CN113075648B - Clustering and filtering method for unmanned cluster target positioning information - Google Patents
Clustering and filtering method for unmanned cluster target positioning information Download PDFInfo
- Publication number
- CN113075648B CN113075648B CN202110293752.1A CN202110293752A CN113075648B CN 113075648 B CN113075648 B CN 113075648B CN 202110293752 A CN202110293752 A CN 202110293752A CN 113075648 B CN113075648 B CN 113075648B
- Authority
- CN
- China
- Prior art keywords
- clustering
- unmanned
- target
- positioning
- time
- 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
Links
- 238000001914 filtration Methods 0.000 title claims abstract description 22
- 238000000034 method Methods 0.000 title claims abstract description 20
- 230000004927 fusion Effects 0.000 claims abstract description 27
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 15
- 238000001514 detection method Methods 0.000 claims description 6
- 230000010485 coping Effects 0.000 abstract 1
- 230000000694 effects Effects 0.000 description 6
- 238000002474 experimental method Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 2
- 230000005484 gravity Effects 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000002123 temporal effect Effects 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
- G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
- G01S13/06—Systems determining position data of a target
-
- 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
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
- G01S7/414—Discriminating targets with respect to background clutter
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computer Networks & Wireless Communication (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Probability & Statistics with Applications (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a clustering and filtering method of unmanned cluster target positioning information, which belongs to the technical field of multi-source data fusion and association, and is used for unifying time references of positioning data of unmanned equipment in an unmanned cluster; after unifying time references of unmanned equipment is completed, clustering and filtering are carried out on target positioning points output by different unmanned equipment, and clustering rings are formed while false alarms are removed, so that clustering of different positioning values of the same target is completed; and using cluster center distance weighting to complete fusion of a plurality of positioning values so as to realize multi-source data space registration. The invention clusters the positions of unmanned equipment and the positions positioned to the target in the unmanned cluster by means of a mean shift clustering method. By analyzing the characteristics of false alarms and the error distribution of each sensor, targeted adjustment and algorithm improvement are performed on the basis of mean shift clustering. A clustering method capable of eliminating clutter, continuously tracking targets and coping with the problem of multi-object meeting is formed.
Description
Technical Field
The invention belongs to the technical field of multi-source data fusion and association, and particularly relates to a clustering and filtering method for unmanned cluster target positioning information.
Background
In the unmanned cluster comprehensive command center, the unmanned cluster autonomous positioning target, patrol path setting, recognition, tracking, interception, attack target and other contents need to be completed. Wherein information of the position, time, speed, etc. of a plurality of unmanned equipments positioned to the same target is deviated. In practical experiments, part of positioning information is false alarm under the influence of sea clutter. These factors result in that in all the target positioning information received by the command center, a situation that a plurality of positioning points correspond to the same target and a plurality of positioning points are false alarms often occurs. Judging whether a plurality of positioning points are the same target, distinguishing whether the positioning points are false alarms, and determining the corresponding relation between each positioning point and an enemy target or unmanned equipment on the my side is a difficult point, so that great challenges are generated for sensing and reading battlefield situations by a command center.
Disclosure of Invention
Aiming at the defects or improvement demands of the prior art, the invention provides a clustering and filtering method for target positioning information of an unmanned cluster, and solves the problem of clustering and filtering of multiple unmanned targets on multiple positioning information of the same target in the unmanned cluster.
In order to achieve the above object, the present invention provides a method for clustering and filtering unmanned cluster target positioning information, including:
unifying time references of positioning data of all unmanned equipment in the unmanned cluster;
After unifying time references of unmanned equipment is completed, clustering and filtering are carried out on target positioning points output by different unmanned equipment, and clustering rings are formed while false alarms are removed, so that clustering of different positioning values of the same target is completed;
And using cluster center distance weighting to complete fusion of a plurality of positioning values so as to realize multi-source data space registration.
In some optional embodiments, the time reference unifying the positioning data of each unmanned equipment in the unmanned cluster includes:
Receiving data messages of each unmanned equipment in an unmanned cluster, respectively obtaining time stamp information of each unmanned equipment, and obtaining the local time of a fusion program as real time;
Continuously acquiring data for a plurality of seconds to obtain data of each acquired sample, and calculating the time stamp and the real time difference value of each unmanned equipment to obtain a plurality of groups of time differences;
And calculating the mean value and variance of all groups of time differences, subtracting the mean value from the message time of each unmanned equipment, and completing the time stamp calibration of each unmanned equipment.
In some optional embodiments, after the time references of the unmanned apparatuses are unified, clustering and filtering are performed on target positioning points output by different unmanned apparatuses, forming a clustering ring while removing false alarms, and completing clustering of different positioning values of the same target, including:
based on the last clustering circle, using mean shift clustering, further clustering the clustering circles with the distance smaller than r by using k-means to obtain new positions of the original clustering circles;
removing the locating points already contained in the clustering ring, and performing de-duplication and merging on the rest locating points again by using a mean shift clustering algorithm;
and if the distance between the obtained clustering ring and the original clustering ring is larger than R, judging the clustering ring as a new target, otherwise judging the clustering ring as clutter, and forming a fuzzy clustering ring, wherein R is the radius of the target which can be stably detected by the unmanned radar, and R is the target detection error range.
In some alternative embodiments, the fusing of the plurality of location values using cluster center distance weighting to achieve multi-source data spatial registration includes:
Calculating the weight of each positioning point according to the distance between each positioning point and the center of the clustering ring;
and respectively carrying out weighted summation on the navigational speed, the position and the time stamp of the locating point by using the weight value, and calculating the position, the time stamp and the speed of the fused target.
In some alternative embodiments, the method comprisesCalculating the confidence degree w j of each positioning by utilizing the distances dis (P j, O) of the n unmanned equipment to the positioning position P j and the clustering center position O of the same target;
From the following components Obtaining a Position After fusion of the fused target;
From the following components Obtaining a Time stamp Time After fusion of the fused target;
From the following components The velocity V After fusion of the fused target is obtained.
In general, the above technical solutions conceived by the present invention, compared with the prior art, enable the following beneficial effects to be obtained:
even if the network of the unmanned cluster fails, the radar is abnormal, the targets meet, separate and parallel, the unmanned cluster can stably operate. And clustering and filtering as accurately as possible under the condition of error and error data. The unmanned cluster command center and the cluster control center do not need to care about the equipment state, and direct command is directly conducted on the clustering result of the invention. The high-level algorithm is not required to be oriented to unstable and inaccurate sensing data, and is directly oriented to relatively stable and accurate clustering data. Providing more accurate and stable data for subsequent fusion algorithms.
Drawings
FIG. 1 is a schematic diagram of a data input data flow according to an embodiment of the present invention;
Fig. 2 is a schematic flow chart of a clustering and filtering method for unmanned cluster target positioning information provided by an embodiment of the invention;
fig. 3 is a time registration effect diagram provided by the embodiment of the present invention, where (a) represents message time stamps sent by unmanned equipment nos. 1026 and 2049 within 10s, and (b) represents a time registration effect diagram;
FIG. 4 is a graph of clustered effects of the present invention when multiple vessels meet;
fig. 5 is a spatial error experimental diagram provided in an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention. In addition, the technical features of the embodiments of the present invention described below may be combined with each other as long as they do not collide with each other.
After the target positioning information sent by all unmanned equipment in the unmanned cluster is collected, the information of the whole unmanned cluster is integrated and clustered to obtain clustered information, so that the accurate target situation generated through fusion can be conveniently provided for the command center, and support is provided for the command center to perform battlefield understanding and combat command. The invention plays a role in adapting the unmanned cluster comprehensive control center, so that the information of the equipment layer is transparent to the high-level algorithm, and the high-level algorithm can directly and accurately and stably acquire data through the invention without paying attention to the equipment state. The method solves the problems of non-uniform target data time stamp, large positioning error, multi-target track intersection, unstable target data, inaccurate target navigation speed measurement and the like when target positioning data from a plurality of unmanned equipment are fused in an unmanned cluster, and completes target clustering and filtering, thereby providing a data basis for realizing accurate target fusion.
According to the invention, the problem of large time and space system errors of multiple positioning sources is solved through space-time registration of target positioning data of a plurality of unmanned equipment of an unmanned cluster. And obtaining the corresponding relation between the targets detected by each unmanned equipment and the clustered targets through fuzzy clustering association of unmanned cluster multi-domain positioning data. Clustering effects, including temporal registration, clustering and filtering, and spatial registration and fusion, are enhanced by filtering clutter present in the unmanned clusters, as shown in fig. 2.
(1) Time registration
As shown in fig. 1, the time stamp of each piece of unmanned equipment transmitting the target positioning information is based on its own time coordinate system, and the time references of the unmanned equipment which are not uniform in time are inconsistent, so that the time stamps in the target positioning information detected by different pieces of unmanned equipment at the same time are inconsistent, and fusion errors are further caused, and even the valid positioning values are discarded.
Assuming that the procedure of the unmanned equipment can guarantee 200ms of sending a message according to the protocol, the following formula can be obtained.
T Platform ＝T True and true +T Time coordinate system difference +T Network delay
E(T Platform )＝E(T True and true )+E(T Time coordinate system difference )+E(T Network delay )
T Platform denotes a time when the present unmanned equipment transmits, T True and true denotes a real time, T Time coordinate system difference denotes an offset of the present unmanned equipment time from the real time, and T Network delay is a network delay time. The purpose of the time registration is to solve for E (T Platform ), the real time, the time coordinate system offset and the network delay time are independent.
In order to solve E (T Platform ), the invention agrees that the fusion program local time is real time, so that E (T True and true ) can be obtained in the experiment, and the problem is converted into solving E (T Time coordinate system difference )+E(T Network delay ).
To obtain E (T Time coordinate system difference )+E(T Network delay ), after 10s of data (including 50 sample points) were collected, 50 sets of T Platform -T True and true were obtained, assuming that these 50 sets of data were designated Y, and Y could be used to obtain E (T Time coordinate system difference )+E(T Network delay ). Since both the present unmanned equipment time and the local time are stable, T Time coordinate system difference is a constant. And network delay can be considered as a normally distributed random variable.
According to the big theorem, the mean and variance of Y are calculated, which should be equal to E (T Time coordinate system difference )+E(T Network delay ) and var (T Time coordinate system difference +T Network delay ). Since T Time coordinate system difference is a constant, the mean of Y is T Time coordinate system difference +E(T Network delay and the variance of Y is var (T Network delay ).
The average value of Y is thus taken as E (T Time coordinate system difference )+E(T Network delay ). And the average value of the message time-Y of each unmanned equipment is obtained, so that the calibrated time can be obtained. The error after calibration is var (Y), and the error before calibration is var (Y) +E (Y). The accuracy after time registration is expressed as follows:
The invention collects the time stamps of the messages sent by unmanned equipment 1026 and 2049, respectively, with the time length of 10s, as shown in figure 3. Where the y-axis is the time axis, the x-axis is also the time axis, the units are ms, the middle line is the standard time line, and it comes out according to the y=x point. As shown in fig. 3 (a), the lowermost line and the uppermost line represent message timestamps sent by unmanned equipment nos. 1026 and 2049 within 10s, respectively. Due to the problem of inconsistent time coordinate system, the time of 2049 unmanned equipment is advanced to the real time, and 1026 unmanned equipment is delayed to the real time.
As a result of the experiment, as in fig. 3 (b), the uppermost line represents the time line before adjustment, and the two lowermost coincident lines are the standard time line and the time line after adjustment.
It can be seen that there is a large deviation in the timeline of the unmanned equipment before calibration. After calibration, the adjusted line is basically overlapped with the standard time line, only one point is caused by variance fluctuation caused by a network, and the time registration accuracy after the improvement reaches 99.4%.
(2) Clustering and filtering
After the unification of the time references of the unmanned equipment is completed, clustering and filtering are carried out on target positioning points output by different unmanned equipment, and clustering rings are formed while false alarms are removed, so that clustering of different positioning values of the same target is completed.
(A) Mean shift clustering
The mean shift clustering method comprises the steps of firstly setting each locating point on a lake surface as a circle center, setting super parameters as radii, and converting locating point sets into circle sets; counting the positioning points contained in each circle respectively, sliding the center of each circle onto the center of gravity of the positioning point of the inner circle, counting the positioning points contained in each sliding circle again after one-time sliding is completed, and sliding each circle onto the center of gravity of the positioning point of the inner circle again until the number of the positioning points of the inner circle reaches the maximum value, wherein each sliding mode can enable each circle to smoothly reach the area with the maximum local density; and finally, de-duplicating and merging the sliding circle sets to obtain a plurality of clustering circles, wherein the density of locating points in each clustering circle is the local maximum, and the density maximum represents the maximum probability of existence of a real target.
The method has the advantages that the radius is set very close to the scene of the invention, and the invention can adjust the radius of the clustering ring through the target detection error range. Thus, the probability of fusing the same object into two objects or fusing two objects into one object can be effectively reduced.
(B) Improvement of mean shift clustering pertinence
According to the scene of the unmanned cluster, the mean shift clustering algorithm is adjusted in a targeted manner. Assuming that the radius of the stable detection and tracking range of the unmanned equipment radar is R and the detection error range is R, the main improvement of the invention is as follows:
the anchor point that suddenly appears within the unmanned equipment radius R is a false alarm and not a target.
The reason is that the enemy object must be detected and identified as a cluster ring during the approach from a distance to within the unmanned equipment distance R, and thus points that suddenly appear within R are unlikely to be objects.
(II) the radius r of the clustering ring is set to be the average positioning error of the unmanned equipment.
Since most of the positioning errors are within r, setting the radius of the clustered circle to r can effectively circle the real target. The setting of the r parameter can be obtained through statistical analysis.
(III) clustering loops do not need to be regenerated each time and can be inherited.
The clustering circle represents the target, and the moving distance of the target in one algorithm iteration is generally smaller than r, so that the clustering circle obtained by the previous round of calculation is continued to drift by taking the clustering circle as a reference, and the clustering circle can be slid to the position of the target in the round of calculation.
And (IV) selecting k-means clusters without de-duplication and merging when the distance between the cluster circles is smaller than r.
According to the traditional mean shift algorithm, the distance between clustering rings is less than r, and the deduplication and merging are needed. But the realistic meaning of a cluster ring in the context of the present invention is a target. A distance between two objects should not be classified as one object when it is less than r. The present invention cancels the merging operation. However, when the distance between two cluster circles is smaller than r, the situation that the two cluster circles overlap is easy to generate in the next iteration, so that the method uses a k-means clustering algorithm for the adjacent cluster circles, and the k-means has the advantages of good clustering effect and unknown clustering number because the number of the cluster circles is known, so that the k-means is selected to be the best choice for clustering when the cluster circles are close. Moreover, the initial position of the cluster ring is known, which also solves the problem of initializing the k-means cluster center.
As shown in fig. 4, the circles in the inner layer are clustering circles, and two points in the inner layer circles are two different clustering centers, which respectively correspond to the targets and unmanned equipment on the lake in the real environment. x is the target located, + is the unmanned equipment location information, and the outermost circle is the safe distance.
(3) Spatial registration and fusion
And a clustering center distance weighting algorithm is used for completing fusion of a plurality of positioning values, so that multi-source data space registration is realized.
Assuming that the error of each unmanned equipment to the target positioning is a random error, the probability of a real target at the cluster center O is maximum. The confidence w j for each location is thus calculated using the distances dis (P j, O) of the n unmanned equipment to the location position P j and the center position O of the same object. Finally, the Position of the fused target After fusion , the Time stamp Time After fusion , the velocity V After fusion , and the like can be calculated by weighting by the distance.
In a simulation experiment, a plurality of unmanned equipment are arranged to locate a target, and the accuracy of the clustered error relative to each unmanned equipment is checked.
The error is defined as above. Bias merged refers to the error after fusion and distance specifies the bit distance.
9 Experiments were performed. 2,4, 6, 8, 10, 12, 14, 16, 18 unmanned equipment were simulated to locate the same target, respectively. In each experiment, the error of each unmanned equipment after spatial registration is represented by a bar graph, resulting in fig. 5.
From the upper left corner to the lower right corner, the experimental results of 2, 4, 6, 8, 10, 12, 14, 16, 18 pieces of unmanned equipment were obtained. The leftmost column of each graph is the 2% error specified in the index, and the remaining columns are the errors of the different unmanned equipment after spatial registration. The completion index is equivalent to the remaining columns being lower than the leftmost column.
It can be seen that the error in positioning 2, 4 unmanned equipment to the same target exceeds 2%. The error is gradually reduced with the increase of unmanned equipment, and the effect of spatial registration is gradually improved. When 6 unmanned equipment is provided, the average error is about 1%; when 12 pieces of unmanned equipment are provided, the average error is about 0.8 percent; the average error was about 0.6% for 18 unmanned equipment.
Experiments show that the more the number of unmanned equipment is, the more accurate the spatial registration is, and in simulation experiments, the accuracy of more than 6 unmanned equipment can reach 98%.
Example 1
(1) Time registration
Receiving data messages of unmanned equipment A and unmanned equipment B, respectively obtaining time stamp information of the unmanned equipment A and the unmanned equipment B, and obtaining local time of a fusion program as real time; continuously collecting 10s of data, when each unmanned equipment sends a message at 200ms time intervals, 50 samples can be obtained after the time collection is finished, and the time stamp and the real time difference value of each unmanned equipment are calculated to obtain 50 groups of time differences; calculating the mean and variance of 50 groups of time differences; and subtracting the average value from the message time of each unmanned equipment to finish the time stamp calibration of each unmanned equipment.
(2) Clustering and filtering
Assuming that the radius of the unmanned equipment radar capable of stably detecting the target is R, the error range of detection is R, and utilizing the improved mean shift clustering algorithm to complete multi-source data clustering and filtering: 1) Using mean shift clustering based on the last clustering circle; 2) Further clustering the clustering rings with the distance less than r by using k-means; 3) Obtaining new positions of the original clustering rings from the steps 1) and 2); 4) Removing the anchor points already contained in the clustering ring, and using a traditional mean shift algorithm (requiring duplication removal and merging) again for the rest anchor points; 5) The obtained cluster ring is judged to be a new target if the distance from the original cluster ring is larger than R. Otherwise, the noise is determined.
(3) Spatial registration and fusion
1) Clustering the locating points by using an improved mean shift clustering algorithm to form a fuzzy clustering ring; 2) Calculating the weight of each positioning point according to the distance between each positioning point and the center of the clustering ring; 3) And respectively carrying out weighted summation on the navigational speed, the position and the time stamp of the locating point by using the weight value, and calculating the position, the time stamp and the speed of the fused target.
It should be noted that each step/component described in the present application may be split into more steps/components, or two or more steps/components or part of operations of the steps/components may be combined into new steps/components, according to the implementation needs, to achieve the object of the present application.
It will be readily appreciated by those skilled in the art that the foregoing description is merely a preferred embodiment of the invention and is not intended to limit the invention, but any modifications, equivalents, improvements or alternatives falling within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (4)
1. The clustering and filtering method for the unmanned cluster target positioning information is characterized by comprising the following steps of:
unifying time references of positioning data of all unmanned equipment in the unmanned cluster;
After unifying time references of unmanned equipment is completed, clustering and filtering are carried out on target positioning points output by different unmanned equipment, and clustering rings are formed while false alarms are removed, so that clustering of different positioning values of the same target is completed;
Fusion of a plurality of positioning values is completed by using cluster center distance weighting so as to realize multi-source data space registration;
After unifying time references of all unmanned equipment is completed, clustering and filtering are carried out on target positioning points output by different unmanned equipment, a clustering ring is formed when false alarms are removed, and clustering of different positioning values of the same target is completed, and the method comprises the following steps:
Based on the last clustering circle, using mean shift clustering, further clustering the clustering circles with the distance smaller than r by using k-means to obtain new positions of the original clustering circles;
removing the locating points already contained in the clustering ring, and performing de-duplication and merging on the rest locating points again by using a mean shift clustering algorithm;
and if the distance between the obtained clustering ring and the original clustering ring is larger than R, judging the clustering ring as a new target, otherwise judging the clustering ring as clutter, and forming a fuzzy clustering ring, wherein R is the radius of the target which can be stably detected by the unmanned radar, and R is the target detection error range.
2. The method of claim 1, wherein the time reference unifying the positioning data of each unmanned equipment in the unmanned cluster comprises:
Receiving data messages of each unmanned equipment in an unmanned cluster, respectively obtaining time stamp information of each unmanned equipment, and obtaining the local time of a fusion program as real time;
Continuously acquiring data for a plurality of seconds to obtain data of each acquired sample, and calculating the time stamp and the real time difference value of each unmanned equipment to obtain a plurality of groups of time differences;
And calculating the mean value and variance of all groups of time differences, subtracting the mean value from the message time of each unmanned equipment, and completing the time stamp calibration of each unmanned equipment.
3. The method of claim 2, wherein the fusing of the plurality of location values using cluster center distance weighting to achieve multi-source data spatial registration comprises:
Calculating the weight of each positioning point according to the distance between each positioning point and the center of the clustering ring;
and respectively carrying out weighted summation on the navigational speed, the position and the time stamp of the locating point by using the weight value, and calculating the position, the time stamp and the speed of the fused target.
4. A method according to claim 3, characterized by the fact that, byCalculating the weight w j of each positioning by using the distances dis (P j, O) of the n unmanned equipment to the positioning position P j and the clustering center position O of the same target;
From the following components Obtaining a Position After fusion of the fused target;
From the following components Obtaining a Time stamp Time After fusion of the fused target;
From the following components The velocity V After fusion of the fused target is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110293752.1A CN113075648B (en) | 2021-03-19 | 2021-03-19 | Clustering and filtering method for unmanned cluster target positioning information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110293752.1A CN113075648B (en) | 2021-03-19 | 2021-03-19 | Clustering and filtering method for unmanned cluster target positioning information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113075648A CN113075648A (en) | 2021-07-06 |
| CN113075648B true CN113075648B (en) | 2024-05-17 |
Family
ID=76613455
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110293752.1A Active CN113075648B (en) | 2021-03-19 | 2021-03-19 | Clustering and filtering method for unmanned cluster target positioning information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113075648B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113589251B (en) * | 2021-07-29 | 2024-01-05 | 上海无线电设备研究所 | Unit average constant false alarm detection method after Mean-Shift echo clustering |
| CN113537411B (en) * | 2021-09-15 | 2021-11-23 | 湖北工业大学 | Improved fuzzy clustering method based on millimeter wave radar |
| CN114114192B (en) * | 2021-12-02 | 2023-05-23 | 电子科技大学 | Cluster target detection method |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005301789A (en) * | 2004-04-14 | 2005-10-27 | Nara Institute Of Science & Technology | Cluster analysis device, cluster analysis method and cluster analysis program |
| EP2860661A1 (en) * | 2013-10-09 | 2015-04-15 | Delphi Technologies, Inc. | Mean shift tracking method |
| WO2015067119A1 (en) * | 2013-11-07 | 2015-05-14 | 华为技术有限公司 | Method for clustering position points of interest and related device |
| CN104885122A (en) * | 2012-12-25 | 2015-09-02 | 本田技研工业株式会社 | Vehicle periphery monitoring device |
| CN105243396A (en) * | 2015-11-06 | 2016-01-13 | 百度在线网络技术（北京）有限公司 | User position information generation method and device |
| CN105261004A (en) * | 2015-09-10 | 2016-01-20 | 西安电子科技大学 | Mean shift and neighborhood information based fuzzy C-mean image segmentation method |
| CN106646339A (en) * | 2017-01-06 | 2017-05-10 | 重庆邮电大学 | Online matching and positioning method in wireless position fingerprint indoor positioning |
| CN107948930A (en) * | 2017-12-31 | 2018-04-20 | 电子科技大学 | Indoor positioning optimization method based on location fingerprint algorithm |
| WO2018121414A1 (en) * | 2016-12-30 | 2018-07-05 | 中兴通讯股份有限公司 | Electronic device, and target image recognition method and apparatus |
| CN109041206A (en) * | 2018-07-03 | 2018-12-18 | 东南大学 | A kind of indoor positioning floor method of discrimination based on improvement fuzzy kernel clustering |
| CN109117861A (en) * | 2018-06-29 | 2019-01-01 | 浙江大学宁波理工学院 | A kind of multi-level cluster analysis method of point set for taking spatial position into account |
| CN110442143A (en) * | 2019-07-05 | 2019-11-12 | 北京航空航天大学 | A kind of unmanned plane situation data clustering method based on combination multiple target dove group's optimization |
| CN110569892A (en) * | 2019-08-28 | 2019-12-13 | 武汉大学 | Ubiquitous positioning signal static positioning method and system |
| CN111079859A (en) * | 2019-12-31 | 2020-04-28 | 哈尔滨工程大学 | Passive multi-station multi-target direction finding cross positioning and false point removing method |
| CN111340701A (en) * | 2020-02-24 | 2020-06-26 | 南京航空航天大学 | Circuit board image splicing method for screening matching points based on clustering method |
-
2021
- 2021-03-19 CN CN202110293752.1A patent/CN113075648B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005301789A (en) * | 2004-04-14 | 2005-10-27 | Nara Institute Of Science & Technology | Cluster analysis device, cluster analysis method and cluster analysis program |
| CN104885122A (en) * | 2012-12-25 | 2015-09-02 | 本田技研工业株式会社 | Vehicle periphery monitoring device |
| EP2860661A1 (en) * | 2013-10-09 | 2015-04-15 | Delphi Technologies, Inc. | Mean shift tracking method |
| WO2015067119A1 (en) * | 2013-11-07 | 2015-05-14 | 华为技术有限公司 | Method for clustering position points of interest and related device |
| CN105261004A (en) * | 2015-09-10 | 2016-01-20 | 西安电子科技大学 | Mean shift and neighborhood information based fuzzy C-mean image segmentation method |
| CN105243396A (en) * | 2015-11-06 | 2016-01-13 | 百度在线网络技术（北京）有限公司 | User position information generation method and device |
| WO2018121414A1 (en) * | 2016-12-30 | 2018-07-05 | 中兴通讯股份有限公司 | Electronic device, and target image recognition method and apparatus |
| CN106646339A (en) * | 2017-01-06 | 2017-05-10 | 重庆邮电大学 | Online matching and positioning method in wireless position fingerprint indoor positioning |
| CN107948930A (en) * | 2017-12-31 | 2018-04-20 | 电子科技大学 | Indoor positioning optimization method based on location fingerprint algorithm |
| CN109117861A (en) * | 2018-06-29 | 2019-01-01 | 浙江大学宁波理工学院 | A kind of multi-level cluster analysis method of point set for taking spatial position into account |
| CN109041206A (en) * | 2018-07-03 | 2018-12-18 | 东南大学 | A kind of indoor positioning floor method of discrimination based on improvement fuzzy kernel clustering |
| CN110442143A (en) * | 2019-07-05 | 2019-11-12 | 北京航空航天大学 | A kind of unmanned plane situation data clustering method based on combination multiple target dove group's optimization |
| CN110569892A (en) * | 2019-08-28 | 2019-12-13 | 武汉大学 | Ubiquitous positioning signal static positioning method and system |
| CN111079859A (en) * | 2019-12-31 | 2020-04-28 | 哈尔滨工程大学 | Passive multi-station multi-target direction finding cross positioning and false point removing method |
| CN111340701A (en) * | 2020-02-24 | 2020-06-26 | 南京航空航天大学 | Circuit board image splicing method for screening matching points based on clustering method |
Non-Patent Citations (11)
| Title |
|---|
| Design of Mean Shift Tracking Algorithm Based on Target Position Prediction;Wang, H;《2019 IEEE INTERNATIONAL CONFERENCE ON MECHATRONICS AND AUTOMATION (ICMA)》;20191231;全文 * |
| Improvement in Localization of a Moving Vehicle using K-means Clustering;Akhilesh Kumar;《2020 IEEE International Instrumentation and Measurement Technology Conference (I2MTC)》;20200630;全文 * |
| 基于WiFi与PDR融合的井下定位方法;沈辉;李世银;;煤矿机械;20201010(第10期);全文 * |
| 基于均值漂移聚类的扩展目标量测集划分算法;刘风梅;葛洪伟;杨金龙;李鹏;;计算机工程;20141215(第12期);全文 * |
| 基于无人机空中加油的目标追踪算法;王卓雅;屈耀红;闫建国;吴佳驹;;飞行力学;20180502(第04期);全文 * |
| 基于粒子Mean Shift迁移的红外人体目标跟踪算法;云廷进;郭永彩;高潮;;计算机学报;20090615(第06期);全文 * |
| 基于聚类技术的多阈值图像分割技术;陈强;;太赫兹科学与电子信息学报;20180828(第04期);全文 * |
| 基于非参数聚类和多尺度图像的目标跟踪;江焯林;黎绍发;贾西平;祝红丽;;华南理工大学学报(自然科学版);20090115(第01期);全文 * |
| 多站点低空防御系统关键技术研究;王炳琪;《制导与引信》;20191215;第40卷(第4期);摘要，第1-5章 * |
| 应用模糊增强及均值漂移实现红外目标分割;张坤华;张力;杨炬;;计算机工程与应用(第30期);全文 * |
| 模糊聚类分析在多目标跟踪中的应用;张晓丽;;国防技术基础;20080615(第06期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113075648A (en) | 2021-07-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113075648B (en) | Clustering and filtering method for unmanned cluster target positioning information | |
| CN106646338B (en) | A kind of quickly accurate indoor orientation method | |
| US9170324B2 (en) | Statistical movement analysis | |
| CN113777600B (en) | Multi-millimeter wave radar co-location tracking method | |
| US20140372073A1 (en) | Two-stage filtering based method for multiple target tracking | |
| CN107728140A (en) | A kind of surveillance radar multiple target multi-channel parallel tracking processing method | |
| US20060092074A1 (en) | Process for sensor resources management | |
| CN106443598A (en) | Convolutional neural network based cooperative radar network track deception jamming discrimination method | |
| CN112313536B (en) | Object state acquisition method, movable platform and storage medium | |
| US5604683A (en) | Evaluating target tracking when using multiple sensors | |
| CN106291498A (en) | A kind of detecting and tracking combined optimization method based on particle filter | |
| CN111366900B (en) | Tracking radar track quality evaluation method, system and medium based on residual statistics | |
| CN111279215A (en) | Target detection method and device, track management method and device and unmanned aerial vehicle | |
| CN117872354B (en) | Fusion method, device, equipment and medium of multi-millimeter wave Lei Dadian cloud | |
| KR20170103260A (en) | A method for target location using the tdoa information matching | |
| Solonar et al. | General construction principles and performance features of trajectory processing by data from one radar data source | |
| CN117724059A (en) | Multi-source sensor fusion track correction method based on Kalman filtering algorithm | |
| Cook et al. | Real-time radar-based tracking and state estimation of multiple non-conformant aircraft | |
| CN113390406B (en) | Multi-target data association and positioning method based on passive multi-sensor system | |
| EP3671255A1 (en) | Method for determining the location of a plurality of targets | |
| CN114513849B (en) | Outdoor non-line-of-sight propagation single-station positioning method based on scattering region model | |
| KR102361816B1 (en) | Method for detecting target and readable medium | |
| US10267895B1 (en) | Doppler-assisted sensor fusion | |
| CN114594467A (en) | Course angle determining method and device, electronic equipment and storage medium | |
| Yang et al. | Altitude estimation using multipath with a two-dimensional radar over spherical earth |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |