Unlock instant, AI-driven research and patent intelligence for your innovation.

A multi-target tracking data association method and system

A multi-target tracking and data association technology, applied in the field of multi-target tracking data association methods and systems, can solve the problems of low correct association rate and high computational complexity, achieve fast calculation speed, improve association accuracy, and avoid combination explosion problem effect

Active Publication Date: 2021-11-19
SUN YAT SEN UNIV
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The present invention provides a method and system for multi-target tracking data association, which is used to overcome the defects of low correct association rate or relatively large calculation complexity in the prior art, and realize improvement of correct association rate and reduction of calculation complexity

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 multi-target tracking data association method and system
  • A multi-target tracking data association method and system
  • A multi-target tracking data association method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0029] as attached Figure 1-7 As shown, the embodiment of the present invention provides a multi-target tracking data association method, which is not only applicable to the data association in the process of single-target tracking, but also applicable to the data association in the process of tracking two or more targets. Please refer to figure 1 , including:

[0030] Step S1, constructing a reinforcement learning data association model for predicting the target position at the current moment in combination with the state of the target at the previous moment and the motion attributes;

[0031] The current moment here is known in the training process as the radar’s real measurement point data (excluding clutter measurement data) of the target in a certain period of time and the real track point data of the target in the certain period of time. The process is known to be the measurement point data (including clutter point data and real data) of the radar target after the abov...

Embodiment 2

[0135] Based on the first embodiment above, the present invention provides a multi-target tracking data association system, including a memory and a processor, the memory stores a multi-target tracking data association program, and the processor runs the multi-target tracking data association program Execute the steps in any embodiment of the method above.

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 discloses a multi-target tracking data association method and system. The method includes considering the track of the known starting time period of the target measurement point as a reinforcement learning training process according to the multi-target tracking data association characteristics, and in one step the known measurement Random clutter is generated around the point, and both clutter points and known measurement points are regarded as radar acquisition measurement points; candidate measurement points are selected from the measurement points according to the tracking gate, and target motion characteristics are used for motion matching and reinforcement learning Data association is performed on all candidate measurement points according to the matching degree and position distribution law, and the experience matrix of the reinforcement learning model is trained by one-step known measurement points to test the correlation results; according to the trained experience matrix, combined with motion matching, the target enters the clutter area Data association is carried out on the track points, and the experience matrix is ​​continuously optimized by the association results until the track association is completed. Solve the problems of low correct association rate and high computational complexity in the prior art, improve the correct association rate and reduce computational complexity.

Description

technical field [0001] The invention relates to the technical field of multi-target tracking, in particular to a multi-target tracking data association method and system. It is suitable for multi-target tracking data association in a multi-clutter environment. Background technique [0002] The basic concept of multi-target tracking was first proposed by Wax in 1955. In 1964, Sutler conducted in-depth research on the theory of multi-target tracking and data association, and made pioneering progress. However, it was not until the early 1970s that the theory of maneuvering target tracking really attracted people's attention. During this period, the multi-target tracking technology marked by the organic combination of data association technology and Kalman filtering technology pioneered by Bar-shalom and Singer has made a breakthrough. However, target tracking data association in a dense clutter environment has always been a difficult problem in the field of multi-target track...

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 张艳曲承志苏东杨雪榕张鑫
Owner SUN YAT SEN UNIV