Array group target data association method and system based on adaptive clustering

Abstract

The invention provides an array group target data association method and system based on adaptive clustering. The method comprises the following steps: accumulating the same load data; performing clustering processing on the load data to obtain initial array group division; screening a single clustering target set; non-array group targets are removed according to an array group judgment condition; respectively describing array group target compiling features and formation features at the front moment and the back moment, and modeling feature vectors; corresponding basic probability distribution functions are calculated according to array group target correlation matching models of compiled features, formation features and external topological structure features; evidence synthesis is carried out, and a comprehensive basic probability distribution function is established; and performing association decision on the comprehensive basic probability distribution function based on two-dimensional distribution to obtain an array group target matching result. The method can effectively reduce the influence caused by the motion state change of the array group target, the leak detection of the sensor and the observation error, and remarkably improves the correct correlation rate of the target.

Description

Technical Field

[0001] The present invention relates to the technical field of data association, and in particular to an array target data association method and system based on adaptive clustering. Background Art

[0002] Due to the limitation of the motion cycle, the satellite visits the surveillance area intermittently, and the time interval between two visits is relatively long, usually in the order of hours, so the revisit rate of satellite information is low; at the same time, due to the changes in the motion state of the array group targets, the missed detection of the sensors and the influence of observation errors, the acquired target information has certain uncertainty and incompleteness. Under the above background conditions, it is difficult to achieve stable and effective identification and data association for each target in the surveillance area.

[0003] Patent document CN117310642A (application number: CN202311239149.0) discloses a multi-radar dense group target track association method based on multi-scale clustering, which is characterized in that it includes the following steps: Step 1: Obtain radar sampling data and preprocess the data; Step 2: Estimate the track speed based on the preprocessed data to obtain the speed of the track point at each processing moment; Step 3: Obtain the predicted landing point based on the target position and the track point speed, project the track into the longitude and latitude network, perform coarse clustering based on connected domain clustering, and obtain target grouping; Step 4: For each subgroup in the target group, determine the track similarity within the group through the weighted Hausdorff distance measurement criterion to obtain a fine clustering result, that is, a preliminary association relationship between the two tracks; Step 5: According to the preliminary association relationship in step 4, select the path with the minimum weighted distance between the two targets as the optimal association relationship.

[0004] Patent document CN108665478A (application number: CN201810450301.2) discloses a multi-target data association method in a complex scene, which is characterized in that it includes the following steps: Step 1: From the binary image after the current frame image is detected, the coordinate information of all target points in the image is obtained, and the number of target points is counted; Step 2: The coordinate information of the target points is clustered into multiple target blocks; Step 3: Then the target blocks are data aligned and all target block information is eliminated and judged; Step 4: The target blocks that meet the requirements are numbered; Step 5: The number of the target block in the image is selected for tracking, that is, the position information of the target block is used for tracking.

[0005] Traditional target recognition and tracking algorithms have great limitations when dealing with situations where target location information is sparse and location accuracy errors are large, making it difficult to achieve stable and effective recognition and data association for each target in the surveillance area. Summary of the invention

[0006] In view of the defects in the prior art, an object of the present invention is to provide a method and system for associating array target data based on adaptive clustering.

[0007] The array group target data association method based on adaptive clustering provided by the present invention comprises:

[0008] Step S1: Divide the area, obtain the grid division boundary threshold, and accumulate the target position information obtained by the same load within the area;

[0009] Step S2: clustering the data of each load in the order of entering the field of view to obtain the initial array group division;

[0010] Step S3: Screening a single clustering target set and removing target points whose similarity does not meet the preset requirements;

[0011] Step S4: removing non-array group targets according to the array group determination conditions;

[0012] Step S5: describing the formation characteristics of the array targets at two moments respectively;

[0013] Step S6: describing the formation characteristics of the target group at two moments respectively;

[0014] Step S7: Modeling the feature vectors of the formation features and formation features of the array targets at two moments before and after respectively;

[0015] Step S8: Calculate the corresponding basic probability distribution function according to the array group target association matching model with the formation characteristics;

[0016] Step S9: Calculate the corresponding basic probability distribution function according to the array target association matching model of the formation characteristics;

[0017] Step S10: Calculate the corresponding basic probability distribution function according to the array group target association matching model of the external topological structure characteristics;

[0018] Step S11: Using the DS evidence theory, evidence synthesis is performed on the basic probability distribution function based on the association of formation, formation and external topological structure features to establish a comprehensive basic probability distribution function;

[0019] Step S12: Perform an association decision on the comprehensive basic probability allocation function based on the two-dimensional allocation to obtain the array group target matching result.

[0020] Preferably, step S2 comprises:

[0021] Step S2.1: Calculate the distance between every two targets and construct a distance matrix;

[0022] Step S2.2: Find the threshold of density calculation according to the distance matrix, where the number of points around each point whose distance is less than the density threshold accounts for 1%-4% of the total number of points on average;

[0023] Step S2.3: Calculate the local density of each target point and construct a local density vector;

[0024] Step S2.4: Calculate the nearest neighbor distance of each target point and construct the nearest neighbor distance vector;

[0025] Step S2.5: adaptively obtain cluster centers;

[0026] Step S2.6: classify the non-cluster center data points;

[0027] The step S2.2 comprises:

[0028] Step S2.2.1: parameter setting, including the upper and lower limits of the number ratio threshold, and the distance range threshold;

[0029] Step S2.2.2: Set the initial density threshold to the average of the maximum distance plus the minimum distance;

[0030] Step S2.2.3: Calculate the proportion of points with density less than the threshold;

[0031] Step S2.2.4: If the proportion is within the set range, output the current density threshold; otherwise, update the current maximum distance, minimum distance and density threshold;

[0032] Step S2.2.5: If the current distance range is less than the set threshold, output the current density threshold; otherwise return to step S2.3 for iterative calculation;

[0033] The step S2.4 comprises:

[0034] Step S2.4.1: sort the local density vectors from large to small according to the density value;

[0035] Step S2.4.2: Calculate the distance between each point and other points with a greater density than the point;

[0036] Step S2.4.3: Get the index of the nearest neighbor point of each point and the distance between the two points;

[0037] The step S2.5 comprises:

[0038] Step S2.5.1: Calculate a joint density vector based on the local density vector and the nearest neighbor distance vector;

[0039] Step S2.5.2: Calculate the mean and variance of the joint density vector;

[0040] Step S2.5.3: Traverse each data point and take the points that exceed the set threshold range as cluster centers;

[0041] The step S2.6 comprises:

[0042] Step S2.6.1: Initialize the cluster number of each data point, and the cluster number of non-cluster centers is uniformly set to -1;

[0043] Step S2.6.2: sort the local density vectors from large to small according to the density value;

[0044] Step S2.6.3: Update the cluster number of the non-cluster center data point to be the same as the cluster number of the point that is closest to it and has a greater density than it.

[0045] Preferably, step S3 comprises:

[0046] Step S3.1: Obtain corresponding coordinates according to the cluster center index;

[0047] Step S3.2: According to the cluster center and each target cluster number, obtain the position coordinates of each data point in the same cluster;

[0048] Step S3.3: Calculate the local density of each point in a single cluster;

[0049] Step S3.4: Calculate the similarity between each data point in the same cluster and the cluster center;

[0050] Step S3.5: Calculate the comprehensive similarity of each data point in the same cluster based on the similarity and local density;

[0051] Step S3.6: Calculate the mean and variance of the comprehensive similarity vector of data points in a single cluster;

[0052] Step S3.7: Add the data points whose comprehensive similarity values ​​exceed the threshold range as new cluster centers and update their cluster numbers.

[0053] Preferably, the step S4 comprises:

[0054] Step S4.1: Obtain the target belonging to the same cluster according to the cluster number corresponding to each data point;

[0055] Step S4.2: assembling data for a single cluster that meets the set array group construction threshold;

[0056] Step S4.3: Data points that do not meet the array group formation conditions are regarded as isolated points and no subsequent array group matching is performed.

[0057] Preferably, the step S5 comprises:

[0058] Step S5.1: Calculate the target type number of each target in each array group;

[0059] Step S5.2: Calculate the number of members of each target type contained in each array group.

[0060] Preferably, step S6 comprises:

[0061] Step S6.1: Calculate the distance between every two targets in the array;

[0062] Step S6.2: Construct the target formation feature vector of each array group.

[0063] Preferably, the step S8 comprises:

[0064] Step S8.1: Obtain the total number of target types of the two array groups and assign weights according to their importance;

[0065] Step S8.2: Normalize the weight set, the formula is:

[0066]

[0067] in, is the weight set according to the importance, N ij is the array x at time k-1 i and the array group y at time k j The total number of target types included;

[0068] Step S8.3: pre-processing the formation characteristic vectors of different array groups to make the lengths of the formation characteristic vectors of the array groups consistent;

[0069] Step S8.4: Calculate the difference in the formation characteristic vector between the array groups;

[0070] Step S8.5: Calculate the weighted inter-array formation feature difference vector;

[0071] Step S8.6: Construct a basic probability distribution function based on the feature difference vector, the formula is:

[0072]

[0073] in, is the array x at time k-1 i The encoding feature vector C i and the array group y at time k j The encoding feature vector C j The difference between .

[0074] Preferably, the step S9 comprises:

[0075] Step S9.1: Calculate the distance matrix between the characteristic elements of the array formation at time k-1 and time k;

[0076] Step S9.2: extracting elements from the distance matrix to form a vector;

[0077] Step S9.3: Calculate and construct a basic probability distribution function based on the formation feature vector;

[0078] The step S9.2 comprises:

[0079] Step S9.2.1: Select the smaller value of the two array formation feature dimensions as the number of iterations;

[0080] Step S9.2.2: Obtain the maximum and minimum values ​​in the distance matrix, and the corresponding positions;

[0081] Step S9.2.3: Add the minimum value in the distance matrix to the distance feature vector;

[0082] Step S9.2.4: Delete the row and column corresponding to the minimum value of the distance matrix and update the distance matrix;

[0083] Step S9.2.5: Determine whether the number of iterations has been reached. If not, return to step 9.2.2 for iterative calculation; otherwise, output the formation feature vector.

[0084] Preferably, the step S10 includes:

[0085] Step S10.1: Calculate the error matrix of the array group target at time k-1 and time k;

[0086] Step S10.2: Construct a basic probability distribution function based on the error value, the formula is:

[0087]

[0088] in, is the array x at time k-1 i and the array group y at time k j The topological structure error value of

[0089] The step S10.1 comprises:

[0090] Step S10.1.1: Setting the initialization update point set;

[0091] Step S10.1.2: Use Euclidean distance to find the closest point pair of the corresponding point set;

[0092] Step S10.1.3: Calculate the centroid of the source point cloud and the centroid of the target point cloud;

[0093] Step S10.1.4: Calculate the covariance matrix between the centroids;

[0094] Step S10.1.5: performing singular value decomposition on the covariance matrix;

[0095] Step S10.1.6: Calculate the rotation matrix and centroid translation vector between the two point clouds, and update the transformation matrix;

[0096] Step S10.1.7: Calculate the updated point set according to the updated transformation matrix and calculate the error value.

[0097] According to the array group target data association system based on adaptive clustering provided by the present invention, the array group target data association method based on adaptive clustering is adopted, and includes the following modules:

[0098] Module M1: Divide the area, obtain the grid division boundary threshold, and accumulate the target position information obtained by the same load within the area;

[0099] Module M2: cluster the data of each load in the order of entering the field of view to obtain the initial array group division;

[0100] Module M3: Screen a single clustering target set and remove target points whose similarity does not meet the preset requirements;

[0101] Module M4: remove non-array targets according to the array determination conditions;

[0102] Module M5: Describe the characteristics of the target formation at two moments respectively;

[0103] Module M6: Describe the characteristics of the target formation at two moments respectively;

[0104] Module M7: Modeling the feature vectors of the formation characteristics and formation characteristics of the array target at two moments before and after respectively;

[0105] Module M8: Calculate the corresponding basic probability distribution function according to the array target association matching model with formation characteristics;

[0106] Module M9: Calculate the corresponding basic probability distribution function according to the array target association matching model of the formation characteristics;

[0107] Module M10: Calculate the corresponding basic probability distribution function based on the array target association matching model of the external topological structure characteristics;

[0108] Module M11: Through DS evidence theory, evidence synthesis is performed on the basic probability distribution function based on the association of formation, formation and external topological structure characteristics, and a comprehensive basic probability distribution function is established;

[0109] Module M12: Make association decisions based on the comprehensive basic probability allocation function based on two-dimensional allocation to obtain the array target matching results.

[0110] Compared with the prior art, the present invention has the following beneficial effects:

[0111] The present invention divides the sparse target set into array groups, and makes full use of the relatively stable characteristics of the composition, formation and external topological structure characteristics of the array group targets, realizes the establishment of a multi-feature mathematical model through evidence synthesis, and constructs a comprehensive probability distribution function to perform correlation matching of target points between each array group with a unified standard, which can effectively reduce the influence of the change of the motion state of the array group targets, the missed detection of the sensor and the observation error, and significantly improve the correct association rate of the target. BRIEF DESCRIPTION OF THE DRAWINGS

[0112] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0113] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0114] Figure 2 It is a schematic diagram of the process of performing array group division based on the clustering algorithm in the present invention;

[0115] Figure 3 A schematic diagram of the process of screening the clustering target set and removing target points with low similarity in the present invention;

[0116] Figure 4 A schematic diagram of a flow chart for calculating a basic probability allocation function based on a composition feature in the present invention;

[0117] Figure 5 A schematic diagram of a flow chart for calculating a basic probability allocation function based on formation characteristics in the present invention;

[0118] Figure 6 A schematic diagram of a process for calculating a basic probability distribution function based on external topological structure characteristics in the present invention;

[0119] Figure 7a to Figure 7i A schematic diagram of the spatial distribution of the target for each load acquisition of a set of data verified in the present invention;

[0120] Figure 8 Schematic diagram of the spatial distribution of target trajectory points of the verification data in FIG7 of the present invention;

[0121] Figure 9a to Figure 9i It is a schematic diagram of the spatial distribution of clustering results of the target acquisition of each load of the verification data in FIG7 of the present invention;

[0122] Fig.10It is a schematic diagram of the trajectory association result of the verification data in FIG7 of the present invention. DETAILED DESCRIPTION

[0123] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0124] Example 1

[0125] The present invention provides a method for associating array target data based on adaptive clustering, such as Figure 1 As shown, the method specifically includes the following:

[0126] Step S1: Perform preprocessing operations to obtain the grid division boundary threshold and accumulate the same load data.

[0127] Among them, in step S1, a preprocessing operation is performed to obtain a grid division boundary threshold and accumulate the same load data, including:

[0128] Step S1.1: Divide the area and obtain the grid division boundary threshold;

[0129] Step S1.2: accumulating target position information obtained by the same load within the area;

[0130] Step S1.3: Process the load data in the order in which they enter the field of view.

[0131] Step S2: clustering the load data to obtain initial array group divisions.

[0132] In step S2, each load data is clustered to obtain the initial array group division, such as Figure 2 ,include:

[0133] Step S2.1: Calculate the distance between every two targets and construct a distance matrix;

[0134] Step S2.2: Find the threshold of density calculation according to the distance matrix, requiring that the number of points around each point whose distance is less than the density threshold should account for 1%-4% of the total number of points.

[0135] Wherein, the step S2.2 includes the following sub-steps:

[0136] Step S2.2.1: parameter setting, including the upper and lower limits of the number ratio threshold and the distance range threshold;

[0137] Step S2.2.2: Set the initial density threshold as the average of the maximum distance plus the minimum distance;

[0138] Step S2.2.3: Calculate the proportion of points with density less than the threshold;

[0139] Step S2.2.4: If the proportion is within the set range, the current density threshold is output; otherwise, the current maximum distance, minimum distance and density threshold are updated;

[0140] Step S2.2.5: If the current distance range is less than the set threshold, output the current density threshold; otherwise, return to step S2.3 for iterative calculation.

[0141] Step S2.3: Calculate the local density of each target point and construct a local density vector;

[0142] Step S2.4: Calculate the nearest neighbor distance of each target point and construct the nearest neighbor distance vector;

[0143] Wherein, the step S2.4 includes the following sub-steps:

[0144] Step S2.4.1: sort the local density vectors from large to small according to the density value;

[0145] Step S2.4.2: Calculate the distance between each point and other points with a greater density than the point;

[0146] Step S2.4.3: Get the index of the nearest neighbor point of each point and the distance between the two points.

[0147] Step S2.5: adaptively obtain cluster centers;

[0148] Wherein, the step S2.5 includes the following sub-steps:

[0149] Step S2.5.1: Calculate a joint density vector based on the local density vector and the nearest neighbor distance vector;

[0150] Step S2.5.2: Calculate the mean and variance of the joint density vector;

[0151] Step S2.5.3: Traverse each data point and take the points that exceed the set threshold range as cluster centers.

[0152] Step S2.6: Classify non-cluster center data points.

[0153] Wherein, the step S2.6 includes the following sub-steps:

[0154] Step S2.6.1: Initialize the cluster number of each data point, and the cluster number of non-cluster centers is uniformly set to -1;

[0155] Step S2.6.2: sort the local density vectors from large to small according to the density value;

[0156] Step S2.6.3: Update the cluster number of the non-cluster center data point to be the same as the cluster number of the point that is closest to it and has a greater density than it.

[0157] Step S3: Screen a single cluster target set and remove target points with low similarity.

[0158] like Figure 3 , the step S3 comprises the following sub-steps:

[0159] Step S3.1: Obtain corresponding coordinates according to the obtained cluster center index;

[0160] Step S3.2: According to the cluster center and each target cluster number, obtain the position coordinates of each data point in the same cluster;

[0161] Step S3.3: Calculate the local density of each point in a single cluster;

[0162] Step S3.4: Calculate the similarity between each data point in the same cluster and the cluster center;

[0163] Step S3.5: Calculate the comprehensive similarity of each data point in the same cluster based on the similarity and local density;

[0164] Step S3.6: Calculate the mean and variance of the comprehensive similarity vector of data points in a single cluster;

[0165] Step S3.7: Add the data points whose comprehensive similarity values ​​exceed the threshold range as new cluster centers and update their cluster numbers.

[0166] Step S4: removing non-array group targets according to the array group determination conditions.

[0167] Wherein, the step S4 includes the following sub-steps:

[0168] Step S4.1: Obtain the target belonging to the same cluster according to the cluster number corresponding to each data point;

[0169] Step S4.2: assembling data for a single cluster that meets the set array group construction threshold;

[0170] Step S4.3: Data points that do not meet the array group formation conditions are regarded as isolated points and are not subjected to subsequent array group matching.

[0171] Step S5: Describe the formation characteristics of the array target at two moments respectively.

[0172] Wherein, the step S5 includes the following sub-steps:

[0173] Step S5.1: Calculate the target type number of each target in each array group;

[0174] Step S5.2: Calculate the number of members of each target type contained in each array group, which is the target composition feature of the array group.

[0175] Step S6: Describing the characteristics of the target formation at two moments respectively.

[0176] The step S6 of describing the formation features of the target group at two moments respectively includes the following sub-steps:

[0177] Step S6.1: Calculate the distance between every two targets in the array;

[0178] Step S6.2: Construct the target formation feature vector of each array group.

[0179] Step S7: Modeling the feature vectors of the formation features and formation features of the array targets at two moments in time.

[0180] Step S8: Calculate the corresponding basic probability distribution function according to the array group target association matching model with formation characteristics.

[0181] like Figure 4 In step S8, the corresponding basic probability distribution function is calculated according to the array group target association matching model with the formation characteristics, including the following sub-steps:

[0182] Step S8.1: Obtain the total number of target types of the two array groups and assign weights according to their importance;

[0183] Step S8.2: Normalize the weight set according to the following formula:

[0184]

[0185] in, is the weight set according to the importance, N ij is the array x at time k-1 i and the array group y at time k j The total number of target types included.

[0186] Step S8.3: pre-processing the formation feature vectors of different array groups, filling 0 at the corresponding position where there is no target type in the formation feature, so that the length of the formation feature vector of the array group is consistent;

[0187] Step S8.4: Calculate the difference in the formation characteristic vector between the array groups;

[0188] Step S8.5: Calculate the weighted inter-array formation feature difference vector;

[0189] Step S8.6: Construct a basic probability distribution function based on the feature difference vector according to the following formula:

[0190]

[0191] in, is the array x at time k-1 i The encoding feature vector C i and the array group y at time k j The encoding feature vector C j The difference between .

[0192] Step S9: Calculate the corresponding basic probability distribution function according to the array target association matching model of the formation characteristics.

[0193] like Figure 5 , the step S9 includes the following sub-steps:

[0194] Step S9.1: Calculate the distance matrix between the characteristic elements of the array formation at time k-1 and time k;

[0195] Step S9.2: extracting elements from the distance matrix to form a vector;

[0196] Wherein, the step S9.2 includes the following sub-steps:

[0197] Step S9.2.1: Select the smaller value of the two array formation feature dimensions as the number of iterations;

[0198] Step S9.2.2: Obtain the maximum and minimum values ​​in the distance matrix, and the corresponding positions;

[0199] Step S9.2.3: Add the minimum value in the distance matrix to the distance feature vector;

[0200] Step S9.2.4: Delete the row and column corresponding to the minimum value of the distance matrix and update the distance matrix;

[0201] Step S9.2.5: Determine whether the number of iterations has been reached. If not, return to step 9.2.2 for iterative calculation; otherwise, output the formation feature vector.

[0202] Step S9.3: Calculate and construct a basic probability distribution function based on the formation feature vector. The calculation formula is as follows:

[0203]

[0204] in, L ijis the array x at time k-1 i and the array group y at time k j The distance feature vector, N min is the number of iterations.

[0205] Step S10: Calculate the corresponding basic probability distribution function according to the array group target association matching model of the external topological structure characteristics.

[0206] like Figure 6 , the step S10 includes the following sub-steps:

[0207] Step S10.1: Calculate the error matrix of the array group target at time k-1 and time k;

[0208] Wherein, the step S10.1 includes the following sub-steps:

[0209] Step S10.1.1: Setting the initialization update point set;

[0210] Step S10.1.2: Use Euclidean distance to find the closest point pair of the corresponding point set;

[0211] Step S10.1.3: Calculate the centroid of the source point cloud and the centroid of the target point cloud;

[0212] Step S10.1.4: Calculate the covariance matrix between the centroids;

[0213] Step S10.1.5: performing singular value decomposition on the covariance matrix;

[0214] Step S10.1.6: Calculate the rotation matrix and centroid translation vector between the two point clouds, and update the transformation matrix;

[0215] Step S10.1.7: Calculate the updated point set according to the updated transformation matrix and calculate the error value.

[0216] Step S10.2: Construct a basic probability distribution function based on the error value, and the calculation formula is as follows:

[0217]

[0218] in, is the array x at time k-1 i and the array group y at time k j The topological structure error value.

[0219] Step S11: Using the DS evidence theory, evidence synthesis is performed on the basic probability distribution function based on the association of formation, formation and external topological structure features to establish a comprehensive basic probability distribution function;

[0220] Step S12: Based on the two-dimensional allocation, an association decision is made on the comprehensive basic probability allocation function to obtain the array group target matching result according to the following formula:

[0221]

[0222] Satisfy the constraints:

[0223] m ij >m_ref

[0224]

[0225] The optimal matching solution is

[0226] Among them, m ij Represents the matching metric matrix The (i, j)th element of ij Represents the allocation matrix The (i, j)th element of is a binary variable; N k-1 N represents the number of array targets at time (k-1); k represents the number of array targets at time (k); m_ref is the set confidence threshold; (z 1i , z 2j ) indicates a successfully matched target pair.

[0227] Step S13: Wait for the next moment to process the array target information.

[0228] The effects of the present invention can be further illustrated by the following simulation results.

[0229] 1. Simulation data:

[0230] The multi-payload observation of STK simulation was used to obtain the position information sequence of the maritime targets within the specified area, and the data processing was completed through PyCharm 2017 platform simulation.

[0231] 2. Simulation content and results:

[0232] In order to verify the effectiveness of the array target data association method based on adaptive clustering provided by the present invention, experimental tests are conducted on multiple sets of target data sets with unknown array information. The following is a set of experimental data and processing results of the position information sequence of 30 marine moving targets observed by 9 types of payloads within 23 minutes for illustration.

[0233] Figure 7 is a schematic diagram of the spatial distribution of 30 marine moving targets observed by 9 types of payloads within 23 minutes of a simulation;

[0234] Figure 8 Schematic diagram of the spatial distribution of target trajectory points in the scenario described in FIG7 ;

[0235] FIG9 is a schematic diagram of the spatial distribution of clustering results of each load acquisition target in the scenario described in FIG7 ;

[0236] Fig.10 Schematic diagram of trajectory association results in the scenario described in FIG7 .

[0237] Example 2

[0238] The present invention also provides an array group target data association system based on adaptive clustering, wherein the array group target data association based on adaptive clustering can be implemented by executing the process steps of the array group target data association method based on adaptive clustering, that is, those skilled in the art can understand the array group target data association method based on adaptive clustering as a preferred implementation of the array group target data association system based on adaptive clustering. The system specifically includes the following contents:

[0239] Module M1: Perform preprocessing operations to obtain the grid division boundary threshold and accumulate the same load data.

[0240] The module M1 includes:

[0241] Module M1.1: Divide the area and obtain the grid division boundary threshold;

[0242] Module M1.2: Accumulate the target position information obtained by the same payload within the area;

[0243] Module M1.3: Process the data of each load in the order in which they enter the field of view.

[0244] Module M2: Cluster the load data to obtain the initial array group division.

[0245] The module M2 includes:

[0246] Module M2.1: Calculate the distance between every two targets and construct a distance matrix;

[0247] Module M2.2: Find the threshold for density calculation based on the distance matrix, requiring that the number of points around each point whose distance is less than the density threshold should account for 1%-4% of the total number of points);

[0248] Among them, module M2.2 includes:

[0249] Module M2.2.1: Parameter settings, including the upper and lower limits of the number percentage threshold and the distance range threshold;

[0250] Module M2.2.2: Set the initial density threshold as the average of the maximum distance plus the minimum distance;

[0251] Module M2.2.3: Calculate the percentage of points with density less than the threshold;

[0252] Module M2.2.4: If the proportion is within the set range, the current density threshold is output; otherwise, the current maximum distance, minimum distance and density threshold are updated;

[0253] Module M2.2.5: If the current distance range is less than the set threshold, output the current density threshold; otherwise, return to module M2.3 for iterative calculation.

[0254] Module M2.3: Calculate the local density of each target point and construct the local density vector;

[0255] Module M2.4: Calculate the nearest neighbor distance of each target point and construct the nearest neighbor distance vector;

[0256] Among them, module M2.4 includes:

[0257] Module M2.4.1: Sort the local density vectors by density value from large to small;

[0258] Module M2.4.2: Calculate the distances between each point and all other points that are denser than it;

[0259] Module M2.4.3: Get the index of the nearest neighbor of each point and the distance between the two points.

[0260] Module M2.5: Adaptive acquisition of cluster centers;

[0261] Among them, module M2.5 includes:

[0262] Module M2.5.1: Calculate the joint density vector based on the local density vector and the nearest neighbor distance vector;

[0263] Module M2.5.2: Calculate the mean and variance of the joint density vector;

[0264] Module M2.5.3: Traverse each data point and take the points that exceed the set threshold range as cluster centers.

[0265] Module M2.6: Classify non-cluster center data points.

[0266] Among them, module M2.6 includes:

[0267] Module M2.6.1: Initialize the cluster number of each data point, and the cluster number of non-cluster centers is uniformly set to -1;

[0268] Module M2.6.2: Sort the local density vectors by density value from large to small;

[0269] Module M2.6.3: Update the cluster number of non-cluster center data points to the same cluster number as the point that is closest to it and has a greater density than it.

[0270] Module M3: Filter a single clustering target set and remove target points with low similarity.

[0271] Among them, module M3 includes:

[0272] Module M3.1: Obtain the corresponding coordinates according to the obtained cluster center index;

[0273] Module M3.2: Obtain the position coordinates of each data point in the same cluster according to the cluster center and each target cluster number;

[0274] Module M3.3: Calculate the local density of each point in a single cluster;

[0275] Module M3.4: Calculate the similarity between each data point in the same cluster and the cluster center;

[0276] Module M3.5: Calculate the comprehensive similarity of each data point in the same cluster based on similarity and local density;

[0277] Module M3.6: Calculate the mean and variance of the comprehensive similarity vector of data points in a single cluster;

[0278] Module M3.7: Add data points whose comprehensive similarity values ​​exceed the threshold range as new cluster centers and update their cluster numbers.

[0279] Module M4: Remove non-array targets based on the array determination conditions.

[0280] Among them, module M4 includes:

[0281] Module M4.1: Obtain the targets belonging to the same cluster according to the cluster number corresponding to each data point;

[0282] Module M4.2: Data assembly for individual clusters that meet the set array construction threshold;

[0283] Module M4.3: Data points that do not meet the array group formation conditions are regarded as isolated points and no subsequent array group matching is performed.

[0284] Module M5: Describe the characteristics of the target group composition at two moments in time.

[0285] Among them, module M5 includes:

[0286] Module M5.1: Calculate the target type number of each target in each array group;

[0287] Module M5.2: Calculate the number of members of each target type contained in each array group, which is the target composition feature of the array group.

[0288] Module M6: Describe the characteristics of the target formation at two moments in time.

[0289] Among them, module M6 describes the characteristics of the target formation at two moments respectively, including:

[0290] Module M6.1: Calculate the distance between every two targets in the array;

[0291] Module M6.2: Construct the target formation feature vector for each array group.

[0292] Module M7: Model the feature vectors of the target group formation characteristics and formation characteristics at two moments before and after respectively.

[0293] Module M8: Calculate the corresponding basic probability distribution function based on the array target association matching model with organizational characteristics.

[0294] Among them, the module M8 calculates the corresponding basic probability distribution function according to the array group target association matching model of the formation characteristics, including:

[0295] Module M8.1: Obtain the total number of target types in the two array groups and assign weights according to their importance;

[0296] Module M8.2: Normalize the weight set according to the following formula:

[0297]

[0298] in, is the weight set according to the importance, N ij is the array x at time k-1 i and the array group y at time k j The total number of target types included.

[0299] Module M8.3: Preprocess the formation feature vectors of different arrays, fill 0 in the corresponding positions of the target type that does not exist in the formation feature, so that the length of the formation feature vectors of the arrays is consistent;

[0300] Module M8.4: Calculate the difference in formation characteristic vectors between array groups;

[0301] Module M8.5: Calculate the weight-based inter-array formation feature difference vector;

[0302] Module M8.6: Construct a basic probability distribution function based on the feature difference vector according to the following formula:

[0303]

[0304] in, is the array x at time k-1 i The encoding feature vector C i and the array group y at time k j The encoding feature vector C j The difference between .

[0305] Module M9: Calculate the corresponding basic probability distribution function based on the array target association matching model of the formation characteristics.

[0306] Among them, module M9 includes:

[0307] Module M9.1: Calculate the distance matrix between the characteristic elements of the formation at time k-1 and time k;

[0308] Module M9.2: Extract elements from the distance matrix to form a vector;

[0309] Among them, module M9.2 includes:

[0310] Module M9.2.1: Select the smaller value of the two array formation characteristic dimensions as the number of iterations;

[0311] Module M9.2.2: Get the maximum and minimum values ​​in the distance matrix and the corresponding positions;

[0312] Module M9.2.3: Add the minimum value in the distance matrix to the distance feature vector;

[0313] Module M9.2.4: Delete the row and column corresponding to the minimum value of the distance matrix and update the distance matrix;

[0314] Module M9.2.5: Determine whether the number of iterations has been reached. If not, return to module M9.2.2 for iteration counting.

[0315] Calculate, otherwise output the formation feature vector.

[0316] Module M9.3: Calculate the basic probability distribution function based on the formation feature vector. The calculation formula is as follows:

[0317]

[0318] in, L ij is the array x at time k-1 i and the array group y at time k j The distance feature vector, N min is the number of iterations.

[0319] Module M10: Calculate the corresponding basic probability distribution function based on the array target association matching model based on the external topological structure characteristics.

[0320] Among them, module M10 includes:

[0321] Module M10.1: Calculate the error matrix of the array target at time k-1 and time k;

[0322] Among them, module M10.1 includes:

[0323] Module M10.1.1: Set the initialization update point set;

[0324] Module M10.1.2: Use Euclidean distance to find the closest pair of corresponding points.

[0325] Module M10.1.3: Calculate the centroid of the source point cloud and the centroid of the target point cloud;

[0326] Module M10.1.4: Calculate the covariance matrix between centroids;

[0327] Module M10.1.5: Perform singular value decomposition on the covariance matrix;

[0328] Module M10.1.6: Calculate the rotation matrix and centroid translation vector between two point clouds and update the transformation matrix;

[0329] Module M10.1.7: Calculate the updated point set based on the updated transformation matrix and calculate the error value.

[0330] Module M10.2: Construct a basic probability distribution function based on the error value. The calculation formula is as follows:

[0331]

[0332] in, is the array x at time k-1 i and the array group y at time k j The topological structure error value.

[0333] Module M11: Through DS evidence theory, evidence synthesis is performed on the basic probability distribution function based on the association of formation, formation and external topological structure characteristics, and a comprehensive basic probability distribution function is established;

[0334] Module M12: Based on the two-dimensional allocation, the integrated basic probability allocation function is associated with the decision to obtain the array target matching result according to the following formula:

[0335]

[0336] Satisfy the constraints:

[0337] mij >m_ref

[0338]

[0339]

[0340] The optimal matching solution is

[0341] Among them, m ij Represents the matching metric matrix The (i, j)th element of ij Represents the allocation matrix The (i, j)th element of is a binary variable; N k-1 N represents the number of array targets at time (k-1); k represents the number of array targets at time (k); m_ref is the set confidence threshold; (z 1i , z 2j ) indicates a successfully matched target pair.

[0342] Module M13: Wait for the next moment to process the array target information.

[0343] Those skilled in the art know that, in addition to implementing the system, device and its various modules provided by the present invention in a purely computer-readable program code, it is entirely possible to implement the same program in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers and embedded microcontrollers by logically programming the method steps. Therefore, the system, device and its various modules provided by the present invention can be considered as a hardware component, and the modules included therein for implementing various programs can also be considered as structures within the hardware component; the modules for implementing various functions can also be considered as both software programs for implementing the method and structures within the hardware component.

[0344] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A method for associating array target data based on adaptive clustering, characterized in that: include: Step S1: Divide the area, obtain the grid division boundary threshold, and accumulate the target position information obtained by the same load within the area; Step S2: clustering the data of each load in the order of entering the field of view to obtain the initial array group division; Step S3: Screening a single clustering target set and removing target points whose similarity does not meet the preset requirements; Step S4: removing non-array group targets according to the array group determination conditions; Step S5: describing the formation characteristics of the array targets at two moments respectively; Step S6: describing the formation characteristics of the target group at two moments respectively; Step S7: Modeling the feature vectors of the formation features and formation features of the array targets at two moments before and after respectively; Step S8: Calculate the corresponding basic probability distribution function according to the array group target association matching model with the formation characteristics; Step S9: Calculate the corresponding basic probability distribution function according to the array target association matching model of the formation characteristics; Step S10: Calculate the corresponding basic probability distribution function according to the array group target association matching model of the external topological structure characteristics; Step S11: Using the DS evidence theory, evidence synthesis is performed on the basic probability distribution function based on the association of formation, formation and external topological structure features to establish a comprehensive basic probability distribution function; Step S12: Perform an association decision on the comprehensive basic probability allocation function based on the two-dimensional allocation to obtain the array group target matching result.

2. The array target data association method based on adaptive clustering according to claim 1, characterized in that: The step S2 comprises: Step S2.1: Calculate the distance between every two targets and construct a distance matrix; Step S2.2: Find the threshold of density calculation according to the distance matrix, where the number of points around each point whose distance is less than the density threshold accounts for 1%-4% of the total number of points on average; Step S2.3: Calculate the local density of each target point and construct a local density vector; Step S2.4: Calculate the nearest neighbor distance of each target point and construct the nearest neighbor distance vector; Step S2.5: adaptively obtain cluster centers; Step S2.6: classify the non-cluster center data points; The step S2.2 comprises: Step S2.2.1: parameter setting, including the upper and lower limits of the number ratio threshold, and the distance range threshold; Step S2.2.2: Set the initial density threshold to the average of the maximum distance plus the minimum distance; Step S2.2.3: Calculate the proportion of points with density less than the threshold; Step S2.2.4: If the proportion is within the set range, output the current density threshold; otherwise, update the current maximum distance, minimum distance and density threshold; Step S2.2.5: If the current distance range is less than the set threshold, output the current density threshold; otherwise return to step S2.3 for iterative calculation; The step S2.4 comprises: Step S2.4.1: sort the local density vectors from large to small according to the density value; Step S2.4.2: Calculate the distance between each point and other points with a greater density than the point; Step S2.4.3: Get the index of the nearest neighbor point of each point and the distance between the two points; The step S2.5 comprises: Step S2.5.1: Calculate a joint density vector based on the local density vector and the nearest neighbor distance vector; Step S2.5.2: Calculate the mean and variance of the joint density vector; Step S2.5.3: Traverse each data point and take the points that exceed the set threshold range as cluster centers; The step S2.6 comprises: Step S2.6.1: Initialize the cluster number of each data point, and the cluster number of non-cluster centers is uniformly set to -1; Step S2.6.2: sort the local density vectors from large to small according to the density value; Step S2.6.3: Update the cluster number of the non-cluster center data point to be the same as the cluster number of the point that is closest to it and has a greater density than it.

3. The array target data association method based on adaptive clustering according to claim 1, characterized in that: The step S3 comprises: Step S3.1: Obtain corresponding coordinates according to the cluster center index; Step S3.2: According to the cluster center and each target cluster number, obtain the position coordinates of each data point in the same cluster; Step S3.3: Calculate the local density of each point in a single cluster; Step S3.4: Calculate the similarity between each data point in the same cluster and the cluster center; Step S3.5: Calculate the comprehensive similarity of each data point in the same cluster based on the similarity and local density; Step S3.6: Calculate the mean and variance of the comprehensive similarity vector of data points in a single cluster; Step S3.7: Add the data points whose comprehensive similarity values ​​exceed the threshold range as new cluster centers and update their cluster numbers.

4. The array target data association method based on adaptive clustering according to claim 1, characterized in that: The step S4 comprises: Step S4.1: Obtain the target belonging to the same cluster according to the cluster number corresponding to each data point; Step S4.2: assembling data for a single cluster that meets the set array group construction threshold; Step S4.3: Data points that do not meet the array group formation conditions are regarded as isolated points and no subsequent array group matching is performed.

5. The array target data association method based on adaptive clustering according to claim 1, characterized in that: The step S5 comprises: Step S5.1: Calculate the target type number of each target in each array group; Step S5.2: Calculate the number of members of each target type contained in each array group.

6. The array target data association method based on adaptive clustering according to claim 1, characterized in that: The step S6 comprises: Step S6.1: Calculate the distance between every two targets in the array; Step S6.2: Construct the target formation feature vector of each array group.

7. The array target data association method based on adaptive clustering according to claim 1, characterized in that: The step S8 comprises: Step S8.1: Obtain the total number of target types of the two array groups and assign weights according to their importance; Step S8.2: Normalize the weight set, the formula is: in, is the weight set according to the importance, N ij is the array x at time k-1 i and the array group y at time k j The total number of target types included; Step S8.3: pre-processing the formation characteristic vectors of different array groups to make the lengths of the formation characteristic vectors of the array groups consistent; Step S8.4: Calculate the difference in the formation characteristic vector between the array groups; Step S8.5: Calculate the weighted inter-array formation feature difference vector; Step S8.6: Construct a basic probability distribution function based on the feature difference vector, the formula is: in, is the array x at time k-1 i The encoding feature vector C i and the array group y at time k j The encoding feature vector C j The difference between .

8. The array target data association method based on adaptive clustering according to claim 1, characterized in that: The step S9 comprises: Step S9.1: Calculate the distance matrix between the characteristic elements of the array formation at time k-1 and time k; Step S9.2: extracting elements from the distance matrix to form a vector; Step S9.3: Calculate and construct a basic probability distribution function based on the formation feature vector; The step S9.2 comprises: Step S9.2.1: Select the smaller value of the two array formation feature dimensions as the number of iterations; Step S9.2.2: Obtain the maximum and minimum values ​​in the distance matrix, and the corresponding positions; Step S9.2.3: Add the minimum value in the distance matrix to the distance feature vector; Step S9.2.4: Delete the row and column corresponding to the minimum value of the distance matrix and update the distance matrix; Step S9.2.5: Determine whether the number of iterations has been reached. If not, return to step 9.2.2 for iterative calculation; otherwise, output the formation feature vector.

9. The array target data association method based on adaptive clustering according to claim 1, characterized in that: The step S10 comprises: Step S10.1: Calculate the error matrix of the array group target at time k-1 and time k; Step S10.2: Construct a basic probability distribution function based on the error value, the formula is: in, is the array x at time k-1 i and the array group y at time k j The topological structure error value of The step S10.1 comprises: Step S10.1.1: Setting the initialization update point set; Step S10.1.2: Use Euclidean distance to find the closest point pair of the corresponding point set; Step S10.1.3: Calculate the centroid of the source point cloud and the centroid of the target point cloud; Step S10.1.4: Calculate the covariance matrix between the centroids; Step S10.1.5: performing singular value decomposition on the covariance matrix; Step S10.1.6: Calculate the rotation matrix and centroid translation vector between the two point clouds, and update the transformation matrix; Step S10.1.7: Calculate the updated point set according to the updated transformation matrix and calculate the error value.

10. An array target data association system based on adaptive clustering, characterized in that: The array group target data association method based on adaptive clustering according to any one of claims 1 to 9 comprises the following modules: Module M1: Divide the area, obtain the grid division boundary threshold, and accumulate the target position information obtained by the same load within the area; Module M2: cluster the data of each load in the order of entering the field of view to obtain the initial array group division; Module M3: Screen a single clustering target set and remove target points whose similarity does not meet the preset requirements; Module M4: remove non-array targets according to the array determination conditions; Module M5: Describe the characteristics of the target formation at two moments respectively; Module M6: Describe the characteristics of the target formation at two moments respectively; Module M7: Modeling the feature vectors of the formation characteristics and formation characteristics of the array target at two moments before and after respectively; Module M8: Calculate the corresponding basic probability distribution function according to the array target association matching model with formation characteristics; Module M9: Calculate the corresponding basic probability distribution function according to the array target association matching model of the formation characteristics; Module M10: Calculate the corresponding basic probability distribution function based on the array target association matching model of the external topological structure characteristics; Module M11: Through DS evidence theory, evidence synthesis is performed on the basic probability distribution function based on the association of formation, formation and external topological structure characteristics, and a comprehensive basic probability distribution function is established; Module M12: Make association decisions based on the comprehensive basic probability allocation function based on two-dimensional allocation to obtain the array target matching results.

Images