CN109298413A - A kind of method that specific aim solves the problems, such as the multiple target tracking data correlation under complex electromagnetic environment - Google Patents
A kind of method that specific aim solves the problems, such as the multiple target tracking data correlation under complex electromagnetic environment Download PDFInfo
- Publication number
- CN109298413A CN109298413A CN201811017254.9A CN201811017254A CN109298413A CN 109298413 A CN109298413 A CN 109298413A CN 201811017254 A CN201811017254 A CN 201811017254A CN 109298413 A CN109298413 A CN 109298413A
- Authority
- CN
- China
- Prior art keywords
- target
- measurement
- public
- public measurement
- association
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000005259 measurement Methods 0.000 claims abstract description 123
- 238000005070 sampling Methods 0.000 claims abstract description 8
- 239000011159 matrix material Substances 0.000 claims description 35
- 230000000694 effects Effects 0.000 claims description 6
- 238000013210 evaluation model Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 abstract description 4
- 238000011160 research Methods 0.000 abstract description 3
- 238000004364 calculation method Methods 0.000 description 4
- 238000001914 filtration Methods 0.000 description 4
- 230000033001 locomotion Effects 0.000 description 4
- 238000012512 characterization method Methods 0.000 description 3
- 238000012790 confirmation Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000001788 irregular Effects 0.000 description 2
- 241000287196 Asthenes Species 0.000 description 1
- 238000004380 ashing Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/66—Radar-tracking systems; Analogous systems
- G01S13/72—Radar-tracking systems; Analogous systems for two-dimensional tracking, e.g. combination of angle and range tracking, track-while-scan radar
- G01S13/723—Radar-tracking systems; Analogous systems for two-dimensional tracking, e.g. combination of angle and range tracking, track-while-scan radar by using numerical data
- G01S13/726—Multiple target tracking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Mathematical Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computational Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Probability & Statistics with Applications (AREA)
- Evolutionary Biology (AREA)
- Operations Research (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Algebra (AREA)
- Computer Networks & Wireless Communication (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention belongs to electromagnetism research fields, the method of the multiple target tracking data correlation under complex electromagnetic environment is solved the problems, such as more particularly to a kind of specific aim, the following steps are included: obtaining all position measurement information and other characteristic parameter information from target airspace in a certain sampling instant, fan-shaped tracking gate is set around the predicted value of existing each targetpath, the validity of all measurements is determined with this, and the measurement in multiple target following doors intersection region, i.e., public measurement are fallen into from screening in all effective measurements；The method of the present invention improves multiple target tracking precision, it can be achieved that carrying out more real-time effective tracking to unfriendly target by the accuracy of raising data correlation；The method of the present invention has good expansibility, for a variety of different track algorithms, can carry out data correlation using this method, and it can be made to filter convergence rate and be improved significantly；The method of the present invention is adapted to a variety of Nonlinear Parameter type of sports, and robustness is higher.
Description
Technical field
The invention belongs to electromagnetism research fields, and in particular to a kind of specific aim solve the multiple target under complex electromagnetic environment with
The method of track data correlation problem.
Background technique
Under the electromagnetic environment of even more complex and interference effect, the defect of available data association and target following technology is gradually
Show, can not adapt to the new demand of the application scenarios of all kinds of fast developments under the new situation.Probabilistic Data Association Algorithm is applicable in
Measurement-track association problem under clutter environment when monotrack, and measuring the situation for falling into multiple target overlapping region
Under then far from the association accuracy requirement for meeting target following.Joint Probabilistic Data Association algorithm be generally believe at present most at
The data association algorithm of function, especially when carrying out real-time tracking to the multiple target under more clutters, joint probability data association is calculated
Method introduces the concept of " poly- ", is estimated by building joint event each dbjective state, has given full play to its advantage.But work as
When number of targets or clutter number increase, joint event can be in exponential increase, so as to cause calculation amount abruptly increase, or even causes and " combines quick-fried
It is fried " phenomenon, it is unable to satisfy requirement of the target following to real-time.In order to solve this problem, the data of various feature auxiliary are closed
Connection algorithm is suggested in succession, by other features except target position information or during parameter is introduced into data correlation, to
The higher interrelating effect of accuracy is obtained, to improve the precision and real-time of target following.Herein in conjunction with hesitation fuzzy set ash
Change thought and propose a kind of new feature auxiliary data association algorithm, focus on solving multiple short distances are done under complex electromagnetic environment it is small
The problem of angle cross non-linear moving target is tracked.
In conclusion there is the problems such as association accuracy is not high, calculation amount is excessive in the prior art.
Summary of the invention
Solve the problems, such as the multiple target tracking under complex electromagnetic environment, present invention is generally directed to property in multiple target target
When doing the nonlinear motion of short distance low-angle intersection, the precision and real-time of target following are still ensured that.
A kind of method that specific aim solves the problems, such as the multiple target tracking data correlation under complex electromagnetic environment, including following step
It is rapid:
(1) all position measurement information and other characteristic parameter information are obtained from target airspace in a certain sampling instant,
Fan-shaped tracking gate is set around the predicted value of existing each targetpath, determines the validity of all measurements with this, and from institute
Screening falls into the measurement in multiple target following doors intersection region, i.e., public measurement in effectively measuring；
(2) it is directed to each public measurement, determination target relevant to its possibility is joined in conjunction with feature corresponding with public measurement
Number information, seeks the public measurement degree of association associated with each target under various possible radar signal types, is re-introduced into still
The thought of Henan fuzzy set assesses a possibility that public measurement is related to each target, and calculates public measurement phase based on grey possibility degree
For the allocation probability of each target；
(3) each target is distributed into all public measurements according to the probability acquired in step 2, multiple target tracking problem is turned
The monotrack problem for turning to amendment association probability, estimates dbjective state using Probabilistic Data Association Algorithm；
(4) target following error co-variance matrix and each target signature parameter center are updated, and so on iteration is with reality
Now to the real-time tracking of multiple target.
It is described to be directed to each public measurement, determination target relevant to its possibility, in conjunction with feature corresponding with public measurement
Parameter information is sought the public measurement degree of association associated with each target under various possible radar signal types, is re-introduced into
The thought of hesitation fuzzy set assesses a possibility that public measurement is related to each target, and calculates public measurement based on grey possibility degree
Allocation probability relative to each target, comprising:
(2.1) assume to share N in t momentsA public measurement, and for public measurement js, fall intoA target following
The intersection region of door, i.e. js=1,2 ..., Ns,Below for public measurement jsWith target isIt is introduced,
Based on location informationCharacteristic parameter information corresponding with public measurementIt is associated with each target to seek the public measurement under various possible radar signal types
The degree of association；
(2.2) for public measurement js, introduce hesitation fuzzy set thought, under the different type that front is acquired, each feature
The measurement of parameter characterization is arranged with each target association level data, establishes hesitation fuzzy decision matrix
Due toInterior each data are profit evaluation model, therefore programming decision-making matrix
(2.3) programming decision-making matrix is soughtIn each fuzzy member core and upper and lower deviation value；
If the length for the fuzzy member h that hesitates is L, hi(i=1,2 ..., L) regards discrete grey fuzzy number as, claims
For the core for the fuzzy member h that hesitates；Referred to as upper deviation value；Deviation value is referred to as descended,
For the citation form of core and deviation value, incited somebody to action according to defined aboveBe converted to the decision matrix of core Yu upper and lower deviation value
(2.4) reference sequence is formedWhereinForEach maximum element of column center value, k ∈ [sta, PRI, PW, CF]；
According to reference sequence and decision matrixObtain grey incidence coefficientAnd grey relational grade
Indicate the distance of two hesitation Fuzzy Grey numbers, ρ is resolution ratio, ρ=0.5；
(2.5) weighted value Ω=[ω (1), ω (2), ω (3), ω (4)] of each characteristic attribute is sought, wherein
(2.6) according to weighted value, weighted comprehensive attribute value is determined:
Do not subtract axiom according to gray scale it is found thatGray scale be equal to the gray scale maximum value for participating in all grey numbers of operation, and
It will not reduce or reduce；
(2.7) will likely the corresponding weighted comprehensive attribute value of associated objects from core and upper and lower deviation value form be reduced to routine
Grey number form formula:
Wherein,
To it is all may be with public measurement jsAssociated target is compared and seeks respective possibility degree:
It enablesThen ordering vector isAnd
(2.8) public measurement and the allocation probability matrix Y to each target are sought, wherein
Wherein, α is the proportion adjustment factor.
The beneficial effects of the present invention are:
(1) the method for the present invention improves multiple target tracking precision, it can be achieved that enemy by the accuracy of raising data correlation
Target carries out more real-time effective tracking；
(2) there is the method for the present invention good expansibility can use the party for a variety of different track algorithms
Method carries out data correlation, and it can be made to filter convergence rate and be improved significantly；
(3) the method for the present invention is adapted to a variety of Nonlinear Parameter type of sports, and robustness is higher.
Detailed description of the invention
Fig. 1 is the feature auxiliary mark tracking data correlating method block diagram based on hesitation fuzzy set.
Specific embodiment
The present invention is described further with reference to the accompanying drawing.
The present invention is mainly discussed and is solved under interference and the intensive complex electromagnetic environment of clutter, while small to closely doing
When multiple targets of angle cross non-linear movement are tracked, data correlation less effective leads to tracking filter result and target
System real time decline when actual flight path generation relatively large deviation or algorithm complexity surge cause to track target, generates
The problem of even missing optimum reacting time compared with long time delay.The field being related to is mainly data correlation and target following, is
Current important and hot research field.
Under the electromagnetic environment of even more complex and interference effect, the defect of available data association and target following technology is gradually
Show, can not adapt to the new demand of the application scenarios of all kinds of fast developments under the new situation.Probabilistic Data Association Algorithm is applicable in
Measurement-track association problem under clutter environment when monotrack, and measuring the situation for falling into multiple target overlapping region
Under then far from the association accuracy requirement for meeting target following.Joint Probabilistic Data Association algorithm be generally believe at present most at
The data association algorithm of function, especially when carrying out real-time tracking to the multiple target under more clutters, joint probability data association is calculated
Method introduces the concept of " poly- ", is estimated by building joint event each dbjective state, has given full play to its advantage.But work as
When number of targets or clutter number increase, joint event can be in exponential increase, so as to cause calculation amount abruptly increase, or even causes and " combines quick-fried
It is fried " phenomenon, it is unable to satisfy requirement of the target following to real-time.In order to solve this problem, the data of various feature auxiliary are closed
Connection algorithm is suggested in succession, by other features except target position information or during parameter is introduced into data correlation, to
The higher interrelating effect of accuracy is obtained, to improve the precision and real-time of target following.Herein in conjunction with hesitation fuzzy set ash
Change thought and propose a kind of new feature auxiliary data association algorithm, focus on solving multiple short distances are done under complex electromagnetic environment it is small
The problem of angle cross non-linear moving target is tracked.
Solve the problems, such as the multiple target tracking under complex electromagnetic environment, present invention is generally directed to property in multiple target target
When doing the nonlinear motion of short distance low-angle intersection, the precision and real-time of target following are still ensured that.
The present invention realizes that process is as follows: in each sampling period, measuring to all positions obtained out of target airspace
Information carries out availability deciding, that is, judges whether it falls into target following door.For falling into multiple target following doors zone of intersection
Measurement in domain calculates associated allocation in conjunction with hesitation fuzzy set thought then using other characteristic parameter information corresponding thereto
Probability corrects the association probability formula of traditional data association algorithm by the allocation probability.It is general according to final amendment association
Rate estimates the state vector of each target, to realize the update of a plurality of targetpath.Specific step is as follows:
Step 1:
All position measurement information and other characteristic parameter information are obtained from target airspace in a certain sampling instant, existing
Have and set fan-shaped tracking gate around the predicted value of each targetpath, the validity of all measurements is determined with this, and have from all
Effect screens the measurement fallen into multiple target following doors intersection region, i.e., public measurement in measuring.
Step 2:
For each public measurement, determining may relative target.In conjunction with characteristic parameter corresponding with public measurement
Information seeks the degree of association of the public measurement and each target under various possible radar signal types, is re-introduced into hesitation fuzzy set
Thought assess a possibility that public measurement is related to each target size, and based on grey possibility degree calculate public measurement relative to
The allocation probability of each target.
Step 3:
Each target is distributed into all public measurements according to the probability acquired in step 2, multiple target tracking problem is converted
For the monotrack problem for correcting association probability, dbjective state is estimated using Probabilistic Data Association Algorithm.
Step 4:
Target following error co-variance matrix and each target signature parameter center are updated, and so on iteration is to realize pair
The real-time tracking of multiple target.
Compared to probabilistic data association, the feature auxiliary data based on gray relative association the methods of, using be based on hesitation mould
The tracking accuracy of Kalman filter tracking algorithm is improved significantly after the feature auxiliary data correlating method of paste collection；And it compares
Joint probability data association method, this method is not under the premise of tracking accuracy is decreased obviously, hence it is evident that improves multiple target
The real-time of tracking.
(1) the method for the present invention improves multiple target tracking precision, it can be achieved that enemy by the accuracy of raising data correlation
Target carries out more real-time effective tracking.
(2) there is the method for the present invention good expansibility can use the party for a variety of different track algorithms
Method carries out data correlation, and it can be made to filter convergence rate and be improved significantly.
(3) the method for the present invention is adapted to a variety of Nonlinear Parameter type of sports, and robustness is higher.
Fig. 1 is the feature auxiliary mark tracking data correlating method block diagram based on hesitation fuzzy set.
Present invention is further described in detail with specific embodiment with reference to the accompanying drawing.
The present invention provide it is a kind of interfere and the intensive complex electromagnetic environment of clutter under to the non-linear of multiple UNKNOWN TYPEs
When target is tracked, the various features information except binding site carries out tracking and associated new method.And introduce hesitation mould
Collection ashing thought is pasted, decision is carried out to the degree of association size of public measurement and each target, it is accurate in the hope of the more reliable association of acquisition
Rate and higher tracking accuracy and real-time.
The present invention is further described in detail below:
Step 1:
1) all position measurement information Z are obtained from target airspace in sampling instant tt=[ρt,θt] and other characteristic parameters
Information Ct=[PRIt,PWt,CFt], this method chooses pulse repetition period PRI, tri- characteristic parameters of pulse width PW, carrier frequency CF
Auxiliary association is carried out, other characteristic informations can also be chosen as auxiliary parameter to adapt to different application scene.
2) in existing each targetpath predicted valueAround set fan-shaped tracking gate, sentenced with this
The validity of fixed all measurements has for measuring j
gate_ρiWith gate_ θiTwo tracking gate threshold value components of the respectively target i under polar coordinates.
3) confirmation matrix is establishedWhereinIt is binary variable,Indicate that measuring j falls into mesh
In the tracking gate for marking t；Indicate that measuring j does not fall in the tracking gate of target t.N is existing targetpath number, mtFor
Effective measurement number that this sampling instant obtains.The situation that tracking gate intersecting area is fallen into for measurement means that the measurement may
Derived from multiple targets.
4) above-mentioned confirmation matrix is analyzed, screening falls into multiple target following doors zone of intersection from all effective measurements
Measurement in domain, i.e., public measurement.
Step 2:
1) assume to share N in t momentsA public measurement, and for public measurement js, fall intoA target following door
Intersection region, i.e. js=1,2 ..., Ns,Below for public measurement jsWith target isIt is introduced, base
In location informationWith corresponding characteristic parameter informationIt seeks
The degree of association of the public measurement and each target under various possible radar signal types.
(a) based on the degree of association of state quantity measurement
Public measurement jsWith each target prediction valueStatistical distance use respectivelyTable
Show, wherein with i-thsThe state of a target measures statistical distance are as follows:
Wherein,To filter new breath vector,For target isFiltering residuals covariance square
Battle array.The degree of association based on state quantity measurement is denoted as msta, it indicates to measure the influence with the statistical distance of state to correlation degree, I
Define i-thsA target is relative to measurement jsThe degree of association are as follows:
ω1For location information weight distribution.
(b) it is based on the degree of association of pulse repetition period (PRI)
The degree of association based on PRI is denoted as mPRI, it indicates influence of the PRI observation difference to correlation degree, to different PRI types
Target, mPRIDefinition be different.
To the target of repetition fixed type, public measurement jsWith i-thsThe m of a targetPRI,1(js,is) is defined as:
Wherein,AndIt is by target isSystem noise and error in measurement institute really
Fixed PRI measures tolerance, ω2For PRI information weight distribution.
To the target of the irregular type of repetition two, if target isTwo irregular repetition differences bePublic measurement jsWith
isThe m of a targetPRI,2(js,is) is defined as:
(c) it is based on the degree of association of pulse width (PW)
The degree of association based on pulse width is denoted as mPW, it indicates influence of the pulsewidth PW observation difference to correlation degree, is defined as:
Wherein,AndIt is by target isSystem noise and error in measurement determined
PW measure tolerance, ω3For PW information weight distribution.
(d) it is based on the degree of association of working frequency (CF)
The degree of association based on working frequency is denoted as mCF, for the target of different frequency type, the definition of the frequency association factor
It is different.
To the target of fixed frequency type, public measurement jsWith i-thsThe m of a targetCF,1(js,is) is defined as:
Wherein,AndIt is by target isSystem noise and error in measurement determined by
CF measures tolerance, ω4For CF information weight distribution.
To the target of frequency agility type, public measurement jsWith i-thsThe m of a targetCF,2(js,is) is defined as:
WhereinIt is target isFrequency agility range.
To the target of two diversity type of frequency, if the difference of two diversity frequencies is denoted asThen public measurement jsWith i-thsIt is a
The m of targetCF,3(js,is) is defined as:
2) for public measurement js, introduce hesitation fuzzy set thought, under the different type that front is acquired, each characteristic parameter
The measurement of characterization is arranged with each target association level data, establishes hesitation fuzzy decision matrix:
Due toInterior each data are profit evaluation model, therefore programming decision-making matrix
3) programming decision-making matrix is soughtIn each fuzzy member core and upper and lower deviation value.
If the length for the fuzzy member h that hesitates is L, hi(i=1,2 ..., L) regards discrete grey fuzzy number as, claims
For the core for the fuzzy member h that hesitates；Referred to as upper deviation value；Referred to as descend deviation value.
For the citation form of core and deviation value, incited somebody to action according to defined aboveBe converted to the decision matrix of core Yu upper and lower deviation value
4) reference sequence is formedWhereinForRespectively
The maximum element of column center value, k ∈ [sta, PRI, PW, CF].
According to reference sequence and decision matrixObtain grey incidence coefficientAnd grey relational grade
Indicate the distance of two hesitation Fuzzy Grey numbers.ρ is resolution ratio, generally takes ρ=0.5.
5) weighted value Ω=[ω (1), ω (2), ω (3), ω (4)] of each characteristic attribute is sought, wherein
6) according to weighted value, weighted comprehensive attribute value is determined:
Do not subtract axiom according to gray scale it is found thatGray scale be equal to the gray scale maximum value for participating in all grey numbers of operation, and
It will not reduce or reduce.
7) will likely the corresponding weighted comprehensive attribute value of associated objects from core and upper and lower deviation value form be reduced to conventional ash
Number form formula:
Wherein,
To it is all may be with public measurement jsAssociated target is compared and seeks respective possibility degree:
It enablesThen ordering vector isAnd
8) public measurement and the allocation probability matrix Y to each target are sought, wherein
Wherein, α is the proportion adjustment factor.
Step 3:
1) by taking target i as an example, it is known that the pervious effective measurement of t moment collects Zt-1And the m of t momenttA effective measurement is all
In the case where clutter, Z can be obtainedtJoint probability density function be
For j=1,2 ..., mtAny case, ZtJoint probability density function be
Wherein
If assuming, false-alarm measures number MfObedience parameter is λ VtPoisson distribution, can be calculated
Wherein,
2) association probability measured in each target and tracking gate is modified using the allocation probability acquired in step 2,
Correction matrix FcFor to allocation probability matrixExtension:
Wherein, jsAnd isExist to measure j and target iIn corresponding position coordinates.Revised association probability are as follows:
Mi' it is all measurement numbers fallen into target i tracking gate.
3) each dbjective state is estimated with Kalman filtering, updates filtering equations are as follows:
Wherein,Represent the new breath of combination.Kt,iFor filtering gain,For filter forecasting
Value.
Step 4:
1) target following error co-variance matrix is updated:
Wherein,Ht,i
For the calculation matrix of t moment target i.
2) each target component center is updated:
Continue reciprocal iteration to realize the real-time tracking to multiple target.
1. a kind of specific aim solves the problems, such as the multiple target tracking data correlation under complex electromagnetic environment, it is based on hesitation fuzzy set
The data correlation new method of each possible associated objects is distributed in public measurement by thought, and key step is as follows:
Step 1: all position measurement information and other characteristic parameters letter are obtained from target airspace in a certain sampling instant
Breath sets fan-shaped tracking gate around the predicted value of existing each targetpath, determines the validity of all measurements with this, and from
Screening falls into the measurement in multiple target following doors intersection region, i.e., public measurement in all effective measurements.
Step 2: being directed to each public measurement, determining may relevant target to it.In conjunction with spy corresponding with public measurement
Parameter information is levied, seeks the public measurement degree of association associated with each target under various possible radar signal types, then draw
The thought for entering the fuzzy set that hesitates assesses a possibility that public measurement is related to each target, and calculates public amount based on grey possibility degree
Survey the allocation probability relative to each target.
Step 3: each target is distributed into all public measurements according to the probability acquired in step 2, multiple target tracking is asked
Topic is converted into the monotrack problem of amendment association probability, is estimated using Probabilistic Data Association Algorithm dbjective state.
Step 4: update target following error co-variance matrix and each target signature parameter center, and so on iteration with
Realize the real-time tracking to multiple target.
2. a kind of specific aim according to claim 1 solves the multiple target tracking data correlation under complex electromagnetic environment
The data correlation new method of each possible associated objects, feature are distributed in public measurement based on hesitation fuzzy set thought by problem
It is to combine characteristic parameter information corresponding with public measurement, public measurement association associated with each target in step 2
Degree, the thought for being re-introduced into hesitation fuzzy set calculate allocation probability of the public measurement relative to each target, are specifically described as follows:
1) assume to share N in t momentsA public measurement, and for public measurement js, fall intoA target following door
Intersection region, i.e. js=1,2 ..., Ns,Below for public measurement jsWith target isIt is introduced, base
In location informationCharacteristic parameter information corresponding with public measurementIt is associated with each target to seek the public measurement under various possible radar signal types
The degree of association.
2) for public measurement js, introduce hesitation fuzzy set thought, under the different type that front is acquired, each characteristic parameter
The measurement of characterization is arranged with each target association level data, establishes hesitation fuzzy decision matrix
Due toInterior each data are profit evaluation model, therefore programming decision-making matrix
3) programming decision-making matrix is soughtIn each fuzzy member core and upper and lower deviation value.
If the length for the fuzzy member h that hesitates is L, hi(i=1,2 ..., L) regards discrete grey fuzzy number as, claims
For the core for the fuzzy member h that hesitates；Referred to as upper deviation value；Referred to as descend deviation value.
For the citation form of core and deviation value, incited somebody to action according to defined aboveBe converted to the decision matrix of core Yu upper and lower deviation value
4) reference sequence is formedWhereinForRespectively
The maximum element of column center value, k ∈ [sta, PRI, PW, CF].
According to reference sequence and decision matrixObtain grey incidence coefficientAnd grey relational grade
Indicate the distance of two hesitation Fuzzy Grey numbers.ρ is resolution ratio, generally takes ρ=0.5.
5) weighted value Ω=[ω (1), ω (2), ω (3), ω (4)] of each characteristic attribute is sought, wherein
6) according to weighted value, weighted comprehensive attribute value is determined:
Do not subtract axiom according to gray scale it is found thatGray scale be equal to the gray scale maximum value for participating in all grey numbers of operation, and
It will not reduce or reduce.
7) will likely the corresponding weighted comprehensive attribute value of associated objects from core and upper and lower deviation value form be reduced to conventional ash
Number form formula:
Wherein,
To it is all may be with public measurement jsAssociated target is compared and seeks respective possibility degree:
It enablesThen ordering vector isAnd
8) public measurement and the allocation probability matrix Y to each target are sought, wherein
Wherein, α is the proportion adjustment factor.
3. a kind of specific aim according to claim 1 solves the multiple target tracking data correlation under complex electromagnetic environment
The data correlation new method of each possible associated objects, feature are distributed in public measurement based on hesitation fuzzy set thought by problem
It is in step 3 that each target is distributed in all public measurements using the probability acquired in step 2, multiple target tracking problem is turned
The monotrack problem for turning to amendment association probability, estimates dbjective state using Probabilistic Data Association Algorithm.
4. a kind of specific aim according to claim 1 solves the multiple target tracking data correlation under complex electromagnetic environment
The data correlation new method of each possible associated objects, feature are distributed in public measurement based on hesitation fuzzy set thought by problem
It is to update target following error co-variance matrix and each target signature parameter center in step 4.
The present invention provide it is a kind of interfere and the intensive complex electromagnetic environment of clutter under, to the non-linear of multiple UNKNOWN TYPEs
When target is tracked, the various features information except combining position information carries out tracking and associated new method.It is dry in clutter
It disturbs under serious situation, the data correlation problem during multiple target tracking seems sufficiently complex.Especially when multiple targets are close
Distance do low-angle intersection nonlinear motion when, the accuracy of data correlation directly affect target tracking accuracy and in real time
Property.Probabilistic Data Association Algorithm and Joint Probabilistic Data Association algorithm are considered as optimal single goal and Multi-target Data
Association algorithm, but more complicated Nonlinear Multiobjective track demand is faced, traditional association method still is apparent not enough.In order to solve this
One problem, the data association algorithm based on feature auxiliary come into being.The present invention is ashed thought by introducing hesitation fuzzy set, mentions
A kind of feature assists multiple target tracking data correlation new method out, determines to public measurement and the degree of association size of each target
Plan, in the hope of obtaining more reliable association accuracy rate and higher tracking accuracy and real-time.
Claims (2)
1. a kind of method that specific aim solves the problems, such as the multiple target tracking data correlation under complex electromagnetic environment, which is characterized in that
The following steps are included:
(1) all position measurement information and other characteristic parameter information are obtained from target airspace in a certain sampling instant, existing
Have and set fan-shaped tracking gate around the predicted value of each targetpath, the validity of all measurements is determined with this, and have from all
Effect screens the measurement fallen into multiple target following doors intersection region, i.e., public measurement in measuring；
(2) it is directed to each public measurement, determination target relevant to its possibility is believed in conjunction with characteristic parameter corresponding with public measurement
Breath, seeks the public measurement degree of association associated with each target under various possible radar signal types, is re-introduced into hesitation mould
The thought of paste collection assesses a possibility that public measurement is related to each target, and based on the public measurement of grey possibility degree calculating relative to
The allocation probability of each target；
(3) each target is distributed into all public measurements according to the probability acquired in step 2, converts multiple target tracking problem to
The monotrack problem for correcting association probability, estimates dbjective state using Probabilistic Data Association Algorithm；
(4) target following error co-variance matrix and each target signature parameter center are updated, and so on iteration is to realize pair
The real-time tracking of multiple target.
2. determining may be related to it the method according to claim 1, wherein described be directed to each public measurement
Target it is public to seek under various possible radar signal types this in conjunction with characteristic parameter information corresponding with public measurement
Associated with each target degree of association is measured, being re-introduced into the thought of hesitation fuzzy set, to assess the public measurement relevant to each target
Possibility, and allocation probability of the public measurement relative to each target is calculated based on grey possibility degree, comprising:
(2.1) assume to share N in t momentsA public measurement, and for public measurement js, fall intoA target following door
Intersection region, i.e. js=1,2 ..., Ns,Below for public measurement jsWith target isIt is introduced, is based on
Location informationCharacteristic parameter information corresponding with public measurement
Seek the public measurement degree of association associated with each target under various possible radar signal types；
(2.2) for public measurement js, introduce hesitation fuzzy set thought, under the different type that front is acquired, each characteristic parameter table
The measurement of sign is arranged with each target association level data, establishes hesitation fuzzy decision matrix
Due toInterior each data are profit evaluation model, therefore programming decision-making matrix
(2.3) programming decision-making matrix is soughtIn each fuzzy member core and upper and lower deviation value；
If the length for the fuzzy member h that hesitates is L, hi(i=1,2 ..., L) regards discrete grey fuzzy number as, claimsFor still
Henan obscures the core of member h；Referred to as upper deviation value；Deviation value is referred to as descended,For core with
The citation form of deviation value is incited somebody to action according to defined aboveBe converted to the decision matrix of core Yu upper and lower deviation value
(2.4) reference sequence is formedWhereinForEach column
The maximum element of center value, k ∈ [sta, PRI, PW, CF]；
According to reference sequence and decision matrixObtain grey incidence coefficientAnd grey relational grade
Indicate the distance of two hesitation Fuzzy Grey numbers, ρ is resolution ratio, ρ=0.5；
(2.5) weighted value Ω=[ω (1), ω (2), ω (3), ω (4)] of each characteristic attribute is sought, wherein
(2.6) according to weighted value, weighted comprehensive attribute value is determined:
Do not subtract axiom according to gray scale it is found thatGray scale be equal to the gray scale maximum value for participating in all grey numbers of operation, without
It reduces or reduces；
(2.7) will likely the corresponding weighted comprehensive attribute value of associated objects from core and upper and lower deviation value form be reduced to conventional grey number
Form:
Wherein,
To it is all may be with public measurement jsAssociated target is compared and seeks respective possibility degree:
It enablesThen ordering vector isAnd
(2.8) public measurement and the allocation probability matrix Y to each target are sought, wherein
Wherein, α is the proportion adjustment factor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811017254.9A CN109298413A (en) | 2018-09-01 | 2018-09-01 | A kind of method that specific aim solves the problems, such as the multiple target tracking data correlation under complex electromagnetic environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811017254.9A CN109298413A (en) | 2018-09-01 | 2018-09-01 | A kind of method that specific aim solves the problems, such as the multiple target tracking data correlation under complex electromagnetic environment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109298413A true CN109298413A (en) | 2019-02-01 |
Family
ID=65165967
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811017254.9A Pending CN109298413A (en) | 2018-09-01 | 2018-09-01 | A kind of method that specific aim solves the problems, such as the multiple target tracking data correlation under complex electromagnetic environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109298413A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110133636A (en) * | 2019-05-21 | 2019-08-16 | 中国人民解放军海军航空大学 | A kind of robust Data Association based on the region degree of correlation |
| CN110244294A (en) * | 2019-06-24 | 2019-09-17 | 中国人民解放军空军工程大学航空机务士官学校 | A kind of correlating method of the metric data of multisensor |
| CN112684454A (en) * | 2020-12-04 | 2021-04-20 | 中国船舶重工集团公司第七一五研究所 | Track cross target association method based on sub-frequency bands |
| WO2021217491A1 (en) * | 2020-04-29 | 2021-11-04 | 华为技术有限公司 | Data association method and device |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH10160830A (en) * | 1996-11-29 | 1998-06-19 | Mitsubishi Electric Corp | Method and device for target tracking |
| JP2003149330A (en) * | 2001-11-07 | 2003-05-21 | Mitsubishi Electric Corp | Tracking device |
| CN105137418A (en) * | 2015-07-28 | 2015-12-09 | 中国人民解放军海军航空工程学院 | Multi-object tracking and data interconnection method based on whole neighborhood fuzzy clustering |
| CN106872955A (en) * | 2017-01-24 | 2017-06-20 | 西安电子科技大学 | Radar Multi Target tracking optimization method based on Joint Probabilistic Data Association algorithm |
| CN107576959A (en) * | 2017-08-08 | 2018-01-12 | 电子科技大学 | Tracking before a kind of Gao Zhongying Radar Targets'Detection based on area maps ambiguity solution |
| CN108093213A (en) * | 2017-12-13 | 2018-05-29 | 中国人民解放军陆军工程大学 | A kind of target trajectory Fuzzy Data Fusion method based on video monitoring |
| KR20180080004A (en) * | 2017-01-03 | 2018-07-11 | 국방과학연구소 | Target tracking method using feature information in target occlusion condition |
-
2018
- 2018-09-01 CN CN201811017254.9A patent/CN109298413A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH10160830A (en) * | 1996-11-29 | 1998-06-19 | Mitsubishi Electric Corp | Method and device for target tracking |
| JP2003149330A (en) * | 2001-11-07 | 2003-05-21 | Mitsubishi Electric Corp | Tracking device |
| CN105137418A (en) * | 2015-07-28 | 2015-12-09 | 中国人民解放军海军航空工程学院 | Multi-object tracking and data interconnection method based on whole neighborhood fuzzy clustering |
| KR20180080004A (en) * | 2017-01-03 | 2018-07-11 | 국방과학연구소 | Target tracking method using feature information in target occlusion condition |
| CN106872955A (en) * | 2017-01-24 | 2017-06-20 | 西安电子科技大学 | Radar Multi Target tracking optimization method based on Joint Probabilistic Data Association algorithm |
| CN107576959A (en) * | 2017-08-08 | 2018-01-12 | 电子科技大学 | Tracking before a kind of Gao Zhongying Radar Targets'Detection based on area maps ambiguity solution |
| CN108093213A (en) * | 2017-12-13 | 2018-05-29 | 中国人民解放军陆军工程大学 | A kind of target trajectory Fuzzy Data Fusion method based on video monitoring |
Non-Patent Citations (1)
| Title |
|---|
| 朱嘉颖等: "基于犹豫模糊集的特征辅助数据关联算法", 《制导与引信》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110133636A (en) * | 2019-05-21 | 2019-08-16 | 中国人民解放军海军航空大学 | A kind of robust Data Association based on the region degree of correlation |
| CN110244294A (en) * | 2019-06-24 | 2019-09-17 | 中国人民解放军空军工程大学航空机务士官学校 | A kind of correlating method of the metric data of multisensor |
| WO2021217491A1 (en) * | 2020-04-29 | 2021-11-04 | 华为技术有限公司 | Data association method and device |
| CN112684454A (en) * | 2020-12-04 | 2021-04-20 | 中国船舶重工集团公司第七一五研究所 | Track cross target association method based on sub-frequency bands |
| CN112684454B (en) * | 2020-12-04 | 2022-12-06 | 中国船舶重工集团公司第七一五研究所 | Track cross target association method based on sub-frequency bands |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109298413A (en) | A kind of method that specific aim solves the problems, such as the multiple target tracking data correlation under complex electromagnetic environment | |
| CN106443622B (en) | A kind of distributed object tracking based on improvement joint probability data association | |
| CN106872955B (en) | Radar multi-target tracking optimization method based on joint probability data association algorithm | |
| CN104155650B (en) | A kind of method for tracking target based on entropy weight method point mark quality evaluation | |
| CN108802722B (en) | It is a kind of based on tracking before the Faint target detection virtually composed | |
| CN105137418B (en) | Multiple target tracking and data interconnection method based on complete adjacent fuzzy clustering | |
| CN103472445B (en) | Detecting tracking integrated method for multi-target scene | |
| CN106980114A (en) | Target Track of Passive Radar method | |
| CN111007495A (en) | Target track optimization method based on double-fusion maximum entropy fuzzy clustering JPDA | |
| CN101614817A (en) | A kind of multi-object tracking method based on ground moving target indication radar system | |
| KR20110112829A (en) | Method for position estimation using generalized error distributions | |
| CN109996175A (en) | Indoor locating system and method | |
| CN107861123A (en) | A kind of through-wall radar is under complex environment to the method for multiple mobile object real-time tracking | |
| CN110673130B (en) | Moving target track tracking method based on track association | |
| CN105842687A (en) | Detection tracking integrated method based on RCS prediction information | |
| CN109946694A (en) | Circumference SAR multi-object tracking method based on stochastic finite collection | |
| CN106526549A (en) | False target identification method with combination of two-coordinate radar and three-coordinate radar statistics | |
| CN110308442B (en) | GM-PHD target tracking method of phased array radar in strong clutter environment | |
| CN106680783A (en) | Method for withstanding false targets on basis of station's position error fusion algorithm | |
| CN110244294A (en) | A kind of correlating method of the metric data of multisensor | |
| CN109613483A (en) | A kind of multi-target traces initial mode based on Hough transform | |
| CN108732564A (en) | A kind of dual radars amendment sequence Gauss mixing probability hypothesis density filtering method | |
| CN107102293A (en) | The passive co-located method of unknown clutter estimated based on sliding window integral density | |
| CN110488273A (en) | A kind of vehicle tracking detection method and device based on radar | |
| CN109239704A (en) | A kind of adaptively sampled method based on Sequential filter interactive multi-model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190201 |