Globally optimal association method and system based on target topological feature relationship graph

Abstract

The present application belongs to the field of target association. Specifically disclosed are a globally optimal association method and system based on a target topological feature relationship graph. By means of the present application, a target topological feature of a set of associated targets and a target topological feature of a set of targets to be associated are extracted; and then, all combinations of topological features of associated targets and topological features of targets to be associated are traversed, and a combination which maximizes the degree of relationship between the two target topological features is determined, so as to obtain a target matching relationship. The overall degree of relationship in each combination is taken into full consideration, so as to solve the problem of it being difficult to distinguish targets due to multi-target confusion during target association in dense multi-target scenarios.

Description

A Global Optimal Association Method and System Based on Target Topological Feature Relationship Graph [Technical Field]

[0001] This application belongs to the field of target association, and more specifically, relates to a global optimal association method and system based on the target topological feature relationship graph. [Background Technology]

[0002] Multi-radar target association refers to the process of fusing detection data from different radars in a multi-radar system to determine whether these data belong to the same target and to establish a unified trajectory for the target. The core task of target association is to match the data of the target to be associated with existing associated target data to determine whether they belong to the same actual target. Associated targets refer to radar detection data that has been confirmed and successfully associated with a specific target in the multi-radar system. Targets to be associated refer to radar detection data that has not yet been confirmed or associated with a specific target in the multi-radar system; this data may be new target detections or unconfirmed detections of existing targets. Target association is the foundation of target tracking, improving its accuracy and stability.

[0003] However, when multiple targets are very close together, the radar's resolution may be insufficient to distinguish them, leading to target confusion and preventing clear identification of individual targets. Target confusion refers to the situation in a multi-radar system where the similar or close signal characteristics of multiple targets cause the radar to be unable to accurately distinguish these targets, thus incorrectly associating the signals of different targets with the same target. [Summary of the Invention]

[0004] To address the shortcomings of existing technologies, the purpose of this application is to provide a globally optimal association method and system based on target topological feature relationship graphs, aiming to solve the problem of multi-target confusion and difficulty in distinguishing target associations in dense multi-target scenarios.

[0005] The first aspect of this application relates to a globally optimal association method based on a target topological feature relationship graph, which is applied to multi-radar detection target association, including:

[0006] Obtain the set of associated targets and the set of targets to be associated, where each target includes its own position and velocity;

[0007] Construct target topology features for the associated target set and the target set to be associated respectively. The construction method is as follows: For a single target set, select one target as the center point and the other targets as nodes. Calculate the relative distance, relative orientation, relative speed and relative heading of each node relative to the center point, and use them as the edge between the node and the center point to obtain the target topology features corresponding to the center point. Traverse the entire target set until all targets are selected as the center point.

[0008] Traverse all combinations of topological features of associated targets and topological features of targets to be associated, determine the combination that maximizes the relationship between the topological features of the two targets, and obtain the target matching relationship. The relationship between the topological features of the two targets is the sum of the relationship between each node in the topological features of the target and each node in the topological features of the target to be associated. The relationship between the two nodes is the weighted sum of four factors: relative distance, relative bearing, relative speed and relative heading.

[0009] Perform a correlation test on the target matching relationship. If the test is successful, the target is determined to be correlated; otherwise, it is not correlated.

[0010] Preferably, for a target topological feature with center point i and a target topological feature to be associated with center point k, the formula for calculating the relationship degree between the two nodes is as follows:

[0011] Where j represents a node in the target topological feature, l represents a node in the target topological feature to be associated, and λ c Similarity factors The corresponding weight vector, based on factors The importance is determined, and Similarity factors Let τ represent the similarity of the c-th factor between node j of the target topological feature with center point i and node l of the target topological feature to be associated with center point k. ij =[Δdis ij ,Δore ij ,Δspeed ij ,Δcou ij ], representing the relative distance, relative orientation, relative speed, and relative heading of node j relative to center point i, respectively, τ ij (c) is τ ij The c-th element, μ kl =[Δdis kl ,Δore kl ,Δspeed kl ,Δcou kl ], representing the relative distance, relative orientation, relative velocity, and relative heading of node l relative to center point k, respectively, μ kl (c) is μ kl The c-th element; The nodes in the target topological feature with center point i and the nodes in the target topological feature to be associated with center point k are related to each other. The maximum value, The nodes in the target topological feature with center point i and the nodes in the target topological feature to be associated with center point k are related to each other. The minimum value.

[0012] Preferably, the step of traversing all combinations of topological features of associated targets and topological features of targets to be associated, determining the combination that maximizes the relational degree between the topological features of the two targets, and obtaining the target matching relationship is as follows:

[0013] Where H(j,l) = 1 or 0, H(j,l) = 1 indicates that the topological feature relationship of the target is more closely related, and H(j,l) = 0 indicates that the topological feature relationship of the target is more sparse.

[0014] Preferably, the optimal solution is obtained based on the KM algorithm to obtain the target matching relationship.

[0015] Preferably, the following threshold clustering method is used to determine the mutual support relationship:

[0016] Step 1: Given a clustering threshold ε, arbitrarily select relation degree 'a' as the seed value to form cluster e1. Search for the target relation degree 'b' outside cluster e1. If |ba|≤ε exists, then relation degree 'b' is marked as a non-seed value and enters cluster e1; otherwise, go to Step 2.

[0017] Step 2: Identify the relation degree b as the new seed value to form a new cluster e2, until e * There are no other target affinity values;

[0018] Step 3: If each cluster e * If there are still non-seed values ​​in the cluster, return to Step 1; otherwise, the clustering ends and proceed to Step 4.

[0019] Step 4: Find the largest cluster. If the number of items in the cluster exceeds 50% of the total number, the association is considered successful; otherwise, the association is unsuccessful.

[0020] Preferably, the clustering threshold ranges from 0.1 to 0.35.

[0021] The second aspect of this application relates to a globally optimal association system based on a target topological feature relationship graph, comprising: at least one memory for storing a program; and at least one processor for accessing the program stored in the memory, wherein when the program stored in the memory is accessed, the processor is used to access the globally optimal association method as described in the first aspect.

[0022] The third aspect of this application relates to a computer-readable storage medium storing a computer program that, when executed on a processor, causes the processor to enter the globally optimal association method as described in the first aspect.

[0023] It is understood that the beneficial effects of the second and third aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here.

[0024] Overall, the technical solutions conceived in this application have the following beneficial effects compared with the prior art:

[0025] This application proposes a global optimal association method based on the target topology feature relationship graph. By extracting the topology features of each target in the associated target set and the target set to be associated, and then traversing all combinations of the topology features of the associated targets and the topology features of the target sets to be associated, the method determines the combination that maximizes the relationship between the topology features of the two targets, thereby obtaining the target matching relationship. This method fully considers the overall relationship in each combination, thus solving the problem of multi-target confusion and difficulty in distinguishing targets in dense multi-target scenarios. [Attached Image Description]

[0026] Figure 1 is a flowchart of a global optimal association method based on a target topological feature relationship graph provided in this application.

Detailed Implementation Methods

[0027] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0028] In this application, the term "and / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three cases: A existing alone, A and B existing simultaneously, and B existing alone. In this application, the symbol " / " indicates that the related objects are in an "or" relationship, for example, A / B means A or B.

[0029] The terms "first" and "second," etc., used in the specification and claims of this application are used to distinguish different objects, not to describe a specific order of objects. For example, "first response message" and "second response message," etc., are used to distinguish different response messages, not to describe a specific order of response messages.

[0030] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0031] In the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more, for example, multiple processing units means two or more processing units, multiple elements means two or more elements, etc.

[0032] The embodiments of this application are described below with reference to the accompanying drawings.

[0033] As shown in Figure 1, this application discloses a globally optimal association method based on a target topology feature relationship graph. This method is applied to multi-radar detection target association, including:

[0034] Step 1: Obtain the set of associated targets and the set of targets to be associated. Each target includes its own position and velocity.

[0035] Step 2: Construct the target topology features of the associated target set and the target set to be associated respectively. The construction method is as follows: For a single target set, select one target as the center point and the other targets as nodes. Calculate the relative distance, relative orientation, relative speed and relative heading of each node relative to the center point, and use them as the edge between the node and the center point to obtain the target topology features corresponding to the center point. Traverse the entire target set until all targets are selected as the center point.

[0036] Relative distance, relative bearing, relative speed, and relative heading can be mathematically calculated based on the positions and speeds of the two nodes. The details of this calculation are existing technology and will not be elaborated here.

[0037] Step 3: Traverse all combinations of topological features of associated targets and topological features of targets to be associated, determine the combination that maximizes the relationship between the topological features of the two targets, and obtain the target matching relationship. The relationship between the topological features of the two targets is the sum of the relationship between each node in the topological features of the target and the topological features of the target to be associated. The relationship between the two nodes is the weighted sum of four factors: relative distance, relative bearing, relative speed and relative heading.

[0038] Preferably, for a target topological feature with center point i and a target topological feature to be associated with center point k, the formula for calculating the relationship degree between the two nodes is as follows:

[0039] Where j represents a node in the target topological feature, l represents a node in the target topological feature to be associated, and λc Similarity factors The corresponding weight vector, based on factors The importance is determined, and Similarity factors Let τ represent the similarity of the c-th factor between node j of the target topological feature with center point i and node l of the target topological feature to be associated with center point k. ij =[Δdis ij ,Δore ij ,Δspeed ij ,Δcou ij ], representing the relative distance, relative orientation, relative speed, and relative heading of node j relative to center point i, respectively, τ ij (c) is τ ij The c-th element, μ kl =[Δdis kl ,Δore kl ,Δspeed kl ,Δcou kl ], representing the relative distance, relative orientation, relative velocity, and relative heading of node l relative to center point k, respectively, μ kl (c) is μ kl The c-th element; The nodes in the target topological feature with center point i and the nodes in the target topological feature to be associated with center point k are related to each other. The maximum value, The nodes in the target topological feature with center point i and the nodes in the target topological feature to be associated with center point k are related to each other. The minimum value.

[0040] Preferably, the step of traversing all combinations of topological features of associated targets and topological features of targets to be associated, determining the combination that maximizes the relational degree between the topological features of the two targets, and obtaining the target matching relationship is as follows:

[0041] Where H(j,l) = 1 or 0, H(j,l) = 1 indicates a closer relationship between the target topological features, and H(j,l) = 0 indicates a sparser relationship between the target topological features. The constraint is used to ensure that "if a target to be associated can be associated, it has one and only one associated target". n represents the number of topological feature nodes of the associated target, m represents the number of topological feature nodes of the target to be associated, and n ≤ m.

[0042] This application transforms the calculation of topological feature matching into an optimal linear programming problem with a single objective function, namely, finding a combination relationship in the relation degree matrix of two objective topological features such that the sum of the relation degree of the two objective topological features is maximized.

[0043] Preferably, the optimal solution is obtained based on the KM algorithm to obtain the target matching relationship.

[0044] The Kuhn-Munkres algorithm, or KM algorithm, finds a matching in a weighted bipartite graph that maximizes the sum of the weights of all edges in the matching. Specifically, it transforms the maximum weight matching problem into a perfect matching problem by assigning a vertex label (weight) to each vertex. The goal of the algorithm is to find an equal subgraph (a subgraph consisting of edges that satisfy specific conditions) and then find a perfect matching within that subgraph.

[0045] Step 4: Perform a correlation test on the target matching relationship. If the test is successful, the target is determined to be correlated; otherwise, it is not correlated.

[0046] Preferably, the following threshold clustering method is used to determine the mutual support relationship:

[0047] Step 1: Given a clustering threshold ε, arbitrarily select relation degree 'a' as the seed value to form cluster e1. Search for the target relation degree 'b' outside cluster e1. If |ba|≤ε exists, then relation degree 'b' is marked as a non-seed value and enters cluster e1; otherwise, go to Step 2.

[0048] Step 2: Identify the relation degree b as the new seed value to form a new cluster e2, until e * There are no other target affinity values;

[0049] Step 3: If each cluster e * If there are still non-seed values ​​in the cluster, return to Step 1; otherwise, the clustering ends and proceed to Step 4.

[0050] Step 4: Find the largest cluster. If the number of items in the cluster exceeds 50% of the total number, the association is considered successful; otherwise, the association is unsuccessful.

[0051] Preferably, the clustering threshold ranges from 0.1 to 0.35.

[0052] Example

[0053] Step 1: Obtain the set of associated targets and the set of targets to be associated. Each target includes its own position and velocity.

[0054] This embodiment is used in a multi-radar system at sea. The associated target set is {A,B,C,D}, and the target set to be associated is {a,b,c,e,f,g}. Each target includes its own position and velocity.

[0055] Step 2: Construct the target topology features of the associated target set and the target set to be associated respectively. The construction method is as follows: For a single target set, select one target as the center point and the other targets as nodes. Calculate the relative distance, relative orientation, relative speed and relative heading of each node relative to the center point, and use them as the edge between the node and the center point to obtain the target topology features corresponding to the center point. Traverse the entire target set until all targets are selected as the center point.

[0056] For a set of associated targets, containing four targets, we first use A as the center point. We calculate the relative distance, relative orientation, relative speed, and relative heading of B with respect to A and C with respect to A, and use these as edges between B and A. Similarly, we construct edges between C and A, and between D and A, resulting in a target topology feature map with A as the root node and three edges. We then obtain the target topology feature maps for B, C, and D in the same way.

[0057] For the associated target set, obtain six target topology feature maps corresponding to each node in the same way.

[0058] Step 3: Traverse all combinations of topological features of associated targets and topological features of targets to be associated, determine the combination that maximizes the relationship between the topological features of the two targets, and obtain the target matching relationship. The relationship between the topological features of the two targets is the sum of the relationship between each node in the topological features of the target and the topological features of the target to be associated. The relationship between the two nodes is the weighted sum of four factors: relative distance, relative bearing, relative speed and relative heading.

[0059] The number of combinations of all target topological features and target topological features to be associated is: Taking the target topology feature map corresponding to A and the target topology feature map corresponding to a as examples, the degree of relation between B and b, c, e, f, g, C and b, c, e, f, g, and D and b, c, e, f, g are calculated under different association modes, and the sum is used as the degree of relation between the two target topology features of the target topology feature combination under the association mode.

[0060] Find the combination with the largest topological feature relation degree between two targets from all combinations. Then, obtain the target matching relationship based on H(j,l) corresponding to this combination. Specifically, consider all two targets with H(j,l) = 1 as having a matching relationship.

[0061] Step 4: Perform a correlation test on the target matching relationship. If the test is successful, the target is determined to be correlated; otherwise, it is not correlated.

[0062] Successful association is defined as meeting the threshold clustering requirement and having more than 50% of the clusters being clustered. In this embodiment, we obtain a associating with A, b associating with B, and c associating with C, resulting in a new association target set = {a, b, c, e, f, g, D}.

[0063] It should be understood that the above-described device is used to execute the methods in the above embodiments. The implementation principle and technical effect of the corresponding program modules in the device are similar to those described in the above methods. The working process of the device can be referred to the corresponding process in the above methods, and will not be repeated here.

[0064] Based on the methods in the above embodiments, this application provides an electronic device that may include a processor, a communications interface, a memory, and a communication bus, wherein the processor, communications interface, and memory communicate with each other via the communication bus. The processor may invoke logical instructions stored in the memory to execute the methods in the above embodiments.

[0065] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0066] Based on the methods in the above embodiments, this application provides a computer-readable storage medium storing a computer program that, when run on a processor, causes the processor to execute the methods in the above embodiments.

[0067] Based on the methods in the above embodiments, this application provides a computer program product that, when run on a processor, causes the processor to execute the methods in the above embodiments.

[0068] It is understood that the processor in the embodiments of this application can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. A general-purpose processor can be a microprocessor or any conventional processor.

[0069] The method steps in this application embodiment can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, portable hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can reside in an ASIC.

[0070] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0071] It is understood that the various numerical designations used in the embodiments of this application are merely for the convenience of description and are not intended to limit the scope of the embodiments of this application.

[0072] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A globally optimal association method based on target topology feature relationship graph, characterized in that, This method is applied to the association of targets detected by multiple radars, including: Obtain the set of associated targets and the set of targets to be associated, where each target includes its own position and velocity; Construct target topology features for the associated target set and the target set to be associated respectively. The construction method is as follows: For a single target set, select one target as the center point and the other targets as nodes. Calculate the relative distance, relative orientation, relative speed and relative heading of each node relative to the center point, and use them as the edge between the node and the center point to obtain the target topology features corresponding to the center point. Traverse the entire target set until all targets are selected as the center point. Traverse all combinations of topological features of associated targets and topological features of targets to be associated, determine the combination that maximizes the relationship between the topological features of the two targets, and obtain the target matching relationship. The relationship between the topological features of the two targets is the sum of the relationship between each node in the topological features of the target and each node in the topological features of the target to be associated. The relationship between the two nodes is the weighted sum of four factors: relative distance, relative bearing, relative speed and relative heading. Perform a correlation test on the target matching relationship. If the test is successful, the target is determined to be correlated; otherwise, it is not correlated.

2. The globally optimal association method as described in claim 1, characterized in that, For the target topology feature with the center point i and the target topology feature to be associated with the center point k, the relationship degree of the two nodes is calculated according to the following formula: Where j represents a node in the target topological feature, l represents a node in the target topological feature to be associated, and λ c Similarity factors The corresponding weight vector, based on factors The importance is determined, and Similarity factors Let τ represent the similarity of the c-th factor between node j of the target topological feature with center point i and node l of the target topological feature to be associated with center point k. ij =[Δdis ij ,Δore ij ,Δspeed ij ,Δcou ij ], representing the relative distance, relative orientation, relative speed, and relative heading of node j relative to center point i, respectively, τ ij (c) is τ ij The c-th element, μ kl =[Δdis kl ,Δore kl ,Δspeed kl ,Δcou kl ], representing the relative distance, relative orientation, relative velocity, and relative heading of node l relative to center point k, respectively, μ kl (c) is μ ll The c-th element; The nodes in the target topological feature with center point i and the nodes in the target topological feature to be associated with center point k are related to each other. The maximum value, The nodes in the target topological feature with center point i and the nodes in the target topological feature to be associated with center point k are related to each other. The minimum value.

3. The globally optimal association method as described in claim 1, characterized in that, The process involves traversing all combinations of topological features of associated targets and topological features of targets to be associated, determining the combination that maximizes the relational degree between the topological features of the two targets, and obtaining the target matching relationship, as detailed below: Where H(j,l) = 1 or 0, H(j,l) = 1 indicates that the topological feature relationship of the target is more closely related, and H(j,l) = 0 indicates that the topological feature relationship of the target is more sparse.

4. The globally optimal association method as described in claim 3, characterized in that, The optimal solution is obtained based on the KM algorithm to obtain the target matching relationship.

5. The globally optimal association method as described in claim 1, characterized in that, The following threshold clustering method is used to determine the mutual support relationship: Step 1: Given a clustering threshold ε, arbitrarily select relation degree 'a' as the seed value to form cluster e1. Search for the target relation degree 'b' outside cluster e1. If |ba|≤ε exists, then relation degree 'b' is marked as a non-seed value and enters cluster e1; otherwise, go to Step 2. Step 2: Identify the relationship degree b as a new seed value, form a new cluster e2, until e * No other target affinity value outside Step 3: If there are still non-seed values in each cluster e * , return to Step 1; otherwise, the clustering ends and enters Step 4; Step 4: Find the largest cluster. If the number of items in the cluster exceeds 50% of the total number, the association is considered successful; otherwise, the association is unsuccessful.

6. The globally optimal association method as described in claim 5, characterized in that, The clustering threshold ranges from 0.1 to 0.

35.

7. A globally optimal association system based on a target topological feature relationship graph, characterized in that, include: At least one memory for storing programs; At least one processor is configured to access a program stored in the memory, wherein, when the program stored in the memory is accessed, the processor is configured to access the globally optimal association method as described in any one of claims 1 to 6.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed on a processor, causes the processor to enter the globally optimal association method as described in any one of claims 1 to 6.

Images