CN106546975B - A kind of small-sized unmanned plane based on radar data and flying bird classifying identification method - Google Patents

A kind of small-sized unmanned plane based on radar data and flying bird classifying identification method Download PDF

Info

Publication number
CN106546975B
CN106546975B CN201610896005.6A CN201610896005A CN106546975B CN 106546975 B CN106546975 B CN 106546975B CN 201610896005 A CN201610896005 A CN 201610896005A CN 106546975 B CN106546975 B CN 106546975B
Authority
CN
China
Prior art keywords
model
target
moment
small
unmanned plane
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201610896005.6A
Other languages
Chinese (zh)
Other versions
CN106546975A (en
Inventor
陈唯实
李敬
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Academy of Civil Aviation Science and Technology
Original Assignee
China Academy of Civil Aviation Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China Academy of Civil Aviation Science and Technology filed Critical China Academy of Civil Aviation Science and Technology
Priority to CN201610896005.6A priority Critical patent/CN106546975B/en
Publication of CN106546975A publication Critical patent/CN106546975A/en
Application granted granted Critical
Publication of CN106546975B publication Critical patent/CN106546975B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • 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
    • G01S7/411—Identification of targets based on measurements of radar reflectivity

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

The small-sized unmanned plane and flying bird classifying identification method that the invention discloses a kind of based on radar data.The target measurement information that small-sized unmanned plane proposed by the present invention and flying bird classifying identification method are obtained based on radar, pass through four steps such as multiple model target tracking, the extraction of multi-model probability, the judgement of target motor pattern, Target Motion Character extraction, the target signatures such as target movement model conversion frequency finally are extracted, to distinguish small-sized unmanned plane target and flying bird target.The shortcomings that small-sized unmanned plane and flying bird cannot be distinguished the present invention overcomes low altitude airspace radar surveillance system, Target Motion Character is extracted using radar data, to distinguish small-sized unmanned plane target and flying bird target, the recognition capability of low altitude airspace radar surveillance system is promoted.

Description

A kind of small-sized unmanned plane based on radar data and flying bird classifying identification method
Technical field
The small-sized unmanned plane and flying bird classifying identification method that the present invention relates to a kind of based on radar data, belong to radar mesh Tracking technique field is marked, is related to radar signal feature and target identification is classified.
Background technique
In recent years, the small-sized unmanned plane quantity of business level rapidly increases.Small-sized unmanned plane belong to it is typical " it is low, slow, It is small " target, refer to a kind of aircraft that flying height is low, speed is slow, target scattering sections are small, such aircraft it is small in size, winged Row height is low, it is more to be blocked by atural object, it is difficult to detect, identify, it is often more important that, the small-sized unmanned plane of business level is easily terrorist It utilizes, seriously threatens national air defence safety.
Complicated low-altitude surveillance radar is widely used in the low altitude airspace security monitoring in important sensitive area, border region.This Class radar uses solid LDA signal processor technology, can detect the small weak mesh in the strong clutter environment of complicated low altitude airspace Mark carries out networking detection using single portion's X-band, S-band or multi-section X-band, S-band radar, realizes to a wide range of low altitude airspace Inexpensive all weather surveillance.
But the characteristic informations such as scattering section, flying speed, flying height of the biological targets such as flying bird in low altitude airspace Close with small-sized unmanned plane, existing low-altitude surveillance radar is difficult to differentiate between, and easily leads to false-alarm when detecting small-sized unmanned plane.
Summary of the invention
The purpose of the present invention is to solve the above problem, proposes a kind of small-sized unmanned plane based on radar data and fly Bird classifying identification method, the small-sized unmanned plane tracking suitable for complicated low latitude environment, can reject the jamming targets such as flying bird, Promote tracking effect.
A kind of small-sized unmanned plane based on radar data and flying bird classifying identification method, extract mesh using radar data Mark motion feature includes the following steps: to distinguish small-sized unmanned plane target and flying bird target
Step 1: multiple model target tracking；
Step 2: multi-model probability extracts；
Step 3: target motor pattern judges；
Step 4: Target Motion Character extracts.
The present invention has the advantages that
(1) Target Motion Character is extracted using radar data, to distinguish small-sized unmanned plane target and flying bird target, Promote the recognition capability of low altitude airspace radar surveillance system；
It (2), only need to be to radar without being modified to hardware components such as the signal processings of low altitude airspace radar surveillance system The software sections such as data processing carry out upgrading, that is, are able to achieve the rejecting to false-alarm targets such as flying birds.
Detailed description of the invention
Fig. 1 is the flow chart of the small-sized unmanned plane based on radar data and flying bird classifying identification method of the invention；
Fig. 2 is the small-sized unmanned plane of the embodiment of the present invention and the track following schematic diagram of flying bird target；
Fig. 3 is the multi-model probability curve of the target 1 (small-sized unmanned plane) of the embodiment of the present invention；
Fig. 4 is the multi-model probability curve of the target 2 (flying bird) of the embodiment of the present invention；
Fig. 5 is the motion model staircase curve of the target 1 (small-sized unmanned plane) of the embodiment of the present invention；
Fig. 6 is the motion model staircase curve of the target 2 (flying bird) of the embodiment of the present invention.
Specific embodiment
Below in conjunction with drawings and examples, the present invention is described in further detail.
The present invention is a kind of small-sized unmanned plane based on radar data and flying bird classifying identification method, utilizes flying bird and light The jamming targets such as flying bird are rejected in the difference of small drone target movement model, realize the small-sized nothing in the environment of complicated low latitude The tracking and identification of man-machine target, process is as shown in Figure 1, include the following steps:
Step 1: multiple model target tracking；
Since each target may have multi-motion modes during the motion, at a time, multi-model target with Track uses the model concurrent working of n kind, and the state estimation generated by mixing last moment all filters obtains certain model and matches Set the primary condition of filter.
Each model MiAnd MjIn the mixing probability at k momentIt calculates as follows
In formula,For k-1 moment model MiProbability,For normalized parameter, in subsequent time by model MiIt is transformed into Model MjProbability be expressed as:
Then, the Mixed design of each filter is calculated:
Wherein,WithIt is the update mean value and covariance of k-1 moment model i.
To each model Mi, it filters as follows:
In formula, F is used in estimating and updating for standard Kalman filter respectivelyP() and Fu() indicates, ykIt is the k moment It measures,WithIt is k moment model MiPredictive mean value and covariance,WithFor k-1 moment model MiConversion square Battle array and process noise matrix,WithFor k moment model MiMeasurement model matrix and measure noise matrix.In addition, also calculating The measurement similitude of each filter
WhereinTo measure residual error,For model MiThe covariance of part is updated in filtering, N () is Gaussian probability density Distribution function.
Step 2: multi-model probability extracts；
K moment each model MiProbabilityIt calculates as follows:
Wherein c is normalization factor.
Step 3: target motor pattern judges；
In k moment n kind parallel existing model, select probabilityMaximum model MiTarget as the k moment moves Mode Ik, it is denoted as
Step 4: Target Motion Character extracts；
The conversion frequency F for extracting target movement model, is calculated by following formula
That is the target number that motor pattern changes in set time T, NkIt indicates until the k moment, target motor pattern The cumulative number of variation, is calculated by following formula
The mobility of flying bird target be higher than small-sized unmanned plane, given threshold S, if object module conversion frequency be higher than S, It is then flying bird target, conversely, being then small-sized unmanned plane target.
Embodiment:
The tracking of radar target and recognition result are based on thunder to proposed by the present invention in middle two-dimensional space with reference to the accompanying drawing Small-sized unmanned plane up to data is illustrated and is described with flying bird classifying identification method.
General low altitude airspace radar surveillance system is difficult to differentiate between small-sized unmanned plane and flying bird target, is based on radar data Small-sized unmanned plane and flying bird classifying identification method, the radar target tracking suitable for complicated low latitude environment can reject Flying bird target is obviously improved the target identification ability of low altitude airspace radar surveillance system.
Fig. 2-6 is the radar target tracking of the embodiment of the present invention and the schematic diagram of identification process, including small-sized unmanned plane With the tracking of the target trajectory of flying bird, target multi-model probability curve and target movement model staircase curve, to illustrate target following With the overall process of identification.
Small-sized unmanned plane and flying bird classifying identification method based on radar data of the invention, which is characterized in that utilize The jamming targets such as flying bird are rejected in the difference of flying bird and small-sized unmanned plane target motion model, are realized in the environment of complicated low latitude The tracking and identification of small-sized unmanned plane target, process is as shown in Figure 1, include the following steps:
Step 1: multiple model target tracking；
(k=0,2 ..., 50) target 1 is set forth in certain time in Fig. 2 and the path tracking of target 2 is shown Example, and be labeled in satellite map.Since each target may have multi-motion modes during the motion, in this example In, the model concurrent working of n=3 kind is used in all moment multiple model target trackings, by mixing last moment all filters The state estimation of generation obtains certain model (M1, M2, M3) configuration filter primary condition.
Each model MiAnd MjIn the mixing probability at k momentIt calculates as follows
In formula,For k-1 moment model MiProbability,For normalized parameter, in subsequent time by model MiIt is transformed into Model MjProbability be expressed as
Then, the Mixed design of each filter is calculated:
WhereinWithIt is k-1 moment model MiUpdate mean value and covariance.
To each model Mi, filter as follows
In formula, F is used in estimating and updating for standard Kalman filter respectivelyP() and Fu() indicates, ykIt is the k moment It measures,WithIt is k moment model MiPredictive mean value and covariance,WithFor k-1 moment model MiConversion square Battle array and process noise matrix,WithFor k moment model MiMeasurement model matrix and measure noise matrix.In addition, also calculating The measurement similitude of each filter
WhereinTo measure residual error,For model MiThe covariance of part is updated in filtering, N () is Gaussian probability density Distribution function.
Step 2: multi-model probability extracts；
K moment each model MiProbabilityIt calculates as follows:
Wherein c is normalization factor.
In this example, Fig. 3 and Fig. 4 be set forth target 1 and target 2 all moment (k=0,2 ..., 50) 3 kinds of models Probability curve
Step 3: target motor pattern judges；
In this example, in 3 kinds of the k moment parallel existing models, select probabilityMaximum model MiMesh as the k moment Mark motor pattern Ik, it is denoted as
Target 1 and target 2 is set forth in the motor pattern ladder at all moment (k=0,2 ..., 50) in Fig. 5 and Fig. 6 Curve.
Step 4: Target Motion Character extracts；
The conversion frequency F for extracting target movement model, is calculated by following formula
That is the target number that motor pattern changes in set time T, NkIt indicates until the k moment, target motor pattern The cumulative number of variation, is calculated by following formula
The mobility of flying bird target be higher than small-sized unmanned plane, given threshold S, if object module conversion frequency be higher than S, It is then flying bird target, conversely, being then small-sized unmanned plane target.
In this example, T=50 is set, then in entire observation process (k=0,2 ..., 50), F1=(N50-N0)/50=0, F2 =(N50-N0)/50=3/50=0.06；S=0.03 is set, and therefore, target 1 is determined as that unmanned plane target, target 2 are judged to flying Bird target.

Claims (1)

1. a kind of small-sized unmanned plane based on radar data and flying bird classifying identification method, include the following steps:
Step 1: multiple model target tracking；
At a time, multiple model target tracking uses the model concurrent working of n kind, raw by mixing last moment all filters At state estimation, obtain certain model configuration filter primary condition；
If model MiAnd MjIn the mixing probability at k momentAre as follows:
In formula,For k-1 moment model MiProbability,For normalized parameter, in subsequent time by model MiIt is transformed into model MjProbability be expressed as:
Wherein:Respectively indicate the model M at k momentjWith the model M at k-1 momenti, i and j be between 1~n from So number, and i ≠ j；
Calculate the Mixed design of each filter:
Wherein,The hybrid mean value and covariance that k-1 moment all filters generate are respectively indicated,WithIt is K-1 moment model MiUpdate mean value and covariance；
To each model Mi, it filters as follows:
In formula,Indicate k moment model MiUpdate mean value,Indicate k moment model MiCovariance, standard Kalman filtering F is used in estimating and updating for device respectivelyP() and Fu() indicates, ykIt is the measurement at k moment,WithIt is k moment model Mi's Predictive mean value and covariance,WithFor k-1 moment model MiTransition matrix and process noise matrix,WithFor k Moment model MiMeasurement model matrix and measure noise matrix；
Calculate the measurement similitude of each filter:
Wherein:To measure residual error,For model MiThe covariance of part is updated in filtering, N () is Gaussian probability density distribution Function；
Step 2: multi-model probability extracts；
K moment each model MiProbabilityAre as follows:
Wherein: c is normalization factor；For model MiNormalized parameter；
Step 3: target motor pattern judges；
In k moment n kind parallel existing model, select probabilityMaximum model MiTarget motor pattern as the k moment Ik, it is denoted as
Step 4: Target Motion Character extracts；
Extract the conversion frequency F of target movement model:
That is the target number that motor pattern changes in set time T, Nk-TIt indicates until the k-T moment, target motor pattern becomes The cumulative number of change, NkIt indicates until the k moment, the cumulative number of target motor pattern variation is calculated by following formula
Given threshold S is then judged as flying bird target, conversely, being judged as small-sized nothing if the conversion frequency of object module is higher than S Man-machine target.
CN201610896005.6A 2016-10-14 2016-10-14 A kind of small-sized unmanned plane based on radar data and flying bird classifying identification method Active CN106546975B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610896005.6A CN106546975B (en) 2016-10-14 2016-10-14 A kind of small-sized unmanned plane based on radar data and flying bird classifying identification method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610896005.6A CN106546975B (en) 2016-10-14 2016-10-14 A kind of small-sized unmanned plane based on radar data and flying bird classifying identification method

Publications (2)

Publication Number Publication Date
CN106546975A CN106546975A (en) 2017-03-29
CN106546975B true CN106546975B (en) 2019-04-26

Family

ID=58368865

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610896005.6A Active CN106546975B (en) 2016-10-14 2016-10-14 A kind of small-sized unmanned plane based on radar data and flying bird classifying identification method

Country Status (1)

Country Link
CN (1) CN106546975B (en)

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107097810B (en) * 2017-04-30 2018-04-20 中南大学 A kind of Along Railway foreign body intrusion UAV Intelligent identification and method for early warning and system
CN107678024B (en) * 2017-10-23 2020-12-29 中国民航科学技术研究院 Light and small unmanned aerial vehicle fusion tracking method based on radar and infrared combined detection
CN108519587B (en) * 2018-04-25 2021-11-12 东南大学 Real-time aerial target motion mode identification and parameter estimation method
CN108733073B (en) * 2018-05-21 2021-11-23 厦门安胜网络科技有限公司 System and method for controlling unmanned aerial vehicle in region and readable medium
CN109239702B (en) * 2018-10-17 2022-07-19 北京航空航天大学 Airport low-altitude flying bird number statistical method based on target state set
CN111679257B (en) * 2019-12-30 2023-05-23 中国船舶集团有限公司 Target recognition method and device for light unmanned aerial vehicle based on radar detection data
CN111429791B (en) * 2020-04-09 2022-11-18 浙江大华技术股份有限公司 Identity determination method, identity determination device, storage medium and electronic device
CN113537347B (en) * 2021-07-15 2023-07-18 北京航空航天大学 Unmanned aerial vehicle and flying bird target classification method based on track motion characteristics
CN113658225A (en) * 2021-08-19 2021-11-16 天之翼(苏州)科技有限公司 Moving object identification method and system based on aerial photography monitoring
CN117554366A (en) * 2021-11-23 2024-02-13 葛凌云 Agricultural Internet of things application service monitoring platform

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101414006A (en) * 2008-11-28 2009-04-22 北京航空航天大学 Flying bird object identifying and tracking method based on radar image
CN101697007A (en) * 2008-11-28 2010-04-21 北京航空航天大学 Radar image-based flyer target identifying and tracking method
CN103913740A (en) * 2014-04-16 2014-07-09 中国民航科学技术研究院 Bird flock target tracking method based on spatial distribution characteristics

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4884096B2 (en) * 2006-06-20 2012-02-22 アルパイン株式会社 Object identification system
US10310068B2 (en) * 2014-12-08 2019-06-04 Northrop Grumman Systems Corporation Variational track management

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101414006A (en) * 2008-11-28 2009-04-22 北京航空航天大学 Flying bird object identifying and tracking method based on radar image
CN101697007A (en) * 2008-11-28 2010-04-21 北京航空航天大学 Radar image-based flyer target identifying and tracking method
CN103913740A (en) * 2014-04-16 2014-07-09 中国民航科学技术研究院 Bird flock target tracking method based on spatial distribution characteristics

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于空域特性的低空空域雷达目标检测;陈唯实等;《航空学报》;20150925;第36卷(第9期);第3060-3068页

Also Published As

Publication number Publication date
CN106546975A (en) 2017-03-29

Similar Documents

Publication Publication Date Title
CN106546975B (en) A kind of small-sized unmanned plane based on radar data and flying bird classifying identification method
EP3167305B1 (en) A classification system for radar and sonar applications
US20150260838A1 (en) Sparse Array RF Imaging for Surveillance Applications
PL220849B1 (en) Weather and airborne clutter suppression using a cluster shape classifier
Chen et al. Classification of UAV and bird target in low-altitude airspace with surveillance radar data
KR102170632B1 (en) Method and apparatus for detecting anomalous behavior in real-time in cluestered system
CN111401168B (en) Multilayer radar feature extraction and selection method for unmanned aerial vehicle
CN109117776B (en) Aircraft and meteorological clutter classification and identification method based on flight path information
CN111142085B (en) External radiation source radar target classification and identification method based on track feature extraction
CN116794647A (en) Wire and tower classification based on trajectory tracking
Sim et al. Effective ground-truthing of supervised machine learning for drone classification
CN112881993A (en) Method for automatically identifying false tracks caused by radar distribution clutter
JP2000508429A (en) Clutter elimination using adaptive evaluation of clutter probability density function
Aydın et al. Development of a new light-weight convolutional neural network for acoustic-based amateur drone detection
CN114139373A (en) Multi-sensor automatic cooperative management method for unmanned aerial vehicle reverse braking
Mehta et al. Tracking and classification of drones and birds at a far distance using radar data
Sévigny et al. Unmanned aircraft (UA) telemetry data for track modelling and classification
Khan et al. Translearn-yolox: Improved-yolo with transfer learning for fast and accurate multiclass uav detection
Wang et al. Infrared unmanned aerial vehicle targets detection based on multi-scale filtering and feature fusion
McDonald et al. Track-before-detect using swerling 0, 1, and 3 target models for small manoeuvring maritime targets
CN110488279B (en) Pedestrian and tree distinguishing method and device based on radar
CN113625266A (en) Method, device, storage medium and equipment for detecting low-speed target by using radar
CN112257699A (en) Bird target feature analysis and extraction method based on visible light image and SAR image
CN110874847A (en) System and method for countering aerial moving target
Vylegzhanin et al. Fuzzy logic alghorithms for target classification in radar observations

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