Track association method and device, computer equipment and readable storage medium

Abstract

The invention relates to a track association method and device, computer equipment, a readable storage medium and a computer program product. The method comprises the following steps: acquiring tracks corresponding to at least two sensors associated with a target object; determining the track membership degree of any two tracks at the current moment; under the condition that the precision of the first sensor is higher than that of the second sensor, determining the track correlation of any two tracks at the current moment according to a comparison result between the track membership of any two tracks at the current moment and the resolution membership of the second sensor; according to the track correlation, determining track correlation quality values of any two tracks at a historical moment; and under the condition that the track association quality value meets a preset association quality threshold value, carrying out track fusion processing on any two tracks to obtain a target association track of the target object. By adopting the method, the accuracy of a track association fusion result can be improved.

Description

Technical Field

[0001] This application relates to the field of track analysis technology, and in particular to a track association method, apparatus, computer equipment, readable storage medium, and computer program product. Background Technology

[0002] In a distributed sensor fusion system, due to factors such as weather, target, and equipment, sensors are used to track and detect the target, and sensor information from multiple tracks of the same target is organically fused. The fused track information is then output as an optimal and effective track through track association.

[0003] Current track association technologies primarily rely on statistics and fuzzy logic to calculate the correlation between different tracks. However, traditional fuzzy logic-based track association techniques are limited by the number of sensors and target scenes. In high-dimensional scenarios requiring expansion, the membership degree of effectively associated tracks is diluted. Furthermore, they often rely on single-moment observations for cross-sensor association, neglecting the continuous temporal process of tracks. This results in insufficient robustness to outliers (bad points) in the tracks, reducing the accuracy of track association results and thus lowering the success rate of track association.

[0004] Therefore, traditional technologies suffer from low accuracy in track correlation and fusion results in track analysis. Summary of the Invention

[0005] Therefore, it is necessary to provide a trajectory association method, apparatus, computer equipment, readable storage medium, and computer program product that can improve the accuracy of trajectory association fusion results in response to the above-mentioned technical problems.

[0006] In a first aspect, this application provides a track association method, the method comprising the following steps:

[0007] Obtain the tracks corresponding to at least two sensors associated with the target object; for any two tracks among the tracks corresponding to each sensor, determine the track membership degree of the any two tracks at the current time; the any two tracks include the track corresponding to the first sensor and the track corresponding to the second sensor;

[0008] If the accuracy of the first sensor is higher than that of the second sensor, then the correlation between the two tracks at the current moment is determined based on the comparison between the track membership degree of the two tracks at the current moment and the resolution membership degree of the second sensor.

[0009] Based on the track correlation of any two tracks at the current time, determine the track association quality value of any two tracks at at least one historical time before the current time.

[0010] If the trajectory association quality value of any two trajectories meets the preset association quality threshold, trajectory fusion processing is performed on the two trajectories to obtain the target associated trajectory of the target object.

[0011] In one embodiment, acquiring the tracks corresponding to at least two sensors associated with the target object includes:

[0012] Acquire real-time measurement data from each of the aforementioned sensors;

[0013] Feature extraction is performed on the real-time measurement data of each sensor to obtain the trajectory feature information of each trajectory of the target;

[0014] Based on the trajectory feature information of each sensor, the trajectory corresponding to each sensor is determined.

[0015] In one embodiment, the step of performing track fusion processing on any two tracks to obtain the target-related track of the target object includes:

[0016] Obtain the ratio of the maximum value of the track membership degree to the average value of the track membership degree in the track membership degree of the target object;

[0017] Select the target trajectory fusion strategy that matches the ratio from among multiple candidate trajectory fusion strategies;

[0018] According to the target track fusion strategy, track fusion processing is performed on any two tracks to obtain the target associated track of the target object.

[0019] In one embodiment, selecting the target trajectory fusion strategy that matches the ratio from among multiple candidate trajectory fusion strategies includes:

[0020] If the ratio is greater than a preset ratio threshold, a first track fusion strategy is selected from the plurality of candidate track fusion strategies as the target track fusion strategy; the first track fusion strategy is used to select the dominant track among any two tracks as the target associated track of the target object;

[0021] If the ratio is less than a preset ratio threshold, the second track fusion strategy is selected as the target track fusion strategy from the plurality of candidate track fusion strategies; the second track fusion strategy is used to perform weighted fusion on any two tracks, and the track obtained by weighted fusion is used as the target associated track of the target object.

[0022] In one embodiment, when the target track fusion strategy is the second track fusion strategy, the step of performing track fusion processing on any two tracks according to the target track fusion strategy to obtain the target associated track of the target object includes:

[0023] Obtain the fusion weights corresponding to each of the two possible paths; the fusion weights corresponding to the paths include the membership degree based on the resolution of the sensor corresponding to the path.

[0024] According to the fusion weights corresponding to each of the two possible paths, the paths in any two possible paths are weighted and summed to obtain the fused path, and the fused path is used as the target associated path of the target object.

[0025] In one embodiment, determining the track correlation of any two tracks at the current moment based on the comparison result between the track membership degree of any two tracks at the current moment and the resolution membership degree of the second sensor includes:

[0026] If the membership degree of any two tracks at the current time is greater than the resolution membership degree of the second sensor, the correlation between the tracks at the current time is determined to be that the two tracks are the same track.

[0027] If the membership degree of any two tracks at the current time is less than the resolution membership degree of the second sensor, the correlation between the tracks at the current time is determined to be that the two tracks are different tracks.

[0028] Secondly, this application provides a track association device, the device comprising:

[0029] The track membership determination module is used to acquire the tracks corresponding to at least two sensors associated with the target object; for any two tracks among the tracks corresponding to each sensor, the track membership degree of the arbitrary two tracks at the current time is determined; the arbitrary two tracks include the track corresponding to the first sensor and the track corresponding to the second sensor.

[0030] The track correlation determination module is used to determine the track correlation of any two tracks at the current moment based on the comparison result between the track membership degree of any two tracks at the current moment and the resolution membership degree of the second sensor when the accuracy of the first sensor is higher than that of the second sensor.

[0031] The track association quality determination module is used to determine the track association quality value of any two tracks at at least one historical moment before the current moment based on the track correlation of any two tracks at the current moment.

[0032] The track fusion processing module is used to perform track fusion processing on any two tracks when the track association quality value of any two tracks meets a preset association quality threshold, so as to obtain the target associated track of the target object.

[0033] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described track association method.

[0034] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0035] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method.

[0036] The aforementioned track association method, apparatus, computer equipment, readable storage medium, and computer program product acquire tracks corresponding to at least two sensors associated with a target object; for any two tracks among the tracks corresponding to each sensor, determine the track membership degree of any two tracks at the current time; any two tracks include the tracks corresponding to the first sensor and the tracks corresponding to the second sensor; if the accuracy of the first sensor is higher than the accuracy of the second sensor, determine the track correlation of any two tracks at the current time based on the comparison result between the track membership degree of any two tracks at the current time and the resolution membership degree of the second sensor; based on the track correlation of any two tracks at the current time, determine the track association quality value of any two tracks at at least one historical time before the current time; and perform track association on any two tracks. When the quality value meets the preset association quality threshold, track fusion processing is performed on any two tracks to obtain the target associated track of the target object. Based on the track membership degree of any two tracks of the target object, a first threshold decision on track correlation is determined by comparing the track membership degree with the sensor resolution membership degree, and a second threshold decision on track association quality is determined based on the track of the continuous process of historical time. This can achieve accurate association of any two tracks. Furthermore, fusion processing is performed on two tracks whose track association quality values ​​meet the preset association quality threshold, thereby improving the stability of track association decision, enhancing the robustness of track association, and improving the accuracy of track association fusion results. This effectively solves the problem of low accuracy of track association fusion results in track analysis by traditional techniques. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is an application environment diagram of a track association method in one embodiment;

[0039] Figure 2 This is a flowchart illustrating a track association method in one embodiment;

[0040] Figure 3 This is an overall architecture diagram of another track association method in one embodiment;

[0041] Figure 4 This is a structural block diagram of a track association device in one embodiment;

[0042] Figure 5 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0043] 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.

[0044] The track association method provided in this application can be applied to, for example... Figure 1 In the application environment shown, server 104 can acquire the tracks corresponding to at least two sensors associated with the target object; for any two tracks among the tracks corresponding to each sensor, it determines the track membership degree of any two tracks at the current moment; if the accuracy of the first sensor is higher than that of the second sensor, server 104 can determine the track correlation of any two tracks at the current moment based on the comparison result between the track membership degree of any two tracks at the current moment and the resolution membership degree of the second sensor; server 104 can determine the track association quality value of any two tracks at at least one historical moment before the current moment based on the track correlation of any two tracks at the current moment; if the track association quality value of any two tracks meets a preset association quality threshold, server 104 can perform track fusion processing on any two tracks to obtain the target associated track of the target object. The computer equipment can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0045] Before introducing the specific embodiments of this application, the technical terms involved in this application will be explained:

[0046] Track association refers to the process of identifying and matching track sequences belonging to the same real target in multi-source real-time measurement data.

[0047] Membership degree: It can quantitatively describe the strength of the association between an element and a specific set.

[0048] Track membership: In track association, it can be used to characterize the credibility of the current sensor's real-time measurement data belonging to the track of the real target object.

[0049] Track fusion refers to the process of combining multiple successfully correlated sensor tracks into a single globally optimal track. Essentially, it involves intelligently integrating real-time measurement data from multiple sources to generate target state estimates that exceed the accuracy limits of a single sensor.

[0050] In one exemplary embodiment, such as Figure 2 As shown, a track association method is provided. This embodiment applies this method to... Figure 1 The method is illustrated using server 104 as an example. In this embodiment, the method includes the following steps S202 to S208. Wherein:

[0051] Step S202: Obtain the tracks corresponding to at least two sensors associated with the target object; for any two tracks among the tracks corresponding to each sensor, determine the track membership degree of any two tracks at the current time.

[0052] The target object can refer to a moving target whose direction, position, speed, distance, etc. are constantly changing, including but not limited to aircraft, ships, vehicles, etc.

[0053] A track can be the trajectory formed by a target object such as an aircraft, ship, or vehicle during its movement. A complete track can be composed of various track points of the target object at multiple moments. It is understandable that any two tracks include the track corresponding to the first sensor. The trajectory corresponding to the second sensor .

[0054] In practical applications, server 104 can acquire the tracks corresponding to at least two sensors associated with the target object through multiple sensors of the target object; for any two tracks corresponding to each sensor... It is possible to determine any two flight paths At the present moment The following flight path distance Then based on the distance of the flight path Determine any two paths At the present moment The following track membership .

[0055] In step S204, if the accuracy of the first sensor is higher than that of the second sensor, the correlation between any two tracks at the current time is determined based on the comparison between the track membership degree of any two tracks at the current time and the resolution membership degree of the second sensor.

[0056] It should be noted that the higher the accuracy of the sensor, the closer the measurement data obtained by the sensor is to the true data, indicating that the error of the sensor measurement data is smaller, which in turn indicates that the sensor resolution value is smaller. Furthermore, when... At that time, track membership degree It can indicate the first The resolution membership degree of each sensor.

[0057] In this embodiment of the application, two threshold decisions are required for track association, namely the first threshold decision and the second threshold decision.

[0058] In specific implementation, based on sensor accuracy conditions, an association decision is made on the current track membership degree to determine track correlation, which serves as the first threshold decision in track association in this application embodiment. Taking any two sensors as an example, in the first sensor... Its accuracy is higher than that of the second sensor. Under the condition of accuracy, that is, the first sensor The resolution is smaller than that of the second sensor. In terms of resolution Server 104 can be based on any two flight paths At the present moment The following track membership Membership degree with the second sensor's resolution The comparison results between the two tracks can be used to determine the location of any two tracks. Current moment Track correlation .

[0059] Step S206: Based on the track correlation of any two tracks at the current time, determine the track association quality value of any two tracks at at least one historical time before the current time.

[0060] In the specific implementation, the track association quality is determined based on track correlation by combining historical tracks, which serves as the second threshold decision in the track association process of this application embodiment. Throughout the entire track association process, the track association quality detection is equivalent to having a sliding window of a certain length. Threshold calculations are then performed on the track points at each time point within the sliding window, and a judgment is made after the calculation. The sliding window is then moved to the next time point, and the calculation and judgment are performed again for the next time point.

[0061] For example, server 104 can base its actions on any two tracks. At the present moment Track correlation Determine any two paths At the present moment Track-related quality value at least one previous historical moment for:

[0062]

[0063] In equation (1), when Always present hour, ;when Always present hour, ; preset initial time If the flight path exist The first association at time, then .

[0064] Therefore, the track-related quality value It can be determined whether any two tracks are a fixed association pair. If any two tracks are determined to be a fixed association pair, it indicates that the second threshold decision detection in the track association of any two tracks is successful, and there is no need to enter the association detection stage again in subsequent association checks. In the second threshold decision, if any two tracks are determined to be a fixed association pair, then under certain conditions, either of the two tracks determined to be a fixed association pair can be used as the dominant track of the target object.

[0065] Step S208: If the trajectory association quality value of any two trajectories meets the preset association quality threshold, perform trajectory fusion processing on any two trajectories to obtain the target associated trajectory of the target object.

[0066] In practical applications, to reduce computational load, the trajectory correlation quality value between any two trajectories is used. For example, when the preset correlation quality threshold is met, the correlation quality value of the track can meet the threshold for eight consecutive time periods. At that time, the flight path can be specified. The tracks are fixed as a pair; therefore, the server 104 can, according to the target track fusion strategy, perform track fusion processing on any one of the two tracks that are determined to be fixed as a pair, and combine the track with the other track that forms a fixed as a pair to obtain the target associated track of the target object.

[0067] The aforementioned track association method involves acquiring the tracks corresponding to at least two sensors associated with the target object; for any two tracks among the tracks corresponding to each sensor, determining the track membership degree of the two tracks at the current moment; the two tracks include the tracks corresponding to the first sensor and the tracks corresponding to the second sensor; if the accuracy of the first sensor is higher than that of the second sensor, then determining the track correlation of the two tracks at the current moment based on the comparison between the track membership degree of the two tracks at the current moment and the resolution membership degree of the second sensor; based on the track correlation of the two tracks at the current moment, determining the track association quality value of the two tracks at at least one historical moment before the current moment; and ensuring that the track association quality value of the two tracks meets a preset association quality threshold. Under certain conditions, track fusion processing is performed on any two tracks to obtain the target associated track of the target object. Based on the track membership degree of any two tracks of the target object, a first threshold decision on track correlation is determined by comparing the track membership degree with the sensor resolution membership degree, and a second threshold decision on track association quality is determined based on the track of the continuous process of historical time. This can achieve accurate association of any two tracks. Furthermore, fusion processing is performed on two tracks whose track association quality values ​​meet the preset association quality threshold, thereby improving the stability of track association decision, enhancing the robustness of track association, and improving the accuracy of track association fusion results. This effectively solves the problem of low accuracy of track association fusion results in track analysis by traditional techniques.

[0068] In an exemplary embodiment, acquiring the tracks corresponding to at least two sensors associated with the target object includes: acquiring real-time measurement data of each sensor; extracting features from the real-time measurement data of each sensor to obtain track feature information of each track of the target; and determining the track corresponding to each sensor based on the track feature information of each sensor.

[0069] In its implementation, server 104 can acquire real-time measurement data from various sensors associated with the target object, such as radar sensors and infrared sensors, including parameters like heading angle, pitch angle, velocity, and displacement. It can then perform feature extraction on the real-time measurement data of each sensor to obtain the final real-time measurement data of each sensor. The characteristics of time are used to obtain the trajectory characteristic information of each track of the target. Also for each flight path in The track feature information corresponding to the track point at time, where Indicates the number of track features. Indicates the time; based on the trajectory characteristic information of each sensor. Thus, the individual sensors are determined. At the present moment Corresponding flight path .

[0070] Furthermore, based on clustering membership algorithms, according to the flight path... With each flight path The cluster centers are used to obtain the track membership degree between track feature information and cluster centers. Specifically, the following improved cluster center formula can be used to obtain the cluster centers of the track corresponding to each sensor. :

[0071]

[0072] In equation (2), This indicates at least two sensors associated with the target object. Their respective flight paths The following formula, which is an improvement upon the one used in this embodiment, represents the track membership degree:

[0073]

[0074] In equation (3), For scale parameters, This is represented as a membership control parameter. This embodiment uses the Euclidean distance between two tracks to represent the distance between the tracks and the cluster centers. As the track distance. In equation (2), when hour, This can represent the resolution membership degree of each sensor, and the corresponding sensor resolution of each sensor is... ,in, This represents the resolution of the sensor corresponding to each track feature; when hour, It can represent the trajectory corresponding to a sensor. Track corresponding to another sensor The degree of association between the tracks.

[0075] Furthermore, the following objective function for track clustering membership loss is constructed:

[0076]

[0077] In equation (4), Indicates the cluster center. Cluster center matrix; For track membership degree, For the track membership matrix, For scale parameters, This is represented as a membership control parameter.

[0078] In equation (4), This application can represent track anomalies with low membership degrees. The embodiments of this application introduce a probability metric for track anomalies (bad points). and possible track intersections or The objective function of track clustering membership was reconstructed. By optimizing track membership, the robustness to track anomalies was enhanced, and the data dimensionality was adaptively processed in multi-sensor (>2) and multi-target environments. This can avoid the decay of track association accuracy (or association success rate) including anomalies.

[0079] With the goal of minimizing the membership loss of track clustering as expressed in equation (4), the cluster centers are... Track distance Track membership Iterative optimization is performed to obtain the optimal track membership degree. Track membership The higher the membership degree, the stronger the credibility that the two tracks belong to the real target object; the lower the membership degree, the weaker the credibility that the two tracks belong to the real target object.

[0080] The technical solution of this embodiment extracts features from the real-time measurement data of each sensor to obtain track feature information that can characterize the target. Then, based on the track feature information of each sensor, the track corresponding to each sensor is determined, thus providing basic data support for subsequent iterative optimization to obtain track membership degree and track association based on track membership degree.

[0081] In an exemplary embodiment, performing track fusion processing on any two tracks to obtain the target associated track of the target object includes: obtaining the ratio of the maximum track membership degree to the average track membership degree in the track membership degree of the target object; selecting a target track fusion strategy that matches the ratio from multiple candidate track fusion strategies; and performing track fusion processing on any two tracks according to the target track fusion strategy to obtain the target associated track of the target object.

[0082] In the specific implementation, the track association quality value between any two tracks Under the condition that the preset correlation quality threshold is met, for example, when the correlation quality values ​​of the track can meet the requirements for 8 consecutive time periods. At that time, server 104 can target any two flight paths. One of the flight paths Obtain the track membership degree of the target object. Maximum membership degree of the track Average membership degree of the track ratio :

[0083]

[0084] Candidate track fusion strategies include when the ratio The first track fusion strategy when the ratio is greater than a preset threshold, and when the ratio is greater than a preset threshold threshold. The second track fusion strategy is used when the ratio is less than a preset threshold. Therefore, server 104 can select the appropriate strategy from multiple candidate track fusion strategies based on the ratio. A matching target trajectory fusion strategy; then, the server 104 can, according to the target trajectory fusion strategy, perform trajectory fusion processing on one of the two trajectories that are determined to be a fixed association pair, and on the other trajectories that form a fixed association pair with the trajectories, thereby obtaining the target associated trajectory of the target object.

[0085] The technical solution of this embodiment obtains the ratio of the maximum value of the track membership degree to the average value of the track membership degree in the track membership degree of the target object. This upgrades the track membership degree from a track association basis to a track fusion strategy selection basis. Furthermore, it can perform targeted track fusion processing on two tracks according to the target track fusion strategy that matches the ratio, thereby improving the stability of track association decision-making and realizing dynamic optimization and accurate fusion of any two tracks of the target object. This ensures that the obtained target association track is the globally optimal track of the target object.

[0086] In an exemplary embodiment, selecting a target trajectory fusion strategy that matches the ratio from multiple candidate trajectory fusion strategies includes: selecting a first trajectory fusion strategy as the target trajectory fusion strategy from multiple candidate trajectory fusion strategies when the ratio is greater than a preset ratio threshold; and selecting a second trajectory fusion strategy as the target trajectory fusion strategy from multiple candidate trajectory fusion strategies when the ratio is less than the preset ratio threshold.

[0087] The first track fusion strategy selects the dominant track from any two tracks as the target associated track for the target object. It can be understood that the dominant track can be any one of any two tracks that are determined as a fixed association pair if the ratio is greater than a preset ratio threshold and the track association quality value meets a preset association quality threshold.

[0088] The second track fusion strategy is used to perform weighted fusion on any two tracks, and the track obtained by weighted fusion is used as the target associated track of the target object.

[0089] In practical applications, according to equation (5), server 104 can obtain the track membership degree of the target object. Maximum membership degree of the track Average membership degree of the track ratio Among multiple candidate track fusion strategies, the one with the ratio is selected. A matching target trajectory fusion strategy.

[0090] For example, the preset ratio threshold can be 0.7, and the ratio is determined according to the definition of the following formula (6). Matching target track fusion strategy:

[0091]

[0092] Furthermore, in the ratio When the ratio exceeds a preset threshold, through the first track fusion strategy, server 104 can determine any one of the two tracks identified as a fixed association pair in the second threshold decision as the dominant track among the two tracks, and use this dominant track as the target association track of the target object; when the ratio If the ratio is less than the preset threshold, the server 104 can perform weighted fusion on the two tracks that are determined to be fixed association pairs in the second threshold decision through the second track fusion strategy, and use the track obtained by weighted fusion as the target associated track of the target object.

[0093] The technical solution of this embodiment selects a first track fusion strategy as the target track fusion strategy from multiple candidate track fusion strategies when the ratio is greater than a preset ratio threshold, and selects a second track fusion strategy as the target track fusion strategy when the ratio is less than the preset ratio threshold. This clarifies the two track fusion strategies corresponding to the track membership ratio, thereby providing an accurate track fusion strategy basis for targeted track fusion processing of two tracks.

[0094] In an exemplary embodiment, when the target track fusion strategy is the second track fusion strategy, track fusion processing is performed on any two tracks according to the target track fusion strategy to obtain the target associated track of the target object, including: obtaining the fusion weights corresponding to each track in any two tracks; performing a weighted summation on each track in any two tracks according to the fusion weights corresponding to each track in any two tracks to obtain the fused track, and using the fused track as the target associated track of the target object.

[0095] The fusion weights corresponding to the track include the membership degree based on the resolution of the sensor corresponding to the track.

[0096] In practical implementation, the track membership degree of the target object Maximum membership degree of the track Average membership degree of the track ratio Cases where the ratio is less than a preset threshold, for example This situation indicates that in the second threshold decision, the track association quality value of any two tracks... If the preset association quality threshold is met, even though there is no dominant track between the two tracks, the track reliability of the two tracks is balanced. Therefore, track fusion processing can be performed on any two tracks to obtain the target associated track of the target object.

[0097] Specifically, server 104 can connect any two flight paths Sensors corresponding to each flight path ( ) resolution membership As a fusion weight; according to any two paths Fusion weights corresponding to each flight path The following formula (7) can be used to determine the relationship between any two tracks. The weighted sum of the individual flight paths is used to obtain the merged flight path. :

[0098]

[0099] Furthermore, the merged flight paths The target-related track is the target object.

[0100] The technical solution of this embodiment clarifies the specific implementation of the second track fusion strategy, that is, by weighting and fusing the tracks in any two tracks according to the fusion weights corresponding to each track in any two tracks, and using the fused track as the target associated track of the target object, it is possible to efficiently fuse any two tracks that do not have a dominant track but have balanced track reliability into an optimal effective track when the ratio is less than a preset ratio threshold.

[0101] In an exemplary embodiment, the correlation between any two tracks at the current moment is determined based on the comparison between the track membership degree of any two tracks at the current moment and the resolution membership degree of the second sensor. This includes: if the track membership degree of any two tracks at the current moment is greater than the resolution membership degree of the second sensor, the correlation between the tracks at the current moment is determined to be that the two tracks are the same track; if the track membership degree of any two tracks at the current moment is less than the resolution membership degree of the second sensor, the correlation between the tracks at the current moment is determined to be that the two tracks are different tracks.

[0102] In a specific implementation, in the first threshold decision of the embodiments of this application, the first sensor Its accuracy is higher than that of the second sensor. In the case of accuracy, that is, the first sensor The resolution is smaller than that of the second sensor. In terms of resolution For any two tracks At the present moment The following track membership Greater than the resolution membership degree of the second sensor In the case of determining any two paths At the present moment Track correlation The two tracks are considered to be the same track; in any two tracks At the present moment The following track membership The membership degree is smaller than that of the second sensor. In the case of determining any two paths At the present moment Track correlation These two flight paths are different flight paths.

[0103] For example, suppose that the accuracy of the first sensor 1 is higher than that of the second sensor 2, indicating that the error of the measurement data measured by the first sensor 1 is lower than the error of the measurement data measured by the second sensor 2. Then, the resolution of the first sensor is lower than that of the second sensor. For global association decisions in track association, they can generally be made based on the minimum precision sensor, which is the sensor with the largest measurement noise. Under this premise, they can be defined according to the following formula (8):

[0104]

[0105] That is, it can be stipulated that when two tracks At the present moment The following track membership Greater than the resolution membership degree of the second sensor At that time, that is The two flight paths At the present moment track correlation When two tracks At the present moment The following track membership The membership degree is smaller than that of the second sensor. At that time, that is The two flight paths At the present moment Track correlation .

[0106] Furthermore, based on the decision criterion in formula (8) above, the correlation between the two tracks measured by the two sensors at the current moment can be determined as follows:

[0107]

[0108] The technical solution of this embodiment clarifies the specific implementation of the first threshold decision in track association, that is, it clarifies two comparison results between the track membership degree of any two tracks at the current time and the resolution membership degree of the second sensor. Based on the two comparison results, the track correlation of any two tracks can be determined accordingly, thereby providing an accurate basis for the subsequent second threshold decision.

[0109] In one exemplary embodiment, such as Figure 3 Based on the overall architecture shown, this application provides a track association method, including the following steps:

[0110] Step S302, Track Acquisition. The central server can acquire real-time measurement data from multiple sensors in the distributed sensor fusion system associated with the target object, such as radar sensors and infrared sensors, including heading angle, pitch angle, velocity, and displacement. Feature extraction is performed on the real-time measurement data of each sensor to obtain the track feature information of each track of the target. Based on the track feature information of each sensor, the track corresponding to each sensor is determined.

[0111] Step S304, Track Association. The central server can associate any two tracks from the tracks corresponding to each sensor (including the track corresponding to the first sensor and the track corresponding to the second sensor), as follows:

[0112] Based on the trajectory feature information of each sensor, the objective function of trajectory clustering membership degree is reconstructed by introducing a probability measure of trajectory anomalies (bad points), thereby optimizing the trajectory membership degree of any two trajectories at the current time.

[0113] In the first threshold decision of track association, the association decision is made based on the improved track membership degree at the current moment. That is, if the accuracy of the first sensor is higher than that of the second sensor, the track correlation of any two tracks at the current moment is determined by comparing the track membership degree of any two tracks at the current moment with the resolution membership degree of the second sensor.

[0114] In the second threshold decision of track association, the track quality is dynamically evaluated by combining historical tracks. That is, based on the track correlation of any two tracks at the current time, the track association quality value of any two tracks at at least one historical time before the current time is determined. In this way, the continuous process of historical time is introduced into the track association, and the problem of temporal correlation of tracks is considered, which can effectively improve the association accuracy in track intersection scenarios.

[0115] Therefore, this application embodiment constructs a dual guarantee of "instantaneous disturbance resistance + historical deviation correction" through a dual-threshold mechanism. In the first threshold decision, it resists track intersections by improving fuzzy clustering. or The fuzzy membership of the track at the location eliminates the false association spread caused by track anomalies (bad points); in the second threshold decision, the track association quality at historical moments is obtained through a sliding window to suppress the interference of isolated tracks, which can avoid association decision oscillation, thereby forming a full-time suppression capability against continuous interference and instantaneous anomalies, improving the stability of track association decision and enhancing the robustness of track association.

[0116] Step S306, Track Fusion. The central server can fuse any two tracks after track association. It can dynamically optimize the track fusion strategy based on the track membership association matrix and the track association quality value at historical time. Through an adaptive algorithm, it automatically selects the optimal fusion scheme and merges any two tracks that meet the track fusion conditions into an optimal effective track. Specifically, when the track association quality value of any two tracks meets the preset association quality threshold, track fusion processing is performed on any two tracks to obtain the target associated track of the target object.

[0117] Therefore, by upgrading the track membership matrix from a decision criterion to a fusion strategy selector, this application can dynamically switch and adapt the track fusion strategy for any two tracks based on the dominance and uniformity characteristics of track distribution. This can solve the core contradiction of the mismatch between the fixed fusion strategy and the dynamic battlefield environment in traditional solutions, and can deeply mine the information value of track association, thereby improving the accuracy of track association fusion results.

[0118] Step S308, Track Output. The central server can output the target object's associated track as a global track, thereby enabling the output of the target object's accurate associated track.

[0119] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0120] Based on the same inventive concept, this application also provides a track association apparatus for implementing the track association method described above. The solution provided by this apparatus is similar to the implementation scheme described in the above method; therefore, the specific limitations in one or more track association apparatus embodiments provided below can be found in the limitations of the track association method described above, and will not be repeated here.

[0121] In an exemplary embodiment, as shown in FIG4, a track association device is provided, including:

[0122] The track membership determination module 410 is used to acquire the tracks corresponding to at least two sensors associated with the target object; for any two tracks among the tracks corresponding to each sensor, the track membership of any two tracks at the current time is determined; any two tracks include the track corresponding to the first sensor and the track corresponding to the second sensor.

[0123] The track correlation determination module 420 is used to determine the track correlation of any two tracks at the current moment based on the comparison result between the track membership degree of any two tracks at the current moment and the resolution membership degree of the second sensor when the accuracy of the first sensor is higher than that of the second sensor.

[0124] The track association quality determination module 430 is used to determine the track association quality value of any two tracks at at least one historical time before the current time based on the track correlation of any two tracks at the current time.

[0125] The track fusion processing module 440 is used to perform track fusion processing on any two tracks when the track association quality value of any two tracks meets the preset association quality threshold, so as to obtain the target associated track of the target object.

[0126] In one embodiment, the track membership determination module 410 is further configured to acquire real-time measurement data from each sensor; extract features from the real-time measurement data of each sensor to obtain track feature information of each track of the target; and determine the track corresponding to each sensor based on the track feature information of each sensor.

[0127] In one embodiment, the track fusion processing module 440 is specifically used to obtain the ratio of the maximum value of track membership to the average value of track membership in the track membership of the target object; select the target track fusion strategy that matches the ratio from multiple candidate track fusion strategies; and perform track fusion processing on any two tracks according to the target track fusion strategy to obtain the target associated track of the target object.

[0128] In one embodiment, the track fusion processing module 440 is further configured to select a first track fusion strategy as the target track fusion strategy from multiple candidate track fusion strategies when the ratio is greater than a preset ratio threshold; the first track fusion strategy is used to select the dominant track among any two tracks as the target associated track of the target object; when the ratio is less than the preset ratio threshold, select a second track fusion strategy as the target track fusion strategy from multiple candidate track fusion strategies; the second track fusion strategy is used to perform weighted fusion on any two tracks and use the track obtained by weighted fusion as the target associated track of the target object.

[0129] In one embodiment, the track fusion processing module 440 is further configured to obtain the fusion weights corresponding to each track in any two tracks; the fusion weights corresponding to the tracks include the resolution membership degree of the sensor corresponding to the track; according to the fusion weights corresponding to each track in any two tracks, the tracks in any two tracks are weighted and summed to obtain the fused track, and the fused track is used as the target associated track of the target object.

[0130] In one embodiment, the track correlation determination module 420 is specifically used to determine that any two tracks are the same track when the track membership degree of any two tracks at the current time is greater than the resolution membership degree of the second sensor; and to determine that any two tracks are different tracks when the track correlation degree of any two tracks at the current time is less than the resolution membership degree of the second sensor.

[0131] Each module in the aforementioned trajectory association device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0132] In an exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram is shown in Figure 5. The computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is connected to the system bus via the I / O interfaces. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device stores track association data. The I / O interfaces of the computer device are used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a track association method.

[0133] Those skilled in the art will understand that the structure shown in Figure 5 is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or may combine certain components, or may have different component arrangements.

[0134] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the various embodiments of the above-described track association method.

[0135] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the various embodiments of the above-described track association method.

[0136] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps described in various embodiments of the method for calculating track association.

[0137] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0138] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic resistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence processors, etc., and are not limited to these.

[0139] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0140] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for associating flight paths, characterized in that, The method includes: Obtain the tracks corresponding to at least two sensors associated with the target object; for any two tracks among the tracks corresponding to each sensor, determine the track membership degree of the any two tracks at the current time; the any two tracks include the track corresponding to the first sensor and the track corresponding to the second sensor; If the accuracy of the first sensor is higher than that of the second sensor, then the correlation between the two tracks at the current moment is determined based on the comparison between the track membership degree of the two tracks at the current moment and the resolution membership degree of the second sensor. Based on the track correlation of any two tracks at the current time, determine the track association quality value of any two tracks at at least one historical time before the current time. If the trajectory association quality value of any two trajectories meets the preset association quality threshold, trajectory fusion processing is performed on the two trajectories to obtain the target associated trajectory of the target object.

2. The method according to claim 1, characterized in that, The acquisition of the tracks corresponding to at least two sensors associated with the target object includes: Acquire real-time measurement data from each of the aforementioned sensors; Feature extraction is performed on the real-time measurement data of each sensor to obtain the trajectory feature information of each trajectory of the target; Based on the trajectory feature information of each sensor, the trajectory corresponding to each sensor is determined.

3. The method according to claim 1, characterized in that, The step of performing track fusion processing on any two tracks to obtain the target-related track of the target object includes: Obtain the ratio of the maximum value of the track membership degree to the average value of the track membership degree in the track membership degree of the target object; Select the target trajectory fusion strategy that matches the ratio from among multiple candidate trajectory fusion strategies; According to the target track fusion strategy, track fusion processing is performed on any two tracks to obtain the target associated track of the target object.

4. The method according to claim 3, characterized in that, The step of selecting a target trajectory fusion strategy that matches the ratio from multiple candidate trajectory fusion strategies includes: If the ratio is greater than a preset ratio threshold, a first track fusion strategy is selected from the plurality of candidate track fusion strategies as the target track fusion strategy; the first track fusion strategy is used to select the dominant track among any two tracks as the target associated track of the target object; If the ratio is less than a preset ratio threshold, the second track fusion strategy is selected as the target track fusion strategy from the plurality of candidate track fusion strategies; the second track fusion strategy is used to perform weighted fusion on any two tracks, and the track obtained by weighted fusion is used as the target associated track of the target object.

5. The method according to claim 4, characterized in that, When the target track fusion strategy is the second track fusion strategy, the step of performing track fusion processing on any two tracks according to the target track fusion strategy to obtain the target associated track of the target object includes: Obtain the fusion weights corresponding to each of the two possible paths; the fusion weights corresponding to the paths include the membership degree based on the resolution of the sensor corresponding to the path. According to the fusion weights corresponding to each of the two possible paths, the paths in any two possible paths are weighted and summed to obtain the fused path, and the fused path is used as the target associated path of the target object.

6. The method according to claim 1, characterized in that, The step of determining the track correlation of any two tracks at the current moment based on the comparison result between the track membership degree of any two tracks at the current moment and the resolution membership degree of the second sensor includes: If the membership degree of any two tracks at the current time is greater than the resolution membership degree of the second sensor, the correlation between the tracks at the current time is determined to be that the two tracks are the same track. If the membership degree of any two tracks at the current time is less than the resolution membership degree of the second sensor, the correlation between the tracks at the current time is determined to be that the two tracks are different tracks.

7. A track correlation device, characterized in that, The device includes: The track membership determination module is used to acquire the tracks corresponding to at least two sensors associated with the target object; for any two tracks among the tracks corresponding to each sensor, the track membership degree of the arbitrary two tracks at the current time is determined; the arbitrary two tracks include the track corresponding to the first sensor and the track corresponding to the second sensor. The track correlation determination module is used to determine the track correlation of any two tracks at the current moment based on the comparison result between the track membership degree of any two tracks at the current moment and the resolution membership degree of the second sensor when the accuracy of the first sensor is higher than that of the second sensor. The track association quality determination module is used to determine the track association quality value of any two tracks at at least one historical moment before the current moment based on the track correlation of any two tracks at the current moment. The track fusion processing module is used to perform track fusion processing on any two tracks when the track association quality value of any two tracks meets a preset association quality threshold, so as to obtain the target associated track of the target object.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

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

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images