Movatterモバイル変換


[0]ホーム

URL:


CN118731892B - Interrupted track interconnection judgment method based on multi-dimensional feature information - Google Patents

Interrupted track interconnection judgment method based on multi-dimensional feature information
Download PDF

Info

Publication number
CN118731892B
CN118731892BCN202411215158.0ACN202411215158ACN118731892BCN 118731892 BCN118731892 BCN 118731892BCN 202411215158 ACN202411215158 ACN 202411215158ACN 118731892 BCN118731892 BCN 118731892B
Authority
CN
China
Prior art keywords
track
new
old
tracks
target
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202411215158.0A
Other languages
Chinese (zh)
Other versions
CN118731892A (en
Inventor
董云龙
王国庆
曹政
刘宁波
丁昊
胡昊
王杰
孙艳丽
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Naval Aeronautical University
Original Assignee
Naval Aeronautical University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Naval Aeronautical UniversityfiledCriticalNaval Aeronautical University
Priority to CN202411215158.0ApriorityCriticalpatent/CN118731892B/en
Publication of CN118731892ApublicationCriticalpatent/CN118731892A/en
Application grantedgrantedCritical
Publication of CN118731892BpublicationCriticalpatent/CN118731892B/en
Activelegal-statusCriticalCurrent
Anticipated expirationlegal-statusCritical

Links

Classifications

Landscapes

Abstract

The invention relates to an interrupted track interconnection judging method based on multidimensional characteristic information, and belongs to the technical field of track interconnection. The method comprises the following steps of 1, determining a new track set, filtering the head and tail points of the new track set, 2, predicting target prediction states of the new track set and the old track set at middle moments according to different motion models, 3, calculating spatial distances of two prediction positions, determining unified measurement of spatial consistency between the two tracks, defining membership functions of the two tracks, 4, determining membership functions of AD amplitude consistency, 5, forming feature judgment basis for judging whether the new track set corresponds to the same target by the spatial consistency and the AD amplitude consistency, 6, constructing an interconnection relation matrix for describing corresponding relation of the new track set and the old track set, and solving the problem. The method realizes a long-time and non-one-to-one interruption track steady association judgment method, and the method has high accuracy.

Description

Interrupt track interconnection judging method based on multidimensional characteristic information
Technical Field
The invention relates to a track interconnection judging method, in particular to an interrupted track interconnection judging method based on multidimensional characteristic information, and belongs to the technical field of track interconnection.
Background
The early warning detection radar is influenced by various factors such as target maneuver, electronic interference, detection distance limitation, speed blind area, nonuniform detection background and the like, and tracking interruption situations often occur to form a plurality of track segments. The flight path interruption causes a plurality of short flight paths to appear on the same target, and seriously influences the quality of early warning detection information. Therefore, in the post-hoc information analysis process, the interrupted track needs to be analyzed and judged to determine which track segments belong to the same target, and then the track segments are associated together to form continuous observation of the target.
During the track interruption, due to the lack of observation information, the unknowing and description difficulty of a target motion model in a tracking system are increased sharply, even a combination condition of a plurality of motion models can occur, the track prediction is difficult, and the subsequent track association judgment is seriously influenced. Aiming at the technical difficulties, a target motion model description method, a track prediction method and a correlation judgment method with long interruption time are needed to be studied deeply so as to solve the problem of difficult correlation of the interruption track.
The interactive multi-model interrupt track connection association algorithm adopts one-step prediction based on an interactive multi-model to predict backwards from a new track to a final updated point of an old track, then carries out chi-square test, and takes a log-likelihood function as a cost function to carry out final two-dimensional optimal matching. And (3) predicting the new track backwards and predicting the old track forwards by using an interactive multi-model-based method, performing chi-square test on the predicted value after the new track and the old track are predicted to the same moment, constructing a cost function matrix, and performing two-dimensional optimal allocation to obtain an associated track pair. Under the condition that measurement errors exist and the limitation of the existing filtering algorithm on track tracking is caused, the prediction errors on targets are larger, and the situation of wrong association and missed association is caused.
The multi-hypothesis motion model interruption track connection association algorithm based on priori information establishes a plurality of possible target motion models and carries out track prediction based on a multi-hypothesis idea, and a fuzzy correlation function based on position and speed information describes a fuzzy matching relation between a predicted track and a new starting track. The algorithm obviously reduces with the larger and larger association effect of the interrupt interval, and because multiple assumptions need to be made, too many models are selected for more accurate association, so that the calculated amount of the algorithm is increased.
An interrupt track association algorithm based on a statistical double threshold introduces a statistical double threshold principle, and increases the number of associated samples detected by a chi-square distribution threshold. When the target is in large maneuver, the maneuver in the interruption zone cannot be accurately described by the algorithm, and the association effect is seriously deteriorated.
Disclosure of Invention
The invention aims to solve the engineering practical problems of available multidimensional information, long track interruption time, quick change of a target motion mode and the like of an actual engineering scene, and provides an interrupted track interconnection judging method based on multidimensional characteristic information. The method fully utilizes multidimensional information such as space position, motion state, AD amplitude and the like, adopts technical methods such as fuzzy membership function, global nearest neighbor, bidirectional filtering prediction based on head-tail point state information and multiple models and the like, and realizes a long-time and non-one-to-one interruption track steady association judging method.
In order to solve the problems, the application is realized by the following technical scheme:
the interrupted track interconnection judging method based on the multidimensional characteristic information is characterized by comprising the following steps of:
Step 1, determining a new track set and an old track set, and filtering the head and tail points of the new track and the old track based on the head and tail point motion information;
Step 2, predicting a target prediction state of the middle moment of the new and old tracks according to the information of the head and tail points of the new and old tracks obtained in the step 1 and different motion models;
Step 3, calculating the space distance between two predicted positions according to the target predicted state of the new and old tracks at the middle moment obtained in the step 2, determining the unified measure of the space consistency between the two tracks, and defining the membership function;
step 4, determining a membership function of the AD amplitude consistency;
Step 5, the space consistency in the step 3 and the AD amplitude consistency in the step4 form a characteristic judgment basis for judging whether the new track and the old track correspond to the same target;
And 6, constructing an interconnection relation matrix for describing the corresponding relation of the new and old tracks, and solving.
Further, the step 1 specifically includes the following steps:
step 1.1, determining new and old flight path sets, wherein the set formed by all flight paths is formed by the following steps when early warning detection information is subjected to integral analysisWhereinIn the case of a track segment,,
Wherein the method comprises the steps ofRepresents the firstThe starting time of the track is the same,Represent the firstTrack termination time;
Setting upTime of dayDefinition ofFor the earliest track segment end time,The starting moment of the latest track segment;
the old track is gathered into,The new track is gathered into,,;
Step 1.2, filtering the head and tail points of the new and old tracks, namely respectively carrying out forward sequence filtering and reverse sequence filtering from the starting moment and the ending moment of the old track to the interruption moment by utilizing the motion state information of the tracks at the front end and the rear end;
For a certain pair of new and old tracksThe old track isForward filtering toAt the moment, obtain;
New trackInverse filtering toAt the moment, obtain
Further, the step 2 includes the steps of:
Step 2.1, determining target prediction states of two tracks at intermediate moments:
based on the filtering state of the head and tail points of the new and old tracks, namelyUpdate status of time old trackAndUpdate status of new track at momentTo the middle timePredicting the target state of new and old tracks, i.e.
,,
Wherein: AndIn the form of a state transition matrix,Respectively representing target prediction states of the old track and the new track at the middle moment;
Step 2.2, selecting different motion models, and determining a state transition matrix to obtain target prediction states of two tracks at the middle moment;
the method comprises the steps that a target moves in a three-dimensional space, and states of the target are predicted by using state transition matrixes of three motion modes such as uniform speed straight lines, uniform acceleration straight lines and uniform speed turning;
For uniform-speed straight line and uniform acceleration straight line motion modes, the state transition matrix of the target used at the moment is not in coupling relation between different vector coordinatesConsists of three one-dimensional motion state transfer matrices, i.e,
Wherein: is a one-dimensional motion state transition matrix, when uniform linear prediction is performedWhen the uniform acceleration linear prediction is performed;
,,
For the uniform turning motion mode, the coupling relation exists between different vector coordinates, so that the state transition matrix of the target is used at the momentConsists of a two-dimensional uniform turning motion state transfer matrix and a one-dimensional uniform linear motion state transfer matrix, namely,
In the middle ofAndThe state transition matrixes respectively represent the state transition matrixes corresponding to the uniform turning motion of the target in the x-y plane and the uniform motion of the target in the z axis;
,;
and selecting corresponding state transition matrixes according to different motion models, respectively solving the state transition matrixes of the new and the old tracks at the middle moment, and further obtaining the target prediction states of the two tracks at the middle moment.
Further, the step 3 specifically includes the following steps:
step 3.1, calculating the spatial distance between two predicted positions:
step 3.1.1 for a three-dimensional non-coupled constant motion model CV, spatial distance of two predicted positionsThe method comprises the following steps:;
Step 3.1.2 spatial distance between two predicted positions for a three-dimensional non-coupled uniformly accelerated motion model CAThe method comprises the following steps:;
Step 3.1.3, based on the heading angle of the head and tail points of the new and old tracksAndCalculating the turning rate average valueAccording to respectivelyAndSpatial distance between two predicted positionsThe method comprises the following steps:;
Wherein:;
step 3.1.4, turning rate average value based on old track end point and preamble t momentTurning rate average value based on new track head point and subsequent t momentAccording to respectivelyAnd
Spatial distance between two predicted positionsThe method comprises the following steps:;
Wherein:,,
step 3.2, the maximum space distance between the new track and the old trackThe method is defined as a unified measure of the space consistency between two tracks, and a membership function is defined:
Wherein:,
Is the minimum value of the space distance, namely:,
the membership function is:
further, the step 4 specifically includes:
Selecting the maximum value of the AD amplitude in the track as a standard value thereof, and defining a corresponding membership function as follows: , ,
Wherein: AndFor the AD amplitude value corresponding to the new and old tracks,And obtaining or taking the maximum AD amplitude value in the new and old tracks through statistical historical data for the AD amplitude statistical variance of the same track.
Further, the step 5 is specifically that the membership functions of the spatial consistency in the step 3 and the AD amplitude consistency in the step 4 are integrated with the weights corresponding to the membership functions to obtain the correlation decision valueAs a judgment basis:
WhereinIs the weight.
Further, the step 6 specifically includes the following steps:
step 6.1, constructing an interconnection relation matrix and an association decision matrix corresponding to the new track and the old track:
Definition of the definitionIs of the correlation decision matrix of (a)To describe the associated decision values of new and old tracks,,
Wherein: Represent the firstOld track and firstThe association decision value of the new tracks, m and n respectively represent the number of old tracks and the number of new tracks which participate in association decision;
Definition of the definitionIs of the interconnected relation matrix of (a)To describe the correspondence of new and old tracks,,
Indicate the judgment of the firstOld track and firstThe new track of the strip is not associated with,Indicate the judgment of the firstOld track and firstAssociating the new tracks;
because of the non-one-to-one correspondence between new and old tracks, the interconnection matrix satisfies,;
Step 6.2, solving the interconnection relation matrix of step 6.1, which specifically comprises the following steps:
Step 6.2.1, timing sequence initial judgment:
based on old track end timeLess than the new track start timeAnd (3) constraint relation, setting the matrix elements which do not meet the condition in the step 4.1 to zero, namely:,
To interrupt the operation for a time longer than a thresholdIs no longer associated, i.e;
Step 6.2.2, initial judgment of the movement relation:
At maximum speedMaximum accelerationMinimum speedAnd maximum turning rateThe non-zero elements in the interconnection relation matrix are screened and updated, the corresponding element value of the track pair meeting the following conditions is set as 1, and the rest is set as 0:
calculating Euclidean distance between position of old track end time and position of new track start time in track pairSpeed constraint is performed on the track pair according to the speed obtained by dividing the distance by the interruption time:
,,
According to the speed of the end moment of the old trackAnd the speed of the new track start timeDividing the interruption time to obtain the acceleration of the track pair to limit the acceleration of the track pair:,
From previously calculated turn rate meansTo limit the turn rate of the track pair:,
Wherein, maximum speedSelecting according to historical track data of the air and water surface targets, and maximum accelerationSet according to the air and water surface target characteristics or historical data, minimum speedSetting according to historical data of the air and water surface targets;
Maximum turning rate of water surface targetSetting according to the target characteristics of the water surface target, and the maximum turning rate of the air targetSet according to the characteristics of the object in the air or set according to the maximum turning rate of the uninterrupted track multiplied by the amplification factor,,In order to amplify the coefficient of the power,For a maximum turn rate of the uninterrupted track,Is obtained for traversing the track segment;
step 6.2.3, global optimal decision:
selecting an interconnection relation matrixThe element with the highest association decision value corresponding to all the remaining non-zero elements in the system is judged that the corresponding new and old tracks are successfully associated, and then the element is positioned in the row and column elementsAnd (3) setting zero, sequentially carrying out iterative judgment, setting zero until all elements of the interconnection relation matrix are set to zero, and obtaining all new and old track association judgment results.
Based on effective technical measures such as bidirectional filtering prediction and global optimization, the application fully utilizes multidimensional information such as spatial position, motion state, AD amplitude and the like, adopts technical methods such as fuzzy membership function, global nearest neighbor, bidirectional filtering prediction based on head-tail point state information and multiple models and the like, and realizes a long-time and non-one-to-one interruption track steady association judging method.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a leading track association diagram;
FIG. 3 is a graph of method association effects corresponding to different interrupt durations;
FIG. 4 is a graph of the correlation effects of different track numbers corresponding to the method;
FIG. 5 is a graph of the correlation effect of different track loss rate corresponding methods.
Detailed Description
The following description of the present invention will be given with reference to the accompanying drawings, which are used to further explain the constitution of the present invention.
Example 1. An interrupted track interconnection judging method based on multidimensional characteristic information, as shown in fig. 1, comprises the following steps:
Step 1, determining a new track set and an old track set, and filtering the head and tail points of the new track and the old track based on the head and tail point motion information;
step 2, predicting a target prediction state of the middle moment of the new and old tracks according to the information of the new and old track head positions obtained in the step 1 and different motion models;
Step 3, calculating the space distance between two predicted positions according to the target predicted state of the new and old tracks at the middle moment obtained in the step 2, determining the unified measure of the space consistency between the two tracks, and defining the membership function;
step 4, determining a membership function of the AD amplitude consistency;
Step 5, the space consistency in the step 3 and the AD amplitude consistency in the step4 form a characteristic judgment basis for judging whether the new track and the old track correspond to the same target;
And 6, constructing an interconnection relation matrix for describing the corresponding relation of the new and old tracks, and solving.
Further, the step 1 specifically includes the following steps:
step 1.1, determining new and old flight path sets, wherein the set formed by all flight paths is formed by the following steps when early warning detection information is subjected to integral analysisWhereinIn the case of a track segment,,
Wherein the method comprises the steps ofRepresents the firstThe starting time of the track is the same,Represent the firstTrack termination time;
Setting upTime of dayDefinition ofFor the earliest track segment end time,The starting moment of the latest track segment;
the old track is gathered into,;
The new track is gathered into,,
Step 1.2, filtering the head and tail points of the new and old tracks:
forward sequence filtering and reverse sequence filtering are respectively carried out from the starting moment to the stopping moment of the old track and the ending moment of the new track by utilizing the motion state information of the tracks at the front end and the rear end;
For a certain pair of new and old tracksThe old track isForward filtering toAt the moment, obtain;
New trackInverse filtering toAt the moment, obtain
Further, the step 2 includes the steps of:
Step 2.1, determining target prediction states of two tracks at intermediate moments:
based on the filtering state of the head and tail points of the new and old tracks, namelyUpdate status of time old trackAndUpdate status of new track at momentTo the middle timePredicting the target state of new and old tracks, i.e.
,,
Wherein: AndIn the form of a state transition matrix,Respectively representing target prediction states of the old track and the new track at the middle moment;
Step 2.2, selecting different motion models, and determining a state transition matrix to obtain target prediction states of two tracks at the middle moment;
the method comprises the steps that a target moves in a three-dimensional space, and states of the target are predicted by using state transition matrixes of three motion modes such as uniform speed straight lines, uniform acceleration straight lines and uniform speed turning;
For uniform-speed straight line and uniform acceleration straight line motion modes, the state transition matrix of the target used at the moment is not in coupling relation between different vector coordinatesConsists of three one-dimensional motion state transfer matrices, i.e,
Wherein: is a one-dimensional motion state transition matrix, when uniform linear prediction is performedWhen the uniform acceleration linear prediction is performed;
,,
For the uniform turning motion mode, the coupling relation exists between different vector coordinates, so that the state transition matrix of the target is used at the momentConsists of a two-dimensional uniform turning motion state transfer matrix and a one-dimensional uniform linear motion state transfer matrix, namely,
In the middle ofAndThe state transition matrixes respectively represent the state transition matrixes corresponding to the uniform turning motion of the target in the x-y plane and the uniform motion of the target in the z axis;
,;
and selecting corresponding state transition matrixes according to different motion models, respectively solving the state transition matrixes of the new and the old tracks at the middle moment, and further obtaining the target prediction states of the two tracks at the middle moment.
Further, the step 3 specifically includes the following steps:
step 3.1, calculating the spatial distance between two predicted positions:
step 3.1.1 for a three-dimensional non-coupled constant motion model CV, spatial distance of two predicted positionsThe method comprises the following steps:;
Step 3.1.2 spatial distance between two predicted positions for a three-dimensional non-coupled uniformly accelerated motion model CAThe method comprises the following steps:;
Step 3.1.3, based on the heading angle of the head and tail points of the new and old tracksAndCalculating the turning rate average valueAccording to respectivelyAndSpatial distance between two predicted positionsThe method comprises the following steps:;
Wherein:;
step 3.1.4, turning rate average value based on old track end point and preamble t momentTurning rate average value based on new track head point and subsequent t momentAccording to respectivelyAnd
Spatial distance between two predicted positionsThe method comprises the following steps:;
Wherein:,;
step 3.2, the maximum space distance between the new track and the old trackThe method is defined as a unified measure of the space consistency between two tracks, and a membership function is defined:
Wherein:,
Is the minimum value of the space distance, namely:,
the membership function is:
further, the step 4 specifically includes:
Selecting the maximum value of the AD amplitude in the track as a standard value thereof, and defining a corresponding membership function as follows:, ,
Wherein: AndFor the AD amplitude value corresponding to the new and old tracks,And obtaining or taking the maximum AD amplitude value in the new and old tracks through statistical historical data for the AD amplitude statistical variance of the same track.
Further, the step 5 is specifically that the membership functions of the spatial consistency in the step 3 and the AD amplitude consistency in the step 4 are integrated with the weights corresponding to the membership functions to obtain the correlation decision valueAs a judgment basis:
WhereinThe weight is not fixed, the weight of the spatial consistency attribute becomes smaller as the interruption time increases, for example, the weight can be adjusted to 0.5 within 1 minute and adjusted to 0.3 above 5 minutes when being selected to 0.7,1 minutes, and the like, and meanwhile, the spatial consistency weight of the sea surface target can be further improved to 0.8 and is smaller as the interruption time changes.
Further, the step 6 specifically includes the following steps:
step 6.1, constructing an interconnection relation matrix and an association decision matrix corresponding to the new track and the old track:
Definition of the definitionIs of the correlation decision matrix of (a)To describe the associated decision values of new and old tracks,,
Wherein: Represent the firstOld track and firstThe association decision value of the new tracks, m and n respectively represent the number of old tracks and the number of new tracks which participate in association decision;
Definition of the definitionIs of the interconnected relation matrix of (a)To describe the correspondence of new and old tracks,,
Indicate the judgment of the firstOld track and firstThe new track of the strip is not associated with,Indicate the judgment of the firstOld track and firstAssociating the new tracks;
because of the non-one-to-one correspondence between new and old tracks, the interconnection matrix satisfies,;
Step 6.2, solving the interconnection relation matrix of step 6.1, which specifically comprises the following steps:
Step 6.2.1, timing sequence initial judgment:
based on old track end timeLess than the new track start timeAnd (3) constraint relation, setting the matrix elements which do not meet the condition in the step 4.1 to zero, namely:,
To interrupt the operation for a time longer than a thresholdIs no longer associated, i.e;
Step 6.2.2, initial judgment of the movement relation:
At maximum speedMaximum accelerationMinimum speedAnd maximum turning rateThe non-zero elements in the interconnection relation matrix are screened and updated, the corresponding element value of the track pair meeting the following conditions is set as 1, and the rest is set as 0:
calculating Euclidean distance between position of old track end time and position of new track start time in track pairSpeed constraint is performed on the track pair according to the speed obtained by dividing the distance by the interruption time:
,,
According to the speed of the end moment of the old trackAnd the speed of the new track start timeDividing the interruption time to obtain the acceleration of the track pair to limit the acceleration of the track pair:,
From previously calculated turn rate meansTo limit the turn rate of the track pair:,
Wherein, maximum speedSelecting according to historical track data of the air and water surface targets, and maximum accelerationSet according to the air and water surface target characteristics or historical data, minimum speedSetting according to historical data of the air and water surface targets;
Maximum turning rate of water surface targetSetting according to the target characteristics of the water surface target, and the maximum turning rate of the air targetSet according to the characteristics of the object in the air or set according to the maximum turning rate of the uninterrupted track multiplied by the amplification factor,,In order to amplify the coefficient of the power,For a maximum turn rate of the uninterrupted track,Is obtained for traversing the track segment;
step 6.2.3, global optimal decision:
selecting an interconnection relation matrixThe element with the highest association decision value corresponding to all the remaining non-zero elements in the system is judged that the corresponding new and old tracks are successfully associated, and then the element is positioned in the row and column elementsAnd (3) setting zero, sequentially carrying out iterative judgment, setting zero until all elements of the interconnection relation matrix are set to zero, and obtaining all new and old track association judgment results.
Example 2. This example is a simulation of the method described in example 1.
The IMM-TSA algorithm is an interrupt track association algorithm based on an interactive multi-model;
IMMSDP-TSA algorithm is state correlation transfer probability interactive multi-model interruption track association algorithm;
the MHMPI-TSA algorithm is a multi-hypothesis model outage path association algorithm based on prior information.
The method provided by the invention is respectively compared with an IMM-TSA algorithm, a IMMSDP-TSA algorithm and a MHMPI-TSA algorithm through simulation experiments, the average accuracy of the method is higher than that of the three algorithms, and the method has better accuracy and effectiveness when the interruption track dataset is processed.
Experiment 1. The experiment uses a radar to detect the targets, and the method is compared with the performances of the other three algorithms by setting different interrupt time lengths under the assumption that 100 targets exist, 5 state switching of the targets occurs in 600 sampling periods, and no loss occurs in the track.
Experiment 2, the environment is the same as above, the interruption time of the target is 100s, the track is not lost, and different track numbers are set to compare the performance of the method with that of other three algorithms.
Experiment 3, the environment is the same as above, 100 targets exist, the occurrence interruption time of each target is 100s, and different track loss rates are set to compare the performance of the method with that of other three algorithms.
From the experimental results in fig. 3-5, it can be seen that the average correlation accuracy of the method proposed by the present application in various scenarios is above the other three algorithms. The increase of the interruption time length of the track and the increase of the track number can reduce the performance of the algorithm, but the method has less influence.

Claims (10)

CN202411215158.0A2024-09-022024-09-02 Interrupted track interconnection judgment method based on multi-dimensional feature informationActiveCN118731892B (en)

Priority Applications (1)

Application NumberPriority DateFiling DateTitle
CN202411215158.0ACN118731892B (en)2024-09-022024-09-02 Interrupted track interconnection judgment method based on multi-dimensional feature information

Applications Claiming Priority (1)

Application NumberPriority DateFiling DateTitle
CN202411215158.0ACN118731892B (en)2024-09-022024-09-02 Interrupted track interconnection judgment method based on multi-dimensional feature information

Publications (2)

Publication NumberPublication Date
CN118731892A CN118731892A (en)2024-10-01
CN118731892Btrue CN118731892B (en)2025-01-21

Family

ID=92867497

Family Applications (1)

Application NumberTitlePriority DateFiling Date
CN202411215158.0AActiveCN118731892B (en)2024-09-022024-09-02 Interrupted track interconnection judgment method based on multi-dimensional feature information

Country Status (1)

CountryLink
CN (1)CN118731892B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN119003848B (en)*2024-10-242025-04-18中国人民解放军海军航空大学 A method for associating interrupted tracks based on situation graph classification
CN120067819B (en)*2025-04-282025-07-25中国人民解放军海军航空大学Interrupt track association method, system, terminal and medium based on track prediction

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN106980114A (en)*2017-03-312017-07-25电子科技大学Target Track of Passive Radar method
CN118625307A (en)*2018-12-292024-09-10深圳市卓驭科技有限公司 Target track determination method, target tracking system and vehicle
CN113254869B (en)*2021-04-092024-08-02杭州电子科技大学Interrupt track association method based on Gaussian process
CN116106890B (en)*2023-02-082025-07-22南京莱斯电子设备有限公司Radar target track starting method based on quantum particle swarm and LGBM
CN116304536A (en)*2023-03-302023-06-23中航(成都)无人机系统股份有限公司Distributed multi-target tracking method, device, equipment and storage medium
CN117724087B (en)*2024-02-072024-05-28中国人民解放军海军航空大学Radar multi-target tracking double-tag Bernoulli filtering algorithm

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
Contrastive Transformer Network for Track Segment Association with Two-Stage Online Method;Zongqing Cao et al.;remote sensing;20240911;第16卷;全文*

Also Published As

Publication numberPublication date
CN118731892A (en)2024-10-01

Similar Documents

PublicationPublication DateTitle
CN118731892B (en) Interrupted track interconnection judgment method based on multi-dimensional feature information
CN109188423B (en) A distributed multi-target tracking method based on multi-source clustering
CN106443622B (en)A kind of distributed object tracking based on improvement joint probability data association
CN109633589A (en)The Multi-target Data Associations assumed are optimized based on multi-model more in target following
CN105182311B (en)Omnidirectional's radar data processing method and system
CN108152789B (en)Utilize the passive track-corelation data correlation and localization method of RSS information
CN105629198B (en)The indoor multi-target tracking method of fast search clustering algorithm based on density
CN106405537B (en)Radar track initial mode based on location information and doppler information
CN117169818B (en) A radar waveform design method for sea surface maneuvering target tracking
CN109858526B (en) A Sensor-Based Multi-Target Trajectory Fusion Method in Target Tracking
CN110376582B (en) A Maneuvering Target Tracking Method Based on Adaptive GM-PHD
CN108344981B (en) Clutter-oriented multi-sensor asynchronous detection TSBF multi-target tracking method
CN110673130B (en)Moving target track tracking method based on track association
CN103759732A (en)Angle information assisted centralized multi-sensor multi-hypothesis tracking method
CN114002667B (en)Multi-neighbor extended target tracking algorithm based on random matrix method
CN117665795B (en)Poisson Bernoulli extended target tracking method based on track set
CN106291498A (en)A kind of detecting and tracking combined optimization method based on particle filter
CN112799028B (en)False target identification method based on RCS fluctuation statistical characteristic difference
CN117214881B (en)Multi-target tracking method based on Transformer network in complex scene
CN114840018B (en) A method for determining the correlation of UAV trajectories in monitoring blind areas
CN115220002B (en) A method and device for multi-target data association tracking at a fixed single station
CN111811515B (en) A Multi-target Track Extraction Method Based on Gaussian Mixture Probability Hypothesis Density Filter
CN113219425A (en)Test method and system for radar target detection performance
CN115114985A (en) A Distributed Fusion Method for Sensor Systems Based on Set Theory
CN113311420B (en)Radar target detection and tracking method and device based on clustering management

Legal Events

DateCodeTitleDescription
PB01Publication
PB01Publication
SE01Entry into force of request for substantive examination
SE01Entry into force of request for substantive examination
GR01Patent grant
GR01Patent grant

[8]ページ先頭

©2009-2025 Movatter.jp