数进行实时的匹配调整，使得卫星沿轨方向和垂轨方向的平均MTF满足工程需求：When super-agile satellites are "imaging in motion", to ensure that the image MTF during attitude maneuvers requires real-time matching and adjustment of the satellite's TDI-CCD camera parameters, so that the average MTF of the satellite along the orbit and perpendicular to the orbit meets engineering needs. :
MTF平均= (MTF飞行+MTF线阵) / 2MTFaverage = (MTFflight +MTFline array ) / 2
其中，MTF飞行是卫星沿轨方向的MTF值，MTF线阵是卫星垂轨方向的MTF值，MTF平均是卫星沿轨和垂轨方向的平均MTF值。Among them,MTFflight is the MTF value of the satellite in the along-orbit direction,MTFlinear array is the MTF value of the satellite in the perpendicular-orbit direction, andMTFaverage is the average MTF value of the satellite in the along-orbit and perpendicular-orbit directions.
沿轨方向和垂轨方向的MTF公式分别如下：The MTF formulas along the rail direction and perpendicular to the rail direction are as follows:
其中，MTF积分时间匹配、MTF姿态和分别是积分时间匹配、姿态和偏流角改正对应的MTF值，受遥感参数设置影响较大。MTF大气、MTF相机、MTF杂光、MTF空间环境及MTF颤振分别是大气、相机、杂光、空间环境和颤振对应的MTF值，可根据经验获得。Among them,MTFintegration time matching ,MTFattitude and They are the MTF values corresponding to integration time matching, attitude and deflection angle correction, which are greatly affected by remote sensing parameter settings.MTFatmosphere ,MTFcamera ,MTFstray light ,MTFspace environment andMTFflutter are the MTF values corresponding to the atmosphere, camera, stray light, space environment and flutter respectively, which can be obtained based on experience.
积分时间匹配、偏流角、卫星姿态对影像MTF的影响公式如下：The formula for the influence of integration time matching, deflection angle, and satellite attitude on image MTF is as follows:
其中，M表示TDI-CCD相机的积分级数、fc表示空间频率、fN表示奈奎斯特频率、表示行周期误差、/>表示理想行周期、/>表示积分时间匹配相对误差、/>表示偏流角匹配残差、f表示TDI-CCD相机焦距、N表示归一化空间频率、/>表示姿态稳定度。Among them,M represents the integration series of the TDI-CCD camera,fc represents the spatial frequency,fN represents the Nyquist frequency, Indicates row period error,/> Represents the ideal row period,/> Indicates the relative error of integration time matching,/> represents the deflection angle matching residual,f represents the focal length of the TDI-CCD camera,N represents the normalized spatial frequency, /> Indicates posture stability.
将上述公式进行结合，则可构建TDI积分级数（M）与影像MTF之间的定量关系。Combining the above formulas, the quantitative relationship between the TDI integration series (M ) and the image MTF can be constructed.
2）姿态机动角速度与TDI积分级数的映射关系构建2) Construction of the mapping relationship between attitude maneuver angular velocity and TDI integral series
动中成像时，不同的信噪比下，姿态机动角速度（w）与TDI积分级数（M）需按照定量关系进行匹配，最终确定不同信噪比下的姿态机动角速度与影像MTF之间的映射关系。When imaging in motion, under different signal-to-noise ratios, the attitude maneuvering angular velocity (w ) and the TDI integral series (M ) need to be matched according to a quantitative relationship, and the relationship between the attitude maneuvering angular velocity and the image MTF under different signal-to-noise ratios is finally determined. Mapping relations.
则可通过对上述两节映射关系的结合，最终确定不同信噪比下的姿态机动角速度与影像MTF之间的映射关系。The mapping relationship between attitude maneuvering angular velocity and image MTF under different signal-to-noise ratios can be finally determined by combining the mapping relationships in the above two sections.
步骤3，构建以最大化成像覆盖收益、最大化影像MTF和最小化成像任务执行时间为目标函数，条带选星编号、条带成像动作序列编号、条带成像方向编号、条带端点成像时刻的归一化系数为决策变量的基于偏序的Pareto多目标成像任务规划模型；Step 3: Construct the objective function to maximize imaging coverage income, maximize image MTF and minimize imaging task execution time, strip star selection number, strip imaging action sequence number, strip imaging direction number, strip endpoint imaging time A Pareto multi-objective imaging task planning model based on partial ordering where the normalization coefficient is the decision variable;
现有技术中的模型，大多采用优先级排序的多目标成像任务规划模型或基于Pareto的多目标成像任务规划模型，然而，常规用户提出的成像需求一般需优先保证对区域目标的完全成像，在考虑其他目标的最优。尤其面向超敏捷卫星动中成像的任务规划而言，在保证最大化成像覆盖收益前提下，确保影像MTF高且观测效率快是保证超敏捷卫星效能发挥的重要保障。因此，本发明专利提出基于偏序的Pareto多目标成像任务规划模型，实现成像覆盖收益最优前提下，影像MTF和观测效率均衡的一体化优化。Most of the models in the existing technology adopt a prioritized multi-target imaging mission planning model or a Pareto-based multi-target imaging mission planning model. However, the imaging requirements put forward by regular users generally require priority to ensure complete imaging of regional targets. Consider the optimality of other goals. Especially for mission planning of ultra-agile satellite imaging in motion, ensuring high image MTF and fast observation efficiency is an important guarantee for ensuring the effectiveness of ultra-agile satellites on the premise of maximizing imaging coverage benefits. Therefore, the patent of this invention proposes a Pareto multi-objective imaging task planning model based on partial ordering to achieve integrated optimization of balanced image MTF and observation efficiency under the premise of optimal imaging coverage revenue.
实施例的步骤3优选采用的具体实现方式如下：The preferred specific implementation method of step 3 of the embodiment is as follows:
考虑到，在实际的用户成像需求中，尽管其他方面(影像MTF和成像任务执行时间)也可以提高，但最大化成像覆盖收益通常是首要前提，为此，构建基于偏序的Pareto多目标成像任务规划模型，模型优先确保首要优化目标基础上，再实现两个第二优化目标的均衡。以成像条带作为所构建模型的输入，将顾及影像MTF和观测效率的成像任务规划问题转化为与成像时刻相关的多星资源分配和多条带成像动作序列确定的优化问题。本发明构建的基于偏序的Pareto多目标成像任务规划模型如下：Considering that in actual user imaging needs, although other aspects (image MTF and imaging task execution time) can also be improved, maximizing imaging coverage benefits is usually the first prerequisite. To this end, Pareto multi-objective imaging based on partial ordering is constructed In the mission planning model, the model prioritizes ensuring the primary optimization objective and then achieves a balance between the two secondary optimization objectives. Taking imaging strips as the input of the built model, the imaging mission planning problem that takes into account image MTF and observation efficiency is transformed into an optimization problem of multi-satellite resource allocation and multi-strip imaging action sequence determination related to imaging time. The Pareto multi-objective imaging task planning model based on partial order constructed by this invention is as follows:
1）决策变量1) Decision variables
第一类决策变量，条带选星编号：The first type of decision variable, strip selection star number:
其中，j表示成像条带集合中的第j个条带，SN表示成像条带集合中的条带数量，i表示卫星集合中的第i颗卫星，MN表示卫星集合中的卫星数量。Sj为条带选星编号，用于确定每颗卫星待成像的条带和条带数（subSi）。Among them,j represents thej- th strip in the imaging strip set,SN represents the number of strips in the imaging strip set,i represents thei- th satellite in the satellite set, andMN represents the number of satellites in the satellite set.Sj is the strip selection number, which is used to determine the strips and the number of strips to be imaged by each satellite (subSi ).
第二类决策变量，条带成像动作序列编号：The second type of decision variable, strip imaging action sequence number:
其中，i表示卫星集合中的第i颗卫星，MN表示卫星集合中的卫星数量，Pi为条带成像动作序列编号，用于确定Sj得到的各卫星待成像条带的条带成像动作序列（stripSi）。Among them,i represents thei- th satellite in the satellite collection,MN represents the number of satellites in the satellite collection,Pi is the strip imaging action sequence number, which is used to determine the strip imaging action of each satellite to be imaged strip obtained bySj sequence (stripSi ).
第三类决策变量，条带成像方向编号：The third type of decision variable, strip imaging direction number:
其中，i表示卫星集合中的第i颗卫星，MN表示卫星集合中的卫星数量，Qi,k为条带成像方向编号，用于确定Sj和Pi共同确定的条带成像动作序列stripSi中的各条带的成像方向，k表示stripSi中第k个条带，subSi表示stripSi中的条带数，因此，通过上述三类决策变量Sj，Pi，Qi,k，可以为各卫星分配资源并确定各卫星条带端点的成像动作序列（endpointSi）。Among them,i represents thei- th satellite in the satellite collection,MN represents the number of satellites in the satellite collection,Qi,k is the strip imaging direction number, which is used to determine the strip imaging action sequencestripS jointly determined bySj andPi The imaging direction of each strip ini ,k representsthe k -th strip instripSi ,subSi represents the number of strips instripSi , therefore, through the above three types of decision variablesSj ,Pi ,Qi,k , can allocate resources to each satellite and determine the imaging action sequence (endpointSi ) of each satellite strip endpoint.
第四类决策变量，条带端点成像时刻的归一化系数：The fourth type of decision variable, the normalized coefficient of the strip endpoint imaging moment:
其中，i表示卫星集合中的第i颗卫星，MN表示卫星集合中的卫星数量，ti,m表示条带端点成像时刻的归一化系数，用于确定由Sj，Pi，Qi,k三类决策变量确定卫星条带端点的成像动作序列（endpointSi）上每一个条带端点的成像时刻，m表示endpointSi中第m个条带端点，subSi表示条带数，2×subSi表示条带端点数。Among them,i represents thei -th satellite in the satellite collection,MN represents the number of satellites in the satellite collection,ti,m represents the normalized coefficient of the strip endpoint imaging time, which is used to determineSj ,Pi ,Qi ,k three types of decision variables determine the imaging time of each strip endpoint on the imaging action sequence (endpointSi ) of the satellite strip endpoint,m represents themth strip endpoint inendpointSi ,subSi represents the number of strips, 2×subSi represents the number of strip endpoints.
2）目标函数2) Objective function
成像任务规划模型包含三个目标函数，分别是最大化成像覆盖收益、最大化影像MTF和最小化成像任务执行时间。与敏捷卫星相比，超敏捷卫星最显著的提升是可以单次过境完成更多目标的成像，这从本质上提高了其观测效率。用户的成像需求通常旨在最大化成像任务覆盖收益的前提下，提高其观测效率同时确保具有较优的影像MTF。The imaging task planning model contains three objective functions, namely maximizing imaging coverage revenue, maximizing imaging MTF, and minimizing imaging task execution time. Compared with agile satellites, the most significant improvement of ultra-agile satellites is that they can complete imaging of more targets in a single pass, which essentially improves their observation efficiency. Users' imaging needs are usually aimed at maximizing imaging task coverage benefits, improving observation efficiency while ensuring a better image MTF.
然而，对于影像MTF和成像任务执行时间这两个优化目标来说，较优的影像MTF值对应较小的姿态机动角速度，较小的成像任务执行时间对应了较大的姿态机动角速度，显然这两个优化目标为彼此存在冲突和无法进行比较的两个优化目标。因此，目标函数之间具备偏序关系。以最大化成像覆盖收益为首要优化目标，第二优化目标为最大化影像MTF和最小化任务执行时间。其中，首要优化目标与第二优化目标构成优先级排序关系。模型优先确保首要优化目标基础上，再实现两个第二优化目标的均衡。However, for the two optimization goals of image MTF and imaging task execution time, a better image MTF value corresponds to a smaller attitude maneuvering angular velocity, and a smaller imaging task execution time corresponds to a larger attitude maneuvering angular velocity. Obviously, this The two optimization objectives are two optimization objectives that conflict with each other and cannot be compared. Therefore, there is a partial order relationship between the objective functions. The primary optimization goal is to maximize imaging coverage revenue, and the second optimization goal is to maximize image MTF and minimize task execution time. Among them, the primary optimization goal and the second optimization goal form a priority ranking relationship. The model prioritizes ensuring the primary optimization objective and then achieves a balance between the two secondary optimization objectives.
首要目标函数：Primary objective function:
首要目标函数为最大化成像覆盖收益。其中，表示最大化，i表示卫星集合中的第i颗卫星，MN表示卫星集合中的卫星数量，k表示条带成像动作序列（stripSi）中第k个条带，subSi表示stripSi中的条带数，Yi,k表示stripSi中的第k个条带是否可以被观测到，如果能观测到第k个条带，则Yi,k= 1;否则，Yi,k= 0，判断依据如下：stripSi中第k-1个条带的结束端点到stripSi中第k个条带的开始端点的成像时间间隔满足姿态转换时间约束；stripSi中第k个条带的开始端点到stripSi中第k个条带的结束端点的成像时间间隔满足姿态转换时间约束。The primary objective function is to maximize imaging coverage gain. in, represents maximization,i represents thei- th satellite in the satellite set,MN represents the number of satellites in the satellite collection,k represents thek -th strip in the strip imaging action sequence (stripSi ),subSi represents the strip instripSi The number of bands,Yi,k represents whether thek -th strip instripSi can be observed. If thek -th strip can be observed, thenYi,k = 1; otherwise,Yi,k = 0, The basis for judgment is as follows: The imagingtime interval from the end endpoint of thek-1thstrip in stripSito the start endpoint of thekth strip in stripS i satisfies the attitude conversion time constraint; the start endpoint of thekth strip instripSi The imaging time interval to the end endpoint of thek -th strip instripSi satisfies the attitude conversion time constraint.
其中，i表示卫星集合中的第i颗卫星，MN表示卫星集合中的卫星数量，k表示条带成像动作序列（stripSi）中第k个条带，Ti,2k表示stripSi中第k个条带的开始端点的成像时刻，Ti,2k+1表示stripSi中第k个条带的结束端点的成像时刻，表示stripSi中第k-1个条带的结束端点的成像时刻，Yi,k表示stripSi中的第k个条带是否可以被观测到，f_tri,2k表示stripSi中第k-1个条带的结束端点与第k个条带的开始端点之间的最小姿态转换时间，表示stripSi中第k个条带的开始端点与第k个条带的结束端点之间最小姿态转换时间。Among them,i represents thei- th satellite in the satellite collection,MN represents the number of satellites in the satellite collection,k represents thek -th strip in the strip imaging action sequence (stripSi ),Ti,2k represents thek -th strip instripSi The imaging time of the start endpoint of strips,Ti,2k+1 represents the imaging time of the end endpoint of thekth strip instripSi , Indicates the imaging time of the end endpoint of thek-1th strip instripSi ,Yi,k indicates whether thek- th strip instripSi can be observed,f_tri,2k indicates thek-1th strip instripSi The minimum attitude transition time between the end endpoint of the strip and the start endpoint of thekth strip, Indicates the minimum attitude transition time between the start endpoint ofthe k -th strip and the end endpoint ofthe k -th strip instripSi .
Gi,k表示条带成像动作序列（stripSi）中第k个条带的成像覆盖收益。当stripSi中的第k个条带无法完成观测(stripSi中第k个条带不满足姿态转换时间的约束)时，成像覆盖收益表示stripSi中从第1个条带到第k-1个条带的有效覆盖面积百分比之和，即Gi,1到Gi,k-1的和。Gi,k represents the imaging coverage gain ofthe k -th strip in the strip imaging action sequence (stripSi ). When thek- th strip instripSi cannot complete the observation (thek -th strip instripSi does not satisfy the constraint of attitude conversion time), the imaging coverage gain represents the transition from the 1st strip to thek-1 instripSi The sum of the effective coverage area percentages of the strips, that is, the sum ofGi,1 toGi,k-1 .
其中，i表示卫星集合中的第i颗卫星，MN表示卫星集合中的卫星数量，k表示条带成像动作序列（stripSi）中第k个条带，Ai,k为stripSi中第k个条带所包含成像任务的有效面积，A为待观测成像任务的总面积。Among them,i represents thei- th satellite in the satellite collection,MN represents the number of satellites in the satellite collection,k represents thek -th strip in the strip imaging action sequence (stripSi ),Ai,k is thek-th strip instripSi The effective area of the imaging tasks included in the strips,A is the total area of the imaging tasks to be observed.
第二目标函数_1:Second objective function_1:
第二目标函数_1为最大化影像MTF。其中，i表示卫星集合中的第i颗卫星，MN表示卫星集合中的卫星数量，k表示条带成像动作序列（stripSi）中第k个条带，subSi表示stripSi中的条带数，Yi,k表示stripSi中的第k个条带是否可以被观测到，Ai,k为stripSi中第k个条带所包含成像任务的有效面积，MTFi,k为stripSi中第k个条带对应的影像MTF值，通过步骤3所构建的姿态机动角速度与影像MTF的映射关系确定，Ei表示stripSi中能够完成成像的条带所包含的成像任务的面积之和。The second objective function_1 is to maximize the image MTF. Among them,i represents thei- th satellite in the satellite collection,MN represents the number of satellites in the satellite collection,k represents thek -th strip in the strip imaging action sequence (stripSi ),subSi represents the number of strips instripSi ,Yi,k represents whether thek- th strip instripSi can be observed,Ai,k is the effective area of the imaging task included in thek -th strip instripSi ,MTFi,k is the effective area of the imaging task instripSi The image MTF value corresponding tothe k -th strip is determined by the mapping relationship between the attitude maneuvering angular velocity constructed in step 3 and the image MTF.Ei represents the sum of the areas of the imaging tasks contained in the strips instripSi that can complete imaging.
其中，i表示卫星集合中的第i颗卫星，k表示条带成像动作序列（stripSi）中第k个条带，subSi表示stripSi中的条带数，Yi,k表示stripSi中的第k个条带是否可以被观测到，Ai,k为stripSi中第k个条带所包含成像任务的有效面积。Among them,i represents thei -th satellite in the satellite set,k represents thek -th strip in the strip imaging action sequence (stripSi ),subSi represents the number of strips instripSi ,Yi,k representsthe stripSi Whether thek -th strip of can be observed,Ai,k is the effective area of the imaging task included in thek -th strip ofstripSi .
第二目标函数_2:Second objective function_2:
第二目标函数_2为最小化成像任务执行时间。其中，i表示卫星集合中的第i颗卫星，MN表示卫星集合中的卫星数量，m表示条带端点成像动作序列（endpointSi）中第m个条带端点，2*subSi表示endpointSi中的条带端点数，表示endpointSi中从第m-1个条带端点到第m个条带端点之间的成像时间间隔，/>，Y_tri,m表示endpointSi中从m-1个条带端点到第m个条带端点是否满足姿态转换时间约束，判断依据如下：The second objective function_2 is to minimize the imaging task execution time. Among them,i represents thei- th satellite in the satellite collection,MN represents the number of satellites in the satellite collection,m represents them -th strip endpoint in the strip endpoint imaging action sequence (endpointSi ), 2*subSi representsthe endpointSi The number of strip endpoints, Represents the imaging time interval fromthe m-1th strip endpoint to them- th stripe endpoint inendpointSi ,/> ,Y_tri,m indicates whether the attitude conversion time constraint is satisfied fromthe m-1 strip endpoint to them- th strip endpoint inendpointSi . The judgment is based on the following:
其中，i表示卫星集合中的第i颗卫星，m表示条带端点成像动作序列（endpointSi）中第m个条带端点，表示endpointSi中从第m-1个条带端点到第m个条带端点之间的成像时间间隔，Y_tri,m表示endpointSi中从m-1个条带端点到第m个条带端点是否满足姿态转换时间约束，f_tri,m表示endpointSi中从m-1个条带端点到第m个条带端点之间的最小姿态转换时间Among them,i represents thei- th satellite in the satellite set,m representsthe m -th strip endpoint in the strip endpoint imaging action sequence (endpointSi ), represents the imaging time interval from them-1 strip endpoint to them- th strip endpoint inendpointSi ,Y_tri,m representsthe m-1 strip endpoint to them- th strip endpoint inendpointSi Whether the attitude conversion time constraint is met,f_tri,m represents the minimum attitude conversion time fromthe m-1 strip endpoint to them- th strip endpoint inendpointSi
3）约束条件：3) Constraints:
成像时间窗口约束：Imaging time window constraints:
约束条件为成像时间窗口约束，表示条带端点成像动作序列（endpointSi）上第m个条带端点的成像时刻必须满足成像时间窗口约束，其中，Ti,m为endpointSi上第m个条带端点的成像时刻。[ITWi,m-s,ITWi,m-e]表示endpointSi上第m个条带端点的成像时间窗口，ITWi,m-s为条带端点的成像时间窗口的起始时刻，ITWi,m-e为条带端点的成像时间窗口的终止时刻。The constraint is the imaging time window constraint, which means that the imaging moment of them -th strip endpoint on the strip endpoint imaging action sequence (endpointSi ) must satisfy the imaging time window constraint, whereTi,m isthe m -th strip onendpointSi Imaging moment with endpoints. [ITWi,ms ,ITWi,me ] represents the imaging time window ofthe m -th strip endpoint onendpointSi ,ITWi,ms is the starting time of the imaging time window of the strip endpoint,ITWi,me is the strip The ending moment of the imaging time window for the endpoint.
姿态转换时间约束：Attitude transformation time constraints:
约束条件为姿态转换时间约束，表示条带端点成像动作序列（endpointSi）上相邻两个条带端点之间的成像时刻之差（）必须满足相邻两个条带端点之间的姿态转换时间约束（/>）。The constraint is the attitude conversion time constraint, which represents the difference in imaging time between two adjacent strip endpoints on the strip endpoint imaging action sequence (endpointSi ) ( ) must satisfy the attitude conversion time constraint between two adjacent strip endpoints (/> ).
步骤4，针对所构建模型的特殊性，利用一种基于偏序的NSGA-II方法求解模型，该方法优先比较种群中的两个个体之间的覆盖收益进行非支配排序，当且仅当两个个体之间的覆盖收益相等时，再对影像MTF和成像任务执行时间进行非支配排序；Step 4: Based on the particularity of the constructed model, a NSGA-II method based on partial ordering is used to solve the model. This method prioritizes comparing the coverage benefits between two individuals in the population for non-dominated sorting if and only if the two individuals When the coverage gains between individuals are equal, non-dominated sorting is performed on the image MTF and imaging task execution time;
与传统NSGA-II方法相比，本发明专利所提出的基于偏序的NSGA-II方法的区别在于，本发明专利在NSGA-II方法的基础之上，对NSGA-II方法的非支配排序策略进行了改进。该方法不再对模型中的所有目标函数进行非支配排序而是基于步骤3中所构建模型的特殊性，优先比较最大化成像覆盖收益这一目标函数的支配关系，当成像覆盖收益相同时，再判断最大化影像MTF和最小化成像任务执行时间这两个目标函数的支配关系。Compared with the traditional NSGA-II method, the difference between the NSGA-II method based on partial order proposed by the patent of the present invention is that the patent of the present invention adopts the non-dominated sorting strategy of the NSGA-II method based on the NSGA-II method. Improvements have been made. This method no longer performs a non-dominated sorting of all objective functions in the model but is based on the particularity of the model constructed in step 3, giving priority to the dominance relationship of the objective function that maximizes the imaging coverage income. When the imaging coverage income is the same, Then determine the dominance relationship between the two objective functions of maximizing image MTF and minimizing imaging task execution time.
实施例的步骤4优选采用的具体实现方式如下：The preferred specific implementation method of step 4 of the embodiment is as follows:
具体实现方式如下：The specific implementation is as follows:
步骤4.1，对父代种群中的个体进行随机初始化：Step 4.1, randomly initialize the individuals in the parent population:
设置种群规模为N，生成N个初始父代种群，每个个体代表一种顾及动中成像影像MTF退化的超敏捷卫星多星区域成像任务规划方案，最大进化次数设置为Evos；Set the population size toN and generateN initial parent populations. Each individual represents a super-agile satellite multi-satellite regional imaging mission planning scheme that takes into account the MTF degradation of moving imaging images. The maximum number of evolutions is set toEvos ;
其中，粒子的位置表示步骤4中所构建的基于偏序的Pareto多目标成像任务规划模型所述的决策变量；即Sj,evo,l、Pi,evo,l、Qi,k,evo,l、ti,m,evo,l，各粒子的初始值在其各自的取值范围内进行随机取值。Among them, the position of the particle represents the decision variable described in the Pareto multi-objective imaging task planning model based on partial order constructed in step 4; that is,Sj,evo,l ,Pi,evo,l ,Qi,k,evo ,l ,ti,m,evo,l , the initial value of each particle is randomly selected within its respective value range.
其中，evo表示采用基于偏序的NSGA-II方法进行的第evo次进化，evo∈{1,2,…,evo,…,Evos}，l表示第evo次进化时N个父代种群中的第l个个体，l∈{1,2,…,l,…N}；Sj,evo,l表示第evo次进化时父代种群中第l个个体的第j个条带的条带选星编号；Pi,evo,l表示第evo次进化时父代种群中第l个个体的第i颗卫星所包含条带的条带成像动作序列编号；Qi,k,evo,l表示第evo次进化时父代种群中第l个个体的第i颗卫星中的第j个条带的条带成像方向编号；ti,m,evo,l表示第evo次进化时父代种群中第l个个体的第i颗卫星中的第k个条带端点成像时刻的归一化系数；Among them,evo represents theevo- th evolution using the NSGA-II method based on partial ordering,evo ∈ {1,2,…,evo ,…,Evos },l representsthe N parent populations in the evo-th evolution Thel- th individual,l ∈ {1,2,…,l ,…N };Sj,evo,l represents the band selection of thej -th band of thel -th individual in the parent population at theevo- th evolution. Star number;Pi,evo,l represents the strip imaging action sequence number of the strip contained in thei- th satellite of thel-th individual in the parent population at theevo-th evolution;Qi,k,evo,l represents the Thestrip imaging direction number of thej- th strip in the i-th satellite of thel -th individual in the parent population at the time of evo evolution;ti,m,evo,l represents the j-th band in the parent population at the time ofevo evolution. The normalization coefficient of the imaging time of thek -th strip endpoint in the i-th satellite ofl individuals;
步骤4.2，计算当前父代种群中每个个体的规划方案，并得到当前种群中每个个体的适应度值，所述适应度值即为成像覆盖收益、影像MTF和成像任务执行时间。Step 4.2: Calculate the planning plan of each individual in the current parent population, and obtain the fitness value of each individual in the current population. The fitness value is the imaging coverage income, imaging MTF and imaging task execution time.
所述计算当前父代种群每个个体的规划方案，并得到当前种群中每个个体的适应度值为：The calculation plan of each individual in the current parent population is calculated, and the fitness value of each individual in the current population is obtained:
根据决策变量Sj,evo,l、Pi,evo,l、Qi,k,evo,l、ti,m,evo,l，利用步骤3构建的模型和步骤2构建的映射关系，确定在evo次进化过程中的规划方案，得到当前种群中每个个体的适应度值Fitevo,l；According to the decision variablesSj,evo,l ,Pi,evo,l ,Qi,k,evo,l ,ti,m,evo,l , using the model constructed in step 3 and the mapping relationship constructed in step 2, determine Through the planning scheme inthe evo evolution process, the fitness valueFitevo,l of each individual in the current population is obtained;
其中，Fitevo,l表示基于偏序的NSGA-II方法在evo次进化过程中第l个个体对应的适应度值；Among them,Fitevo,l represents the fitness value corresponding to thelth individual in theevo evolution process based on the partial ordering NSGA-II method;
步骤4.3，进行基于偏序的非支配排序和拥挤度距离计算；Step 4.3, perform non-dominated sorting and crowding distance calculation based on partial ordering;
所述基于偏序的非支配排序如下：The non-dominated sorting based on partial ordering is as follows:
相较于其他的多目标求解方法，如PESA-II和SPEA2，NSGA-II方法因其在求解效率、收敛性及解集的分布性等方面具有最佳的表现，同时，考虑到步骤3所构建规划模型的特殊性，需优先确保首要优化目标，即在确保最大化成像覆盖收益前提下，实现影像MTF和成像任务执行时间的均衡优化。然而，传统NSGA-II方法是对所有的目标函数进行非支配排序，无法实现对步骤3中所构建特殊模型求解。因此，本发明提出基于偏序的NSGA-II方法，在求解时，对NSGA-II方法的非支配排序策略进行重新设计，构建基于偏序的非支配排序策略；Compared with other multi-objective solving methods, such as PESA-II and SPEA2, the NSGA-II method has the best performance in terms of solving efficiency, convergence and distribution of solution sets. At the same time, taking into account the requirements in step 3 The particularity of building a planning model requires priority to ensure the primary optimization goal, that is, to achieve balanced optimization of image MTF and imaging task execution time while ensuring maximum imaging coverage revenue. However, the traditional NSGA-II method performs non-dominated sorting on all objective functions and cannot solve the special model constructed in step 3. Therefore, the present invention proposes the NSGA-II method based on partial ordering. When solving, the non-dominated sorting strategy of the NSGA-II method is redesigned to construct a non-dominated sorting strategy based on partial ordering;
其中，基于偏序的非支配排序采用的方式如下：优先比较种群中的两个个体之间的覆盖收益，当个体l的覆盖收益Fitevo,l,cov小于个体r的覆盖收益Fitevo,r,cov时，个体r支配个体l；当个体l的覆盖收益Fitevo,l,cov大于个体r的覆盖收益Fitevo,r,cov时，个体l支配个体r；当个体l和个体r的覆盖收益相等时，即Fitevo,l,cov=Fitevo,r,cov，再对第二目优化目标（最大化影像MTF与最小化任务执行时间）两者的支配关系进行判断，从而实现个体的非支配排序；Among them, the method of non-dominated sorting based on partial order is as follows: first compare the coverage income between two individuals in the population. When the coverage incomeFitevo,l,cov of individuall is less than the coverage incomeFitevo,r of individual r, cov , individualr dominates individuall ; when individuall' s coverage incomeFitevo,l,cov is greater than individualr 's coverage incomeFitevo,r,cov, individuall dominates individualr ; when individuall and individualr 's coverage When the benefits are equal, that is,Fitevo,l,cov =Fitevo,r,cov , then the dominance relationship between the second optimization goal (maximizing image MTF and minimizing task execution time) is judged, so as to achieve individual non-dominated sorting;
其中，Fitevo,l,cov和Fitevo,r,cov分别表示基于偏序的NSGA-II方法在evo次进化过程中第l个个体和第r个个体对应的适应度值中的覆盖收益。Among them,Fitevo,l,cov andFitevo,r,cov respectively represent the coverage gains of the NSGA-II method based on partial ordering in the fitness values corresponding to thel- th individual and ther- th individual in theevo evolution process.
所述拥挤度距离计算如下：The crowding distance is calculated as follows:
执行完基于偏序的非支配排序后，采用传统NSGA-II方法中的拥挤度距离计算方法对具有相同非支配排序层的个体进行选择性的排序，本发明不予赘述；After performing the non-dominated sorting based on partial ordering, the crowding distance calculation method in the traditional NSGA-II method is used to selectively sort the individuals with the same non-dominated sorting layer, which will not be described in detail in the present invention;
步骤4.4，进行选择、交叉和变异操作；Step 4.4, perform selection, crossover and mutation operations;
所述进行选择、交叉和变异操作为：The selection, crossover and mutation operations are as follows:
对父代种群中的决策变量Sj,evo,l、Pi,evo,l、Qi,k,evo,l、ti,m,evo,l采用传统NSGA-II方法的选择、交叉和变异操作生成子代种群中的决策变量S’j,evo,l、P’i,evo,l、Q’i,k,evo,l、t’i,m,evo,l；For the decision variablesSj,evo,l ,Pi,evo,l ,Qi,k,evo,l ,ti,m,evo,l in the parent population, the selection, crossover and sum of traditional NSGA-II methods are used. The mutation operation generates the decision variablesS'j,evo,l ,P'i,evo,l ,Q'i,k,evo,l ,t'i,m,evo,l in the offspring population;
其中，S’j,evo,l表示第evo次进化时子代种群中第l个个体的第j个条带的条带选星编号；P’i,evo,l表示第evo次进化时子代种群中第l个个体的第i颗卫星所包含条带的条带成像动作序列编号；Q’i,k,evo,l表示第evo次进化时子代种群中第l个个体的第i颗卫星中的第j个条带的条带成像方向编号；t’i,m,evo,l表示第evo次进化时子代种群中第l个个体的第i颗卫星中的第k个条带端点成像时刻的归一化系数；Among them,S'j,evo,l represents the strip star selection number of thej -th strip of thel-th individual in the progeny population at theevo -th evolution;P'i,evo,l represents the child at theevo -th evolution. The strip imaging action sequence number of the strip contained in the i-th satellite of thel -th individual in the generation population;Q'i,k,evo,l represents thei -th of thel- th individual in the offspring population during theevo- th evolution The strip imaging direction number of thej -th strip in the satellite;t'i,m,evo,l represents thek -th strip in thei- th satellite of thel -th individual in the progeny population at theevo- th evolution Normalization coefficient with endpoint imaging moment;
步骤4.5，对父代种群和子代种群进行合并；Step 4.5, merge the parent population and the offspring population;
对父代种群和子代种群进行合并形成大小为2N的种群Nt；Merge the parent population and the offspring population to form a populationNt of size2N ;
步骤4.6，进行基于偏序的非支配排序和拥挤度距离计算，生成新的种群；Step 4.6: Perform non-dominated sorting and crowding distance calculation based on partial order to generate a new population;
对合并形成的规模为2N的种群Nt利用步骤4.3所描述的方法，进行基于偏序的非支配排序和拥挤度距离计算，生成新的种群。Use the method described in step 4.3 for the2N populationNt formed by the merger to perform non-dominated sorting and crowding distance calculation based on partial ordering to generate a new population.
步骤4.7，进行优化迭代，获取在确保最大化成像覆盖收益前提下，影像MTF和成像任务执行时间均衡优化的成像任务规划方案；Step 4.7, perform optimization iterations to obtain an imaging task planning solution that balances and optimizes the image MTF and imaging task execution time while ensuring maximum imaging coverage income;
在不满足进化迭代终止条件的情况下，多次执行步骤4.4至4.6，直至达到所设置的最大进化次数Evos，从而得到对目标函数进行非支配排序后的Pareto前沿面上的一组解，用户可以根据自身需求，选择符合用户需求的一个解，即为最优的顾及动中成像影像MTF退化的超敏捷卫星多星区域成像任务规划方案。When the evolutionary iteration termination conditions are not met, steps 4.4 to 4.6 are executed multiple times until the set maximum number of evolutionsEvos is reached, thereby obtaining a set of solutions on the Pareto front after non-dominated sorting of the objective function. The user You can choose a solution that meets the user's needs based on your own needs, which is the optimal ultra-agile satellite multi-satellite regional imaging mission planning solution that takes into account the MTF degradation of moving imaging images.
步骤5，获得顾及动中成像影像MTF退化的超敏捷卫星多星区域成像任务规划方案。Step 5: Obtain a super-agile satellite multi-satellite regional imaging mission planning scheme that takes into account the MTF degradation of moving imaging images.
实施例的步骤5优选采用的具体实现方式如下：The preferred specific implementation method of step 5 of the embodiment is as follows:
通过上述步骤对模型的构建与求解，得到顾及动中成像影像MTF退化的超敏捷卫星多星区域成像任务规划方案。所得到的规划方案包括：Through the construction and solution of the model through the above steps, a super-agile satellite multi-satellite regional imaging mission planning scheme that takes into account the MTF degradation of moving imaging images is obtained. The resulting plans include:
（1）卫星资源配置（每颗卫星分配得到待成像的原子任务集中的成像条带）；(1) Satellite resource configuration (each satellite is assigned an imaging strip focused on the atomic mission to be imaged);
（2）单星成像动作序列（每颗卫星的条带端点成像动作序列）；(2) Single-satellite imaging action sequence (strip endpoint imaging action sequence of each satellite);
（3）观测起始时间；(3) Observation start time;
（4）观测终止时间；(4) Observation termination time;
（5）成像覆盖收益；(5) Imaging coverage income;
（6）影像MTF；(6) Image MTF;
（6）成像任务执行时间。(6) Imaging task execution time.
具体实施时，以上流程可采用计算机软件技术实现自动运行流程。相应的运行流程的系统装置也应当在本发明的保护范围内。During specific implementation, the above process can use computer software technology to realize the automatic operation process. The corresponding system device for running the process should also be within the protection scope of the present invention.
为便于实施参考起见，现提出一个示例，本发明实施例采用4颗超敏捷卫星和一个待成像的区域目标成像任务来说明本发明的具体实施效果。各卫星的轨道参数如表1所示，区域目标成像任务的地理位置信息如表2所示和图2所示。计算步骤2所述姿态机动角速度与影像MTF的映射关系中，影响影像MTF的相关因素如表3所示，空间相机参数如表4所示。For ease of implementation and reference, an example is now provided. This embodiment of the present invention uses four ultra-agile satellites and an imaging mission of a regional target to be imaged to illustrate the specific implementation effect of the present invention. The orbital parameters of each satellite are shown in Table 1, and the geographical location information of the regional target imaging mission is shown in Table 2 and Figure 2. In the mapping relationship between the attitude maneuvering angular velocity and the image MTF described in the calculation step 2, the relevant factors affecting the image MTF are shown in Table 3, and the space camera parameters are shown in Table 4.
表1卫星参数Table 1 Satellite parameters
表2区域目标成像任务属性信息Table 2 Regional target imaging task attribute information
表3影响影像MTF的因素Table 3 Factors affecting image MTF
表4空间相机参数Table 4 Space camera parameters
则经过步骤1将区域目标成像任务预处理为成像条带作为模型输入的原子任务集如图3所示。将带时间戳的成像条带作为模型的输入，经过步骤2的姿态机动角速度与影像MTF的映射关系构建、步骤3的基于偏序的Pareto多目标成像任务规划模型的构建和步骤4的求解，得到最终的一组顾及动中成像影像MTF退化的超敏捷卫星多星区域成像任务规划方案，如表5所示（统计Pareto前沿面上最优MTF、最小成像任务执行时间和MTF=0.1时的三组Pareto解），求解得到一组Pareto前沿面上的解如图4所示。After step 1, the regional target imaging task is preprocessed into imaging strips as the atomic task set as model input, as shown in Figure 3. Using the time-stamped imaging strip as the input of the model, through the construction of the mapping relationship between the attitude maneuvering angular velocity and the image MTF in step 2, the construction of the Pareto multi-target imaging mission planning model based on partial order in step 3, and the solution of step 4, The final set of ultra-agile satellite multi-satellite regional imaging mission planning solutions that take into account the MTF degradation of imaging images in motion is obtained, as shown in Table 5 (statistics of the optimal MTF, minimum imaging mission execution time and MTF=0.1 on the Pareto frontier Three sets of Pareto solutions), and a set of solutions on the Pareto front surface are obtained as shown in Figure 4.
表5顾及动中成像影像MTF退化的超敏捷卫星多星区域成像任务规划方案Table 5 Ultra-agile satellite multi-satellite regional imaging mission planning scheme taking into account MTF degradation of moving imaging images
参见图4及表5，可实现在确保最大化成像覆盖收益前提下，影像MTF和观测效率均衡的一体化优化，用户可根据成像需求选择合适的规划方案。Referring to Figure 4 and Table 5, the integrated optimization of balanced image MTF and observation efficiency can be achieved while ensuring maximum imaging coverage benefits. Users can choose an appropriate planning solution based on imaging needs.
在一些可能的实施例中，本发明还提供了一种顾及动中成像影像MTF退化的超敏捷卫星多星区域成像任务规划系统，包括如下模块：In some possible embodiments, the present invention also provides an ultra-agile satellite multi-satellite regional imaging mission planning system that takes into account MTF degradation of moving imaging images, including the following modules:
任务规划模型构建模块，构建以最大化成像覆盖收益、最大化影像MTF和最小化成像任务执行时间为目标函数，条带选星编号、条带成像动作序列编号、条带成像方向编号、条带端点成像时刻的归一化系数为决策变量的基于偏序的Pareto多目标成像任务规划模型；The mission planning model building module constructs the objective function of maximizing imaging coverage income, maximizing image MTF and minimizing imaging task execution time, strip star selection number, strip imaging action sequence number, strip imaging direction number, strip A Pareto multi-objective imaging task planning model based on partial ordering in which the normalized coefficient of the endpoint imaging moment is the decision variable;
任务规划模型求解模块，针对所构建模型的特殊性，确保在最大化成像覆盖收益前提下，实现影像MTF和成像任务执行时间均衡的一体化优化，构建基于偏序的NSGA-II方法对模型进行求解，确定任务规划方案。The task planning model solving module, based on the particularity of the constructed model, ensures that the integrated optimization of image MTF and imaging task execution time balance is achieved under the premise of maximizing the imaging coverage income, and constructs the NSGA-II method based on partial order to evaluate the model. Solve the problem and determine the mission planning plan.
各模块的具体处理过程可参见以上方法实施例相应步骤，本发明不予赘述。The specific processing procedures of each module can be found in the corresponding steps of the above method embodiments, and will not be described in detail in the present invention.
在一些可能的实施例中，提供一种顾及影像MTF退化的超敏捷卫星任务规划系统，包括处理器和存储器，存储器用于存储程序指令，处理器用于调用存储器中的存储指令执行如上所述的一种顾及影像MTF退化的超敏捷卫星任务规划方法。In some possible embodiments, an ultra-agile satellite mission planning system that takes into account image MTF degradation is provided, including a processor and a memory, the memory is used to store program instructions, and the processor is used to call the storage instructions in the memory to execute the above An ultra-agile satellite mission planning method taking into account image MTF degradation.
在一些可能的实施例中，提供一种顾及影像MTF退化的超敏捷卫星任务规划系统，包括可读存储介质，所述可读存储介质上存储有计算机程序，所述计算机程序执行时，实现如上所述的一种顾及影像MTF退化的超敏捷卫星任务规划方法。In some possible embodiments, an ultra-agile satellite mission planning system that takes into account image MTF degradation is provided, including a readable storage medium, and a computer program is stored on the readable storage medium. When the computer program is executed, the above is implemented The above-mentioned ultra-agile satellite mission planning method takes into account the degradation of image MTF.
本文中所描述的具体实施例仅是对本发明精神作举例说明。本发明所属技术领域的技术人员可以对所描述的具体实施例做各种各样的修改或补充或采用类似的方式代替，但并不会偏离本发明的精神或者超越所附权利要求书所定义的范围。The specific embodiments described herein are merely illustrative of the spirit of the invention. Those skilled in the technical field to which the present invention belongs can make various modifications or additions to the described specific embodiments or use similar methods to replace them, but this will not deviate from the spirit of the present invention or exceed the definition of the appended claims. range.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311636029.4ACN117332624B (en) | 2023-12-01 | 2023-12-01 | Hypersensitivity satellite task planning method and system considering image MTF degradation |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311636029.4ACN117332624B (en) | 2023-12-01 | 2023-12-01 | Hypersensitivity satellite task planning method and system considering image MTF degradation |
| Publication Number | Publication Date |
|---|---|
| CN117332624A CN117332624A (en) | 2024-01-02 |
| CN117332624Btrue CN117332624B (en) | 2024-03-08 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311636029.4AActiveCN117332624B (en) | 2023-12-01 | 2023-12-01 | Hypersensitivity satellite task planning method and system considering image MTF degradation |
| Country | Link |
|---|---|
| CN (1) | CN117332624B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117575371B (en)* | 2024-01-16 | 2024-04-05 | 中国人民解放军战略支援部队航天工程大学 | A multi-satellite imaging mission planning method and system based on human-computer interaction |
| CN120429832B (en)* | 2025-07-04 | 2025-09-02 | 交通运输部天津水运工程科学研究所 | Underwater-based multi-source data feature fusion method |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB201717125D0 (en)* | 2016-11-28 | 2017-11-29 | National Univ Of Defense Technology | Differential evolution method oriented to agile satellite multi-target task planning |
| GB201717138D0 (en)* | 2016-11-28 | 2017-12-06 | National Univ Of Defense Technology | Spark-based imaging satellite task preprocessing parallelization method |
| GB201719540D0 (en)* | 2016-11-28 | 2018-01-10 | National Univ Of Defence Technology | Autonomous Operation and control system of agileimaging satellite and operation process thereof |
| AU2020100051A4 (en)* | 2020-01-10 | 2020-02-13 | Jin, Yawen Miss | Method of the mission planning for the communication between the small satellite clusters |
| CN111666661A (en)* | 2020-05-21 | 2020-09-15 | 武汉大学 | Method and system for planning imaging multi-strip splicing task in single track of agile satellite |
| CN111861230A (en)* | 2020-07-24 | 2020-10-30 | 二十一世纪空间技术应用股份有限公司 | A multi-satellite collaborative observation mission planning method and device for disaster emergency |
| CN112149911A (en)* | 2020-09-30 | 2020-12-29 | 武汉大学 | Hypersensitive short satellite same-orbit multipoint target in-motion imaging task planning method |
| CN115034557A (en)* | 2022-05-02 | 2022-09-09 | 哈尔滨工程大学 | Agile satellite emergency task planning method |
| CN115688568A (en)* | 2022-10-21 | 2023-02-03 | 武汉大学 | Scheduling method for multi-satellite regional imaging tasks of ultra-agile satellites |
| CN116242317A (en)* | 2022-11-24 | 2023-06-09 | 中国空间技术研究院 | An Imaging Method for Agile Optical Satellite Line-array Imaging |
| CN116307535A (en)* | 2023-02-13 | 2023-06-23 | 中国人民解放军战略支援部队航天工程大学 | A Multi-satellite Collaborative Imaging Mission Planning Method Based on Improved Differential Evolution Algorithm |
| CN116341221A (en)* | 2023-03-09 | 2023-06-27 | 长光卫星技术股份有限公司 | Multi-star joint region coverage planning method based on improved multi-objective evolutionary algorithm |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB201717125D0 (en)* | 2016-11-28 | 2017-11-29 | National Univ Of Defense Technology | Differential evolution method oriented to agile satellite multi-target task planning |
| GB201717138D0 (en)* | 2016-11-28 | 2017-12-06 | National Univ Of Defense Technology | Spark-based imaging satellite task preprocessing parallelization method |
| GB201719540D0 (en)* | 2016-11-28 | 2018-01-10 | National Univ Of Defence Technology | Autonomous Operation and control system of agileimaging satellite and operation process thereof |
| AU2020100051A4 (en)* | 2020-01-10 | 2020-02-13 | Jin, Yawen Miss | Method of the mission planning for the communication between the small satellite clusters |
| CN111666661A (en)* | 2020-05-21 | 2020-09-15 | 武汉大学 | Method and system for planning imaging multi-strip splicing task in single track of agile satellite |
| CN111861230A (en)* | 2020-07-24 | 2020-10-30 | 二十一世纪空间技术应用股份有限公司 | A multi-satellite collaborative observation mission planning method and device for disaster emergency |
| CN112149911A (en)* | 2020-09-30 | 2020-12-29 | 武汉大学 | Hypersensitive short satellite same-orbit multipoint target in-motion imaging task planning method |
| CN115034557A (en)* | 2022-05-02 | 2022-09-09 | 哈尔滨工程大学 | Agile satellite emergency task planning method |
| CN115688568A (en)* | 2022-10-21 | 2023-02-03 | 武汉大学 | Scheduling method for multi-satellite regional imaging tasks of ultra-agile satellites |
| CN116242317A (en)* | 2022-11-24 | 2023-06-09 | 中国空间技术研究院 | An Imaging Method for Agile Optical Satellite Line-array Imaging |
| CN116307535A (en)* | 2023-02-13 | 2023-06-23 | 中国人民解放军战略支援部队航天工程大学 | A Multi-satellite Collaborative Imaging Mission Planning Method Based on Improved Differential Evolution Algorithm |
| CN116341221A (en)* | 2023-03-09 | 2023-06-27 | 长光卫星技术股份有限公司 | Multi-star joint region coverage planning method based on improved multi-objective evolutionary algorithm |
| Title |
|---|
| Integrated imaging mission planning modeling method for multi-type targets for super-agile earth observation satellite;Lu, Zezhong, et al.;《IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing》;4156-4169* |
| Multiple super-agile satellite collaborative mission planning for area target imaging;Zezhong, L. U., et al.;《International Journal of Applied Earth Observation and Geoinformation》;1-14* |
| 基于NSGA2算法的成像卫星星座多目标任务规划问题研究;黄瀚;张晓倩;;软件导刊(08);143-146* |
| 基于改进DE算法的敏捷成像卫星前摄式调度;李志亮;李小将;张东来;;系统工程与电子技术(02);353-359* |
| 考虑成像质量的敏捷卫星任务调度模型与算法;李志亮;李小将;孙伟;;宇航学报(06);590-597* |
| 面向天文观测的空间科学卫星任务规划方法研究;吴海燕;孟新;张玉珠;李立钢;;空间科学学报(05);561-568* |
| Publication number | Publication date |
|---|---|
| CN117332624A (en) | 2024-01-02 |
| Publication | Publication Date | Title |
|---|---|---|
| CN117332624B (en) | Hypersensitivity satellite task planning method and system considering image MTF degradation | |
| CN112766813B (en) | A complex task scheduling method for space-space collaborative observation | |
| CN116307535B (en) | Multi-star collaborative imaging task planning method based on improved differential evolution algorithm | |
| US12140974B1 (en) | Method for stochastic inspections on power grid lines based on unmanned aerial vehicle-assisted edge computing | |
| CN110400002B (en) | A Multi-Star Imaging Mission Planning Method | |
| CN108388958B (en) | Method and device for researching two-dimensional attitude maneuvering satellite mission planning technology | |
| CN110926480B (en) | Autonomous aggregation method for remote sensing satellite imaging tasks | |
| CN115688568A (en) | Scheduling method for multi-satellite regional imaging tasks of ultra-agile satellites | |
| CN111897640B (en) | Satellite imaging task planning method and system for regional imaging | |
| CN111178419B (en) | A multi-objective task planning method for agile remote sensing satellites based on task clustering | |
| CN114462681B (en) | Task preprocessing method and system for imaging integrated planning of multiple types of targets of hypersensitive satellites | |
| CN113448703A (en) | Unmanned aerial vehicle swarm dynamic reconnaissance task scheduling system and method based on perception array | |
| CN112149911A (en) | Hypersensitive short satellite same-orbit multipoint target in-motion imaging task planning method | |
| CN113758467A (en) | A Remote Sensing Satellite Regional Observation Method Based on Regional Division and Local Grid Nesting | |
| CN118567233A (en) | An autonomous drone transportation system based on solar power generation | |
| Wu et al. | Multiregion mission planning by satellite swarm using simulated annealing and neighborhood search | |
| CN117029823A (en) | Channel planning method based on reinforcement learning for CSI fingerprint map construction | |
| CN119759521B (en) | Optical imaging satellite task scheduling method based on hybrid heuristic model evolution algorithm | |
| CN119512207A (en) | Satellite-UAV collaborative mission planning method and system for time-sensitive targets | |
| CN116562466B (en) | Planning method for high orbit satellite observation task | |
| Huang et al. | Improving model training with multi-fidelity hyperparameter evaluation | |
| Turki et al. | Generative design for a sustainable urban morphology | |
| CN109284874A (en) | Daily generation prediction technique, device, equipment and the storage medium of photovoltaic plant | |
| CN115841229A (en) | Intelligent scheduling method of internet vehicle connecting system | |
| CN120218573B (en) | Drawing meaning force and double-Q network driven intelligent agent dynamic resource allocation method |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |