Multi-sensor cooperative scheduling method for space target detection based on multi-rule fusion

Abstract

The application relates to a multi-sensor cooperative scheduling method for space target detection based on multi-rule fusion. The method comprises the following steps: constructing constraint conditions for multi-sensor cooperative scheduling according to sensor conversion time, maximum detection capability of the sensor, required detection time of a space target, available detection window and time interval between adjacent detections of the same target; setting the sum of priorities of total detection space targets as a target function, and constructing a multi-sensor cooperative scheduling optimization model for space target detection according to the constraint conditions and the target function; solving the multi-sensor cooperative scheduling optimization model for space target detection according to an evolutionary algorithm and a heuristic rule, and performing multi-sensor cooperative scheduling by using the obtained specific start time and end time of sensor tracking and detection. The method can improve the efficiency of acquiring space target information.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of space situation awareness, in particular to a space target detection multi-sensor collaborative scheduling method and device based on multi-rule fusion, computer equipment and storage medium. BACKGROUND

[0002] With the increasing frequency of human space activities, the number of long-term resident objects in orbit is increasing, which needs to be tracked and detected by ground-based radars, optical telescopes and other sensors, so as to obtain relevant information of space targets in time. Due to the limited number and capacity of space target detection equipment, in order to more efficiently obtain space target information within a specified time, the collaborative use of space target detection equipment needs to be optimized and scheduled. Considering the space-time relationship between space targets and detection equipment, the intersection part of the running track of the space target and the detection range of the detection equipment is the detectable arc segment of the space target, so the space target detection multi-sensor collaborative scheduling problem can generally be converted into a detectable arc segment scheduling problem.

[0003] The existing researches on space target detection multi-sensor collaborative scheduling are mostly based on the strategy of "full arc segment tracking", that is, for a specified space target, once a "detectable arc segment" is assigned, the corresponding detection equipment will track and detect the target in the entire time interval corresponding to the arc segment. In fact, the "detectable arc segment" of some space targets is relatively long, far exceeding the detection time required for cataloging and orbit determination, so the "full arc segment tracking" strategy is easy to cause waste of space target detection resources. SUMMARY

[0004] Therefore, it is necessary to provide a space target detection multi-sensor collaborative scheduling method and device based on multi-rule fusion, computer equipment and storage medium, which can reduce the waste of space target detection resources and improve the efficiency of obtaining space target information.

[0005] A space target detection multi-sensor collaborative scheduling method based on multi-rule fusion, the method comprising:

[0006] obtaining the detection time required for a space target and the available detection window of the space target;

[0007] constructing constraint conditions of the available detection window of the space target according to the sensor detection distance, the sensor elevation angle range and the orbit of the space target, and constructing constraint conditions of the multi-sensor collaborative scheduling according to the sensor switching time, the maximum detection capacity of the sensor, the detection time required for the space target, the available detection window, and the time interval between adjacent detections of the same target;

[0008] The objective function for multi-sensor collaborative scheduling is to maximize the sum of priorities of all detected space targets. An optimization model for multi-sensor collaborative scheduling of space target detection is constructed based on the constraints of the available detection window, the constraints of multi-sensor collaborative scheduling, and the objective function.

[0009] The multi-sensor collaborative scheduling optimization model for space target detection is solved based on evolutionary algorithms and heuristic rules to obtain the specific start and end times of each sensor's tracking and detection.

[0010] Multi-sensor collaborative scheduling of space targets is performed based on the specific start and end times of each sensor's tracking and detection.

[0011] In one embodiment, a multi-sensor cooperative scheduling optimization model for space target detection is constructed based on constraints of the available detection window, constraints of multi-sensor cooperative scheduling, and an objective function, including:

[0012] Based on the constraints of the available detection window, the constraints of multi-sensor cooperative scheduling, and the objective function, the optimization model for multi-sensor cooperative scheduling of space target detection is constructed as follows:

[0013] ;

[0014] st ;

[0015] ;

[0016] ;

[0017] ;

[0018] ; ;

[0019] ;

[0020] in, For radar resource collection, For the target set, For sensors Maximum detection range, The pitch angle of the sensor. It is the minimum pitch angle. Indicates the first The first sensor Detection start time, This corresponds to the end time of sensor detection. For the first Device switching time for each sensor Indicates the first The sensor is for the first The time required to accumulate goals Indicates sensor Detection capabilities Indicate the target The actual start time of the probe, For the corresponding end time, For the goal Required detection time Indicates the start time of the available probe window. That is the corresponding end time. Indicates the scheduling period. This indicates the number of times the target needs to be probed within one scheduling cycle. Represents decision variables, Indicate space target Priority.

[0021] In one embodiment, the multi-sensor cooperative scheduling optimization model for space target detection is solved according to an evolutionary algorithm and heuristic rules to obtain the specific start time and end time of each sensor's tracking and detection, including:

[0022] The optimal detection window for space targets is obtained by solving the multi-sensor collaborative scheduling optimization model for space target detection using an evolutionary algorithm.

[0023] Heuristic rules are used to determine the specific start and end times for each sensor's tracking and detection from the optimal detection window of the space target.

[0024] In one embodiment, the optimal detection window for the space target is obtained by solving the multi-sensor cooperative scheduling optimization model for space target detection using an evolutionary algorithm, including:

[0025] The multi-sensor collaborative scheduling optimization model for space target detection is solved using an evolutionary algorithm to obtain multiple initial solutions to the objective function that satisfy the constraints; the initial solutions represent the available detection window required by the space target during the detection process.

[0026] The objective function is used as the fitness function of the multi-sensor collaborative scheduling optimization model for space target detection; based on the fitness function, the fitness of the initial solution is calculated.

[0027] Based on the selection operator of the evolutionary algorithm and the fitness of the initial solutions, multiple initial solutions are selected to obtain the elite solutions among the initial solutions;

[0028] Candidate solutions are obtained by performing a crossover operation on the elite solutions using a partial matching crossover operator.

[0029] The optimal solution of the multi-sensor cooperative scheduling optimization model for space target detection is obtained by selecting and mutating the candidate solution through mutation operators, including selection mutation, reverse mutation and insertion mutation. The optimal solution includes the best detection window of the space target.

[0030] In one of the embodiments, the heuristic rules include a preceding rule, a following rule and a random rule. The preceding rule is that if the best detection window is executed, the first space target needs to be observed at the beginning of the best detection window, the sensor detection start time is the start time of the corresponding best detection window, and the sensor detection end time depends on the required detection time of the space target, and the later space targets are supplemented in a preset order.

[0031] In one of the embodiments, the following rule is that the end time of the first space target is aligned with the end time of the best detection window, and then the start time of the detection of the space target is rolled back, and finally it is judged whether the last space target can be detected according to the start time and the end time of the best detection window and the required detection time of the last space target.

[0032] In one of the embodiments, the random rule is that when the first space target appears, the first space target is randomly assigned to a window for observation, and when the second space target appears, the second space target is also randomly assigned, but two windows are assigned, and the first space target and the second space target do not overlap in the detection time window.

[0033] A multi-sensor cooperative scheduling device for space target detection based on multi-rule fusion, the device comprising:

[0034] A detection data acquisition module for acquiring the required detection time of the space target and the available detection window of the space target;

[0035] A constraint condition setting module for setting the constraint conditions of the available detection window of the space target according to the sensor detection distance, the sensor pitch angle range and the space target orbit, and setting the constraint conditions of the multi-sensor cooperative scheduling according to the sensor conversion time, the maximum detection capability of the sensor, the required detection time of the space target, the available detection window and the time interval between adjacent detections of the same target;

[0036] A cooperative scheduling optimization model construction module for setting the sum of the priorities of the total detection space targets as the objective function of the multi-sensor cooperative scheduling, and constructing the space target detection multi-sensor cooperative scheduling optimization model according to the constraint conditions of the available detection window, the constraint conditions of the multi-sensor cooperative scheduling and the objective function;

[0037] A model solving and cooperative scheduling module is configured to solve the space target detection multi-sensor cooperative scheduling optimization model according to an evolutionary algorithm and heuristic rules to obtain specific start time and end time of each sensor tracking detection; and the multi-sensor cooperative scheduling is performed by using the specific start time and end time of the sensor tracking detection.

[0038] A computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0039] Obtaining a required detection time of a space target and an available detection window of the space target;

[0040] According to a sensor detection distance, a sensor elevation angle range and a space target orbit, constraint conditions of the available detection window of the space target are constructed, and according to a sensor switching time, a maximum detection capability of the sensor, the required detection time of the space target, the available detection window and a time interval between adjacent detections of the same target, constraint conditions of the multi-sensor cooperative scheduling are constructed;

[0041] Setting a sum of priorities of total detection space targets as a target function of the multi-sensor cooperative scheduling, and constructing a space target detection multi-sensor cooperative scheduling optimization model according to the constraint conditions of the available detection window, the constraint conditions of the multi-sensor cooperative scheduling and the target function;

[0042] Solving the space target detection multi-sensor cooperative scheduling optimization model according to an evolutionary algorithm and heuristic rules to obtain specific start time and end time of each sensor tracking detection;

[0043] Performing the multi-sensor cooperative scheduling on the space target according to the specific start time and end time of each sensor tracking detection.

[0044] A computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the following steps:

[0045] Obtaining a required detection time of a space target and an available detection window of the space target;

[0046] According to a sensor detection distance, a sensor elevation angle range and a space target orbit, constraint conditions of the available detection window of the space target are constructed, and according to a sensor switching time, a maximum detection capability of the sensor, the required detection time of the space target, the available detection window and a time interval between adjacent detections of the same target, constraint conditions of the multi-sensor cooperative scheduling are constructed;

[0047] Set the sum of the priorities of all the detected space targets to maximum as a target function of the multi-sensor cooperative scheduling, and construct a space target detection multi-sensor cooperative scheduling optimization model according to the constraint conditions of the available detection window, the constraint conditions of the multi-sensor cooperative scheduling and the target function;

[0048] Solve the space target detection multi-sensor cooperative scheduling optimization model according to the evolutionary algorithm and the heuristic rules, to obtain the specific start time and end time of the tracking detection of each sensor;

[0049] According to the specific start time and end time of the tracking detection of each sensor, the space target is scheduled cooperatively by the multi-sensor.

[0050] The above space target detection multi-sensor cooperative scheduling method, device, computer equipment and storage medium based on multi-rule fusion first construct the constraint conditions of the available detection window of the space target according to the sensor detection distance, the sensor pitch angle range and the space target orbit, and construct the constraint conditions of the multi-sensor cooperative scheduling according to the sensor conversion time, the maximum detection capability of the sensor, the required detection time of the space target, the available detection window and the time interval between the adjacent two detections of the same target. Set the sum of the priorities of all the detected space targets to maximum as a target function of the multi-sensor cooperative scheduling, and construct a space target detection multi-sensor cooperative scheduling optimization model according to the constraint conditions of the available detection window, the constraint conditions of the multi-sensor cooperative scheduling and the target function. Then, an evolutionary-heuristic algorithm is designed, in which the outer layer is an evolutionary optimization algorithm and the inner layer contains heuristic rules. The outer layer mainly obtains the best candidate "detectable arc segment" of the sensor through evolutionary calculation, and the inner layer determines the specific start time and end time of the tracking detection of the sensor through the assistance of the heuristic rules. The specific observation time of each space target can be effectively obtained, the waste of space target detection resources is reduced, the efficiency of obtaining space target information is improved, the total priority of the space target detection is improved, the sensitivity to the initial value of the parameter is low, the robustness is good, and the detection scheduling result of the space target can be obtained in a complex situation. BRIEF DESCRIPTION OF DRAWINGS

[0051] Figure 1 A flowchart of a space target detection multi-sensor cooperative scheduling method based on multi-rule fusion in an embodiment;

[0052] Figure 2 A schematic diagram for explaining the decision variables in an embodiment;

[0053] Figure 3 A schematic diagram of the "immediately before" rule in an embodiment;

[0054] Figure 4 A schematic diagram of the "immediately after" rule in an embodiment;

[0055] Figure 5 for another embodiment "random" rule for the schematic diagram;

[0056] Figure 6 for an embodiment of a multi-rule fusion-based space target detection multi-sensor collaborative scheduling device structure block diagram;

[0057] Figure 7 for an embodiment of the internal structure diagram of the computer device. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0059] In one embodiment, as shown in Figure 1 A multi-rule fusion-based space target detection multi-sensor collaborative scheduling method is provided, comprising the following steps:

[0060] Step 102, obtaining the required detection time of the space target and the available detection window of the space target.

[0061] Due to the particularity of the space target detection multi-sensor collaborative scheduling problem, due to the limitation of the orbit of the space target, the deployment position of the sensor and the detection range, each space target has multiple available detection time windows in a scheduling period. The purpose of collaborative scheduling is to determine which sensor, in which time period (start and end time) for each space target, and then use heuristic rules to save visible window resources and avoid resource waste. The available detection window of the space target represents the time period in which the sensor can detect the space target in a scheduling period.

[0062] Step 104, constructing the constraint conditions of the available detection window of the space target according to the sensor detection distance, the sensor pitch angle range and the orbit of the space target, and constructing the constraint conditions of the multi-sensor collaborative scheduling according to the sensor conversion time, the maximum detection capability of the sensor, the required detection time of the space target, the available detection window, and the time interval between adjacent two detections of the same target.

[0063] The multi-sensor cooperative detection scheduling problem is a complex combinatorial optimization problem, and its constraint conditions can be expanded from two aspects of sensor resources and detection targets respectively. The constraint conditions of the multi-sensor cooperative scheduling are constructed by using the sensor detection distance, the sensor elevation angle range, the space target orbit, the sensor conversion time, the maximum detection capability of the sensor, the required detection time of the space target, the available detection window and the time interval between adjacent two detections of the same target, which is beneficial to calculate the optimal cooperative scheduling scheme.

[0064] In step 106, the sum of the priorities of all detection space targets is set as the objective function of the multi-sensor cooperative scheduling, and a space target detection multi-sensor cooperative scheduling optimization model is constructed according to the constraint conditions of the available detection window, the constraint conditions of the multi-sensor cooperative scheduling and the objective function.

[0065] In step 108, the space target detection multi-sensor cooperative scheduling optimization model is solved according to the evolutionary algorithm and the heuristic rule, and the specific start time and end time of the sensor tracking detection are obtained; and the space target is cooperatively scheduled according to the specific start time and end time of each sensor tracking detection.

[0066] The evolutionary algorithm improves the application flow of the genetic operator of the traditional genetic algorithm in the multi-sensor cooperative scheduling problem adaptively. In the model solving process, when seeking the optimal solution, firstly, in order to ensure that the alleles in each individual are significantly different, the initialization population is obtained by randomly generating in the present application, and then the real number coding is adopted, and the chromosome length represents the number of space targets. If the first allele is and the second allele is , it indicates that the first space target is detected in the corresponding th arc segment. The second space target is detected in the corresponding th visible arc segment.

[0067] In the process of the evolutionary algorithm, the fitness calculation plays an important role because it affects the probability of the individual being selected to complete the genetic operation. In order to closely combine the evolutionary-heuristic algorithm with the multi-sensor scheduling, the objective function of the cooperative scheduling optimization model is used as the fitness function in this paper. Then, the elite population is selected as the individual of the next generation evolution according to the selection operator, so as to ensure that the fitness is in an upward trend, and finally the crossover and mutation are performed to obtain the best detection window of the space target; the evolutionary-heuristic algorithm set in this application is used to determine the specific start time and end time of the sensor tracking detection from the best detection window of the space target.

[0068] The space target is cooperatively detected according to the specific start time and end time of each sensor tracking detection, and the multi-sensor cooperative scheduling is realized.

[0069] In the multi-rule fusion-based space target detection multi-sensor cooperative scheduling method, first, a constraint condition of an available detection window of a space target is constructed according to a sensor detection distance and a minimum sensor pitch angle, and a constraint condition of multi-sensor cooperative scheduling is constructed according to a sensor switching time, a maximum detection capability of a sensor, a required detection time of a space target, an available detection window, and a time interval between adjacent detections of the same target. A sum of priorities of total detection space targets is set as an objective function of multi-sensor cooperative scheduling, an optimization model of space target detection multi-sensor cooperative scheduling is constructed according to the constraint condition of the available detection window, the constraint condition of multi-sensor cooperative scheduling, and the objective function, an evolutionary-heuristic algorithm with an outer layer of an evolutionary optimization algorithm and an inner layer of heuristic rules is designed, the outer layer mainly obtains a best candidate “detectable arc segment” of a sensor through evolutionary calculation, and the inner layer determines specific start and end times of sensor tracking and detection through the heuristic rules, so that specific observation times of each space target can be effectively obtained, waste of space target detection resources is reduced, space target information efficiency is improved, the total priority of space target detection is improved, the sensitivity to initial values of parameters is low, a certain robustness is provided, and the detection scheduling result of the space target can be obtained in a complex situation.

[0070] In one embodiment, the optimization model of space target detection multi-sensor cooperative scheduling is constructed according to the constraint condition of the available detection window, the constraint condition of multi-sensor cooperative scheduling, and the objective function, and includes:

[0071] The optimization model of space target detection multi-sensor cooperative scheduling constructed according to the constraint condition of the available detection window, the constraint condition of multi-sensor cooperative scheduling, and the objective function is:

[0072] ;

[0073] s.t. ;

[0074] ;

[0075] ;

[0076] ;

[0077] ;

[0078] ;

[0079] ;

[0080] wherein, is a radar resource set, For the target set, For sensors Maximum detection range, The pitch angle of the sensor. It is the minimum pitch angle. Indicates the first The first sensor Detection start time, This corresponds to the end time of sensor detection. For the first Device switching time for each sensor Indicates the first The sensor is for the first The time required to accumulate goals Indicates sensor Detection capabilities Indicate target The actual start time of the probe, For the corresponding end time, For the goal Required detection time Indicates the start time of the available probe window. That is the corresponding end time. Indicates the scheduling period. This indicates the number of times the target needs to be probed within one scheduling cycle. Represents decision variables, Indicate space target Priority.

[0081] In a specific embodiment, the decision variables of the multi-sensor collaborative scheduling optimization model for space target detection are designed. These decision variables include both integer and real number variables. 1 indicates the first The space target is in its corresponding... The first visible window The detection will be carried out at a certain time. If the value is 0, no detection is performed. For example... Figure 2 As shown, Indicates the available probe window. Indicates the first The space target is in its corresponding... A visible window. To address the problem of low utilization of arc segment resources, this invention introduces... A real number variable whose value ranges from the start time to the end time of the window. Therefore, It is a mixed integer variable, that is, it contains both integer variables and real number variables. In this invention, .

[0082] The multi-sensor cooperative detection scheduling problem is a complex combinatorial optimization problem, and its constraints can be expanded from the two aspects of sensor resources and detection targets.

[0083] Regarding sensor resources, the following constraints can be identified:

[0084] C1: Sensor detection range constraint. The radar has a limited range for detecting targets; beyond this range, the target cannot be observed.

[0085] (1)

[0086] For radar resource collection, For the target set, For sensors Maximum detection range.

[0087] C2: Minimum pitch angle constraint for the sensor. The sensor can only observe a target when it is within the available detection window. However, considering that the ground is not level, there may be mountains and other obstacles that hinder the sensor's detection, limiting the effective angle.

[0088] (2)

[0089] The pitch angle of the sensor. It is the minimum pitch angle.

[0090] C3: Sensor switching time constraint. When a sensor detects a target, there needs to be a switching time for adjacent detections by the same sensor.

[0091] (3)

[0092] Indicates the first The first sensor Detection start time, This corresponds to the end time of the detection. For the first Device switching time for each sensor.

[0093] C4: Sensor maximum detection capability constraint. For a given sensor resource, only a certain number of targets can be observed simultaneously. Assume the sensor can detect... One goal. Then:

[0094] (4)

[0095] Indicates the first The sensor is for the first accumulation time of each target, representing the detection capability of the sensor .

[0096] For each space target, the following constraints can be summarized:

[0097] C5: Detection time constraint required for space target. The shortest detection time affects the accuracy of the orbit parameters, which is related to the effective cross section of the target and the radar residence time. The shortest detection time constraint is:

[0098] (5)

[0099] representing the actual start detection time of the target , is the corresponding end time. is the detection time required for the target .

[0100] C6: Available detection window constraint. Only when the target is in the time window of the sensor can it be detected.

[0101] (6)

[0102] representing the start time of the available detection window, is the corresponding end time.

[0103] C7: Sensor detection frequency constraint for the same target. According to the characteristics of the target, the adjacent detection time of the same target cannot be too short. The detection frequency of different targets is different. When the detection times of the same target are greater than 1, the two adjacent detection times must be extended.

[0104] (7)

[0105] representing the scheduling period, representing the number of times the target needs to be detected in a scheduling period.

[0106] In the above constraints, constraints C1 and C2 need to be considered when generating time window information, and constraints C3, C4, C5, C6 and C7 need to be considered when scheduling.

[0107] In one embodiment, the space target multi-sensor collaborative scheduling optimization model is solved according to the evolutionary algorithm and heuristic rules to obtain the specific start time and end time of each sensor tracking detection, including:

[0108] Solving the space target detection multi-sensor collaborative scheduling optimization model according to the evolutionary algorithm to obtain the optimal detection window of the space target;

[0109] Determining the specific start time and end time of each sensor tracking detection from the optimal detection window of the space target by using a heuristic rule.

[0110] In one embodiment, the evolutionary algorithm is an optimized genetic algorithm; solving the space target detection multi-sensor collaborative scheduling optimization model according to the evolutionary algorithm to obtain the optimal detection window of the space target, comprising:

[0111] Solving the space target detection multi-sensor collaborative scheduling optimization model according to the evolutionary algorithm to obtain multiple initial solutions of the objective function satisfying the constraint condition; the initial solution is the available detection window required by the space target in the detection process;

[0112] Taking the objective function as the fitness function of the space target detection multi-sensor collaborative scheduling optimization model; calculating the initial solution according to the fitness function to obtain the fitness of the initial solution;

[0113] Selecting the multiple initial solutions according to the selection operator of the evolutionary algorithm and the fitness of the initial solution to obtain an elite solution in the initial solution;

[0114] Performing a crossover operation on the elite solution according to a partial match crossover operator to obtain a candidate solution;

[0115] Selecting, mutating, and inserting the candidate solution by a mutation operator to obtain the optimal solution of the space target detection multi-sensor collaborative scheduling optimization model; the optimal solution includes the optimal detection window of the space target.

[0116] In a specific embodiment, a double-cutpoint crossover is used in the present application. First, a value between 0 and 1 is randomly generated, and the relationship between the value and the crossover probability is determined. If the value is greater than the crossover probability, the crossover operation is performed, otherwise, it is not performed. Second, two individuals are randomly selected from the population, and two crossover points are randomly generated (the range of the crossover points is the length of the chromosome). Finally, the gene fragments between the two crossover points are crossed to generate two new chromosomes. In the mutation operation, three mutation methods are mainly used in the present application: selection mutation, reversal mutation, and insertion mutation. The process of selection mutation is similar to that of crossover mutation. The process of reversal mutation is as follows: after randomly selecting two cut points, the alleles between the cut points are randomly exchanged. However, since different alleles have different values, a penalty function needs to be added when calculating the fitness. Insertion mutation: it mainly refers to randomly finding several points in the chromosome, and then randomly assigning values to the alleles (the value will not exceed the number of visible windows).

[0117] In one embodiment, the heuristic rules include a first-in rule, a last-in rule, and a random rule. The first-in rule is that if the optimal detection window is executed, the first space object needs to be observed at the beginning of the optimal detection window, the sensor detection start time is the beginning time of the corresponding optimal detection window, and the sensor detection end time depends on the detection time required by the space object, and later space objects are supplemented in a preset order.

[0118] In a specific embodiment, a schematic diagram of the first-in rule is shown in FIG. 3, which indicates that if the optimal detection window is executed, the first space object needs to be observed at the beginning of the window. The sensor detection start time is the beginning time of the corresponding optimal detection window, and the sensor detection end time depends on the detection time required by the space object, and later space objects are supplemented in a preset order. This method avoids the optimization process within the time window, and significantly reduces the computational complexity. Figure 3 In one embodiment, the last-in rule is that the first space object is observed at the end of the optimal detection window, and the following space objects are sequentially arranged. Finally, it is determined whether the last space object can be detected according to the beginning time and the end time of the optimal detection window and the detection time required by the last space object.

[0119] In a specific embodiment, a schematic diagram of the last-in rule is shown in FIG. 4, which indicates that the last-in rule is similar to the first-in rule, but shows the opposite process. The end time of the first space object is aligned with the end time of the optimal detection window, and then the start time of the space object detection is rolled back. If multiple space objects are arranged for detection within the optimal detection window, these space objects should be arranged in the order of priority according to the last-in rule, starting from the high-priority object, and then arranging the detection sensor detection end time of the second high-priority object to the current last time of the optimal detection window to determine whether it can be arranged. Finally, it is determined whether the last space object can be detected according to the beginning time and the end time of the optimal detection window and the detection time required by the last space object. If it can be detected, it is detected, and if it cannot be detected, it is stopped and the sensor detection planning is re-performed.

[0120] Figure 4 In one embodiment, the random rule is that when the first space object appears, the first space object is randomly assigned to a window for observation, and when the second space object appears, the second space object is also randomly assigned, but two windows are assigned, and the first space object and the second space object do not overlap in the detection time window.

[0121] In a specific embodiment, a schematic diagram of the random rule is shown in FIG. 5, which indicates that when the first space object appears, the first space object is randomly assigned to a window for observation, and when the second space object appears, the second space object is also randomly assigned, but two windows are assigned, and the first space object and the second space object do not overlap in the detection time window.

[0122] In a specific embodiment, a schematic diagram of the random rule is shown in FIG. 5, which indicates that when the first space object appears, the first space object is randomly assigned to a window for observation, and when the second space object appears, the second space object is also randomly assigned, but two windows are assigned, and the first space object and the second space object do not overlap in the detection time window. Figure 5 ​As shown, unlike the "immediately before" rule and the "immediately after" rule, the random strategy emphasizes a random process. When the first target appears, it is randomly assigned to a window for observation. When the second target appears, it is also randomly assigned, but two windows are assigned so that they do not overlap.

[0123] The heuristic rule is used to determine the specific start time and end time of sensor tracking detection from the optimal detection window of the space target, including:

[0124] The specific observation optimal detection window corresponding to each space target can be generated by an evolutionary algorithm. For example, assuming that there are 10 space targets (in reality, there are hundreds or thousands, which are used here only for expression) and there is only one sensor (for clarity of expression). After initialization, the chromosome is , which indicates that the first target performs a detection task in the corresponding third visible window, the second target performs a detection task in the corresponding sixth optimal detection window, the third target performs a detection task in the corresponding fifth detection window, and so on. The tenth space target performs a detection task in the corresponding fourth optimal detection window.

[0125] The window generation performed above causes a certain waste of resources (for example, the first target can complete detection in a part of the corresponding third arc segment, and the remaining arc segment resources can be released). Assuming that the normalized start time and end time of the third visible window of the first target are [0.1, 0.4], the normalized start time and end time of the sixth visible window of the second target are [0.2, 0.6], and so on. The start time and end time of the fourth visible window of the tenth target are [0.3, 0.8]. The actual detection time of the first, second, and tenth targets is only 0.1 s. If the traditional method is used, only one target can be detected (because it is full-arc tracking), but if the rule is used, the actual observation time is considered, so that resources can be released.

[0126] Taking the immediately before rule as an example: the first space target appears first, and the time required for the first space target is 0.1 s, so the actual observation time of the first space target is [0.1, 0.2], so the time resource [0.2, 0.4] is released in the system, and at this time the second target is also through the "immediately before rule", the detection time of the second target is [0.2, 0.3], and the detection time of the tenth space target is [0.3, 0.4] by analogy.

[0127] If the tight-after rule is taken as an example: the first space target appears first, and the time required for the first space target is 0.1 s, so the actual observation time of the first space target is [0.3, 0.4], so the time resource of [0.1, 0.3] is released in the system, and at this time the second target is also through the "tight-after rule", the detection time of the second target is [0.2, 0.3], at this time although the tenth space target cannot be detected, but there will be other target window resources.

[0128] If the random rule is taken as an example: the first space target appears first, and the time required for the first space target is 0.1 s, so the actual observation start time of the first space target is between [0.1, 0.4], and the later targets are also taken in this way, as long as they do not coincide with the previous targets.

[0129] It should be noted that the population in each iteration in the above steps will be involved, therefore, the iteration process of the evolutionary algorithm includes the heuristic rule performed each time. This is also the core of the present application.

[0130] In one embodiment, the evolutionary-heuristic algorithm of the present application refers to the outer layer using the evolutionary algorithm to obtain the optimal detection window of the space target, and the inner layer using the heuristic rule to obtain the specific detection time of each space target. By combining the above two methods, the flow of the final evolutionary-heuristic method is shown in the following algorithm.

[0131]

[0132] Therefore, the basic parameters of the evolutionary-heuristic algorithm proposed in this paper are input during program execution, including task number, task priority, observation time, device number, time window, crossover probability and variation probability. Then, it generates initial filling and defines the executable detection window of each target. Further, based on the above solution, the fitness is calculated through the heuristic rule to determine the best individual of each iteration. If the above process is within the iteration range of the algorithm, the optimal individual and elite group are retained for each generation. The rest of the individuals are subjected to crossover, mutation and mutation operations, and the specific process of crossover and mutation is described in detail in the evolutionary algorithm flow. After the algorithm is completed, the best solution can be obtained.

[0133] In one embodiment, the effectiveness of the model and the feasibility of the method of the present application are verified by simulation. Three ground-based radars are used as ground-based sensor resources, and the detection distance of the radar is 500 km, and the sensor latitude and longitude positions are (-75.5966 o , 40.0386 o ), (15.5966 o , -30.0386 o ) and (105.597o -10.0386 o ). The number of space targets is from 500 to 1300, and the priority of each space target is randomly assigned to 0-10, while the required detection time of each space target is from 200s to 500s. The comparison algorithm mainly uses two kinds of heuristic algorithms (first-come-first-served (FCFS) improved first-come-first-served (IFCFS) and traditional genetic algorithm).

[0134] In order to verify the effectiveness of the method proposed in the application, the number of targets is divided into 500-1300, and the traditional method and the method proposed in the application are used to solve the space target detection multi-sensor collaborative scheduling problem, and the experimental results of different algorithms under different target scales are obtained as shown in Table 1.

[0135] Table 1

[0136]

[0137] As can be seen from Table 1, when the scale of the target increases, the total task priority also increases. In addition, the evolutionary-heuristic algorithm proposed in this paper can solve the space target detection multi-sensor collaborative scheduling model, and obtain a better solution than the traditional method.

[0138] It should be understood that, although each step in the flowchart of Figure 1 is shown in order according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps in may include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily sequential, but can be executed with at least part of other steps or other steps. Sub-steps or stages of the stage are alternately or alternately executed.

[0139] In addition, in order to explore the influence of algorithm parameters on experimental results, the parameter sensitivity of the evolutionary-heuristic algorithm is verified by experiments. In this part, the application mainly compares 6 groups of experimental parameters, which are , , , = [0.5 0.4 0.3 0.3], [0.5 0.5 0.4 0.4], [0.7 0.2 0.3 0.4], [0.8 0.1 0.2 0.5], [0.8 0.1 0.3 0.4], and [0.9 0.1 0.4 0.3], the experimental results under different simulation parameters are shown in Table 2 by changing the input parameters of the algorithm, wherein pc / pm / pSwap / pReversion are respectively the crossover probability, the selection mutation probability, the inversion mutation probability and the insertion mutation probability.

[0140] Table 2

[0141]

[0142] It can be seen from Table 2 that when the crossover probability, the mutation probability, the inversion probability and the insertion probability are respectively 0.9, 0.1, 0.4 and 0.3, the solution quality of the algorithm is higher. In addition, compared with the traditional genetic algorithm, the method designed in the application has better effect; and the total priority of the optimal result is within a certain range, which illustrates the robustness of the algorithm.

[0143] In one embodiment, as shown in Figure 6 a multi-sensor cooperative scheduling device for space target detection based on multi-rule fusion is provided, comprising: a detection data acquisition module 602, a constraint condition setting module 604, a cooperative scheduling optimization model construction module 606 and a model solving and cooperative scheduling module 608, wherein:

[0144] The detection data acquisition module 602 is used to acquire the detection time required by the space target and the available detection window of the space target.

[0145] The constraint condition setting module 604 is used to construct the constraint condition of the available detection window of the space target according to the sensor detection distance and the minimum pitch angle of the sensor, and construct the constraint condition of the multi-sensor cooperative scheduling according to the sensor conversion time, the maximum detection capability of the sensor, the detection time required by the space target, the available detection window and the time interval between adjacent detections of the same target.

[0146] The cooperative scheduling optimization model construction module 606 is used to set the sum of the priorities of the total detection space target as the objective function of the multi-sensor cooperative scheduling, and construct the space target detection multi-sensor cooperative scheduling optimization model according to the constraint condition of the available detection window, the constraint condition of the multi-sensor cooperative scheduling and the objective function.

[0147] The model solving and collaborative scheduling module 608 is used to solve the multi-sensor collaborative scheduling optimization model for space target detection based on evolutionary algorithms and heuristic rules, and obtain the specific start time and end time of each sensor's tracking and detection; and to perform multi-sensor collaborative scheduling using the specific start time and end time of sensor tracking and detection.

[0148] In one embodiment, the collaborative scheduling optimization model construction module 606 is further configured to construct a space target detection multi-sensor collaborative scheduling optimization model based on the constraints of the available detection window, the constraints of multi-sensor collaborative scheduling, and the objective function.

[0149] In one embodiment, the model solving and collaborative scheduling module 608 is further used to solve the multi-sensor collaborative scheduling optimization model for space target detection according to evolutionary algorithms and heuristic rules, so as to obtain the specific start and end times of sensor tracking and detection.

[0150] Specific limitations regarding the multi-sensor collaborative scheduling device for space target detection based on multi-rule fusion can be found in the limitations of the multi-sensor collaborative scheduling method for space target detection based on multi-rule fusion described above, and will not be repeated here. Each module in the aforementioned multi-sensor collaborative scheduling device for space target detection based on multi-rule fusion can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0151] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 7 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When executed by the processor, the computer program implements a multi-sensor collaborative scheduling method for space target detection based on multi-rule fusion. The display screen can be an LCD screen or an e-ink display screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0152] Those skilled in the art will understand thatFigure 7 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0153] In an embodiment, a computer device is provided, comprising a memory and a processor, the memory storing a computer program, and the processor implementing the steps of the method in the above embodiments when executing the computer program.

[0154] In an embodiment, a computer storage medium is provided, storing a computer program, and the computer program is executed by a processor to implement the steps of the method in the above embodiments.

[0155] A person of ordinary skill in the art can understand that all or part of the processes of the above-mentioned embodiments can be completed by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer readable storage medium and can include the processes of the above-mentioned embodiments when executed. Any reference to memory, storage, database or other medium used in the embodiments provided by the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct RAM bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0156] Each technical feature of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described, however, as long as the combinations of the technical features do not exist contradictions, they should be considered as the scope of the present application.

[0157] The above-described embodiments are merely illustrative of several embodiments of the present application, which are described in more detail and in a specific and detailed manner, but should not be construed as limiting the scope of the patent. It should be noted that for those skilled in the art, several modifications and improvements can be made without departing from the concept of the present application, and these are all within the scope of the present application. Therefore, the scope of protection of the patent of the present application should be subject to the appended claims.

Claims

1. A multi-sensor cooperative scheduling method based on multi-rule fusion for space target detection, characterized in that, The method comprises: acquiring a required detection time of a space target and an available detection window of the space target; constructing constraint conditions of the available detection window of the space target according to a sensor detection distance, a sensor elevation angle range and an orbit of the space target, and constructing constraint conditions of multi-sensor cooperative scheduling according to a sensor conversion time, a maximum detection capability of the sensor, the required detection time of the space target, the available detection window, and a time interval between adjacent detections of the same target; setting a sum of priorities of total detection space targets as an objective function of the multi-sensor cooperative scheduling, and constructing a space target detection multi-sensor cooperative scheduling optimization model according to the constraint conditions of the available detection window, the constraint conditions of the multi-sensor cooperative scheduling and the objective function; solving the space target detection multi-sensor cooperative scheduling optimization model according to an evolutionary algorithm and a heuristic rule, to obtain a specific start time and an end time of each sensor tracking detection; performing multi-sensor cooperative scheduling on the space target according to the specific start time and the end time of each sensor tracking detection; constructing the space target detection multi-sensor cooperative scheduling optimization model according to the constraint conditions of the available detection window, the constraint conditions of the multi-sensor cooperative scheduling and the objective function, comprises: constructing the space target detection multi-sensor cooperative scheduling optimization model according to the constraint conditions of the available detection window, the constraint conditions of the multi-sensor cooperative scheduling and the objective function, is: s.t. wherein, is a set of radar resources, is a set of targets, is a sensor a maximum detection range of the sensor, is a tilt angle of the sensor, is a minimum tilt angle, denotes the first detection start time of the first sensor, is the corresponding sensor detection end time, is the device switching time of the first sensor, denotes the accumulation time of the first sensor for the first target, denotes the detection capability of the sensor , denotes the actual start detection time of the target , is the corresponding end time, is the required detection time of the target , denotes the start time of the available detection window, is the corresponding end time, denotes the scheduling period, denotes the number of required detections of the target within one scheduling period, denotes the decision variable, denotes the priority of the spatial target .

2. The method of claim 1, wherein, solving the space target detection multi-sensor cooperative scheduling optimization model according to the evolutionary algorithm and the heuristic rule, to obtain a specific start time and a sensor detection end time of each sensor tracking detection, comprises: solving the space target detection multi-sensor cooperative scheduling optimization model according to the evolutionary algorithm, to obtain a best detection window of the space target; determining the specific start time and the end time of each sensor tracking detection from the best detection window of the space target by using the heuristic rule.

3. The method of claim 1, wherein, solving the space target detection multi-sensor cooperative scheduling optimization model according to the evolutionary algorithm, to obtain a best detection window of the space target, comprises: solving the space target detection multi-sensor cooperative scheduling optimization model according to the evolutionary algorithm, to obtain a plurality of initial solutions of the objective function that meet the constraint conditions; the initial solution is an available detection window required by the space target in a detection process; taking the objective function as a fitness function of the space target detection multi-sensor cooperative scheduling optimization model; calculating the initial solution according to the fitness function, to obtain a fitness of the initial solution; selecting a plurality of initial solutions according to a selection operator of the evolutionary algorithm and the fitness of the initial solution, to obtain an elite solution in the initial solutions; performing a crossover operation on the elite solution according to a partial match crossover operator, to obtain a candidate solution; selecting, reversing and inserting the candidate solution by using a mutation operator, to obtain an optimal solution of the space target detection multi-sensor cooperative scheduling optimization model; the optimal solution comprises the best detection window of the space target.

4. The method of claim 3, wherein, The heuristic rules include a front rule, a back rule and a random rule; the front rule is that if the optimal detection window is executed, the first space target needs to be observed at the beginning of the optimal detection window, the sensor detection start time is the start time of the corresponding optimal detection window, and the sensor detection end time depends on the required detection time of the space target, and later space targets are supplemented in a preset order.

5. The method of claim 4, wherein, The back rule is that the end time of the first space target is aligned with the end time of the optimal detection window, and then the start time of the detection of this space target is rolled back, and finally it is judged whether the last space target can be detected according to the start time and the end time of the optimal detection window and the required detection time of the last space target.

6. The method of claim 5, wherein, The random rule is that when the first space target appears, the first space target is randomly allocated to a window for observation, and when the second space target appears, the second space target is also randomly allocated, and the first space target and the second space target do not overlap in the detection time window.

7. A multi-sensor cooperative scheduling device for spatial target detection based on multi-rule fusion, characterized in that, The device comprises: A detection data acquisition module is configured to acquire the required detection time of the space target and the available detection window of the space target. A constraint condition setting module is configured to construct a constraint condition of the available detection window of the space target according to a sensor detection distance, a sensor pitch angle range and a space target orbit, and construct a constraint condition of multi-sensor collaborative scheduling according to a sensor conversion time, a sensor maximum detection capability, a required detection time of the space target, an available detection window and a time interval between adjacent detections of the same target. A collaborative scheduling optimization model construction module is configured to set the sum of the priorities of all detection space targets as a target function of the multi-sensor collaborative scheduling, and construct a space target detection multi-sensor collaborative scheduling optimization model according to the constraint condition of the available detection window, the constraint condition of the multi-sensor collaborative scheduling and the target function, including: The construction of the space target detection multi-sensor collaborative scheduling optimization model according to the constraint condition of the available detection window, the constraint condition of the multi-sensor collaborative scheduling and the target function is: s.t. wherein, is a set of radar resources, is a set of targets, is a sensor a maximum detection range of, is a pitch angle of the sensor, is a minimum pitch angle, denotes the first detection start time of the first sensor, is a corresponding sensor detection end time, is a device switching time of the first sensor, denotes the accumulation time of the first sensor for the first target, denotes the detection capability of the sensor , denotes the actual start detection time of the target , is a corresponding end time, is a required detection time of the target , denotes the start time of the available detection window, is a corresponding end time, denotes a scheduling period, denotes the number of required detections of the target within one scheduling period, denotes a decision variable, denotes a priority of the space target ; A model solving and collaborative scheduling module is configured to solve the space target detection multi-sensor collaborative scheduling optimization model according to an evolutionary algorithm and a heuristic rule, to obtain a specific start time and an end time of tracking detection of each sensor; and perform multi-sensor collaborative scheduling on the space target by using the specific start time and the end time of tracking detection of each sensor. 8.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-7. The processor executes the computer program to realize the steps of the method of any one of claims 1 to 6.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the method of any one of claims 1 to 6.

Images