Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A Radar Target Track Initiation Method Based on Support Vector Machine

A support vector machine and radar target technology, which is applied in the field of track initiation based on support vector machine, can solve the problems of rough sequential processing method rules, large calculation amount of batch processing method, and large batches of measurement data, etc. It takes a long time to start, meets the real-time requirements, and has the effect of small online calculation

Active Publication Date: 2020-04-24
HARBIN INST OF TECH
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to solve the shortcomings of the existing sequential processing method, such as rough rules, the need to set empirical thresholds, and a sharp drop in performance under strong clutter backgrounds; and the batch processing method has a huge amount of calculation and requires many batches of measurement data. In order to solve the problem of long starting time, a radar target track starting method based on support vector machine is proposed

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A Radar Target Track Initiation Method Based on Support Vector Machine
  • A Radar Target Track Initiation Method Based on Support Vector Machine
  • A Radar Target Track Initiation Method Based on Support Vector Machine

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0023] Specific implementation mode one: combine figure 1 Describe this embodiment, a kind of radar target track initiation method based on support vector machine of this embodiment, concrete process is:

[0024] Step 1: Extract the motion information of the radar detection target as the training sample feature;

[0025] Radar detection targets include real targets and false targets;

[0026] Step 2: Use the training sample features to train the support vector machine to obtain the decision function of the optimal hyperplane for radar target classification;

[0027] Step 3: Use the traditional heuristic rule method to pre-select the radar measurement data to be classified to form the primary track;

[0028] Step 4: Use the support vector machine trained in step 2 as a classifier to classify the primary track obtained in step 3, distinguish between real targets and false targets, and obtain the initial result of the track.

specific Embodiment approach 2

[0029] Specific embodiment two, the difference between this embodiment and specific embodiment one is: in the described step one, extract the motion information of the radar detection target as the training sample feature; the specific process is:

[0030] Take the radar detection target as the training sample, the number of training samples is L, the radar detection target includes real targets and false targets (some of which are from real targets, and some are from false targets); each training sample is composed of batches detected by radar signals Measurement combination, set the number of measurement combination points as N, expressed as the following formula:

[0031]

[0032] In the formula, MC k Indicates the kth radar target measurement combination, Indicates the position vector of the i-th batch of points in the k-th radar target measurement combination, 1≤i≤N; N and L are positive integers;

[0033] From the measurement combination MC k Extract the velocity ...

specific Embodiment approach 3

[0050]Specific embodiment three: the difference between this embodiment and specific embodiment one or two is: use training sample feature training support vector machine in described step 2, obtain the decision function of radar target classification optimal hyperplane; Concrete process is:

[0051] The principle of support vector machine training can be summarized as: find a hyperplane that satisfies the classification criteria, guarantees the accuracy requirements, and has the largest interval on both sides of the feature space. Margin maximization is the learning strategy of support vector machine, and the solution problem of hyperplane can be transformed into the solution of a convex quadratic programming problem.

[0052] Set a hyperplane w x+b=0, w is the normal vector, x is the feature vector, and b is the intercept;

[0053] if w·x k +b≥0, then judge x k The sample classification result is 1, otherwise it is -1, set the hyperplane w·x+b=0 to correctly classify all t...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a method for initiating a track of a radar target based on a support vector machine, and relates to a method for initiating a track of a radar target based on a support vector machine, which aims at overcoming the defects of rough rule, setting of an empirical threshold, sudden degrading of property under strong clutter background and the like in the existing sequential processing method, and solving the problems of huge calculation amount and multiple batches of data measuring in a batch processing method. The method specifically comprises the following steps: 1, extracting motion information of a radar detection target as training sample features; 2, utilizing the training sample features to train the support vector machine, so as to obtain a decision function of an optimal hyperplane for classifying of the radar target; 3, utilizing a traditional enlightening type rule method to pre-select to-be-classified measuring data of a radar, so as to form pre-selected tracks; 4, by using the trained support vector machine as a classifier, classifying the pre-selected tracks, and distinguishing a real target from a virtual target, so as to obtain a track initiating result. The method belongs to the fields of radar target data processing and machine learning.

Description

technical field [0001] The invention relates to radar data processing and machine learning, in particular to a track initiation method based on a support vector machine. Background technique [0002] Track initiation is the primary problem of radar multi-target tracking, and its correctness is an effective measure to reduce the computational burden brought by the combined explosion of radar multi-target tracking. If the track starts incorrectly, the target will be lost and it will be impossible to track the target at all. Moreover, at the beginning of the track, the target distance is far away, the radar detection resolution is low, the measurement accuracy is poor, and there is no statistical law for the appearance of true and false targets, so the problem of track start is a difficult problem to deal with. . [0003] The processing method of track initiation is mainly divided into sequential processing method and batch processing method according to the different data pr...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G01S13/72
CPCG01S13/726
Inventor 李宏博刘硕张云白杨
Owner HARBIN INST OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products