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

PHD multi-target tracking method based on variational Bayesian T distributed Kalman filtering

A variational Bayesian and Kalman filtering technology, applied in the field of multi-target tracking and robust PHD tracking, it can solve the problem that it is difficult to effectively handle measurement outliers, the robustness of the filtering algorithm is not strong, and the measurement abnormality cannot be optimized. questions of value

Pending Publication Date: 2020-06-23
JIANGSU UNIV OF TECH
View PDF2 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

UK achieves local linear optimization by finding feature points, which is also suitable for weak linear scenarios and cannot optimize the measurement of outliers
It can be seen that the main problem of the target processing nonlinear tracking scene is that the robustness of the filtering algorithm is not strong, and it is difficult to effectively deal with the measurement outliers

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
  • PHD multi-target tracking method based on variational Bayesian T distributed Kalman filtering
  • PHD multi-target tracking method based on variational Bayesian T distributed Kalman filtering
  • PHD multi-target tracking method based on variational Bayesian T distributed Kalman filtering

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0066] The technical solution of the present invention will be further described in detail below in conjunction with the accompanying drawings.

[0067] The basic theory involved in the present invention is introduced as follows:

[0068] 1. Variational Bayesian inference techniques

[0069] Assuming that the measurement is Z and the target state is X, in the recursive process, if you want to use the function Ψ(Z) to approximate the posterior probability density g(Z|X), you can iteratively approximate it by reducing the KL divergence, that is

[0070]

[0071] make

[0072] L(Ψ)=∫Ψ(Z)logg(Z,X)dZ-∫Ψ(Z)logΨ(Z)dZ

[0073] Then Ψ(Z) can be expressed as logg(X)=KL(Ψ||g)+L(Ψ)

[0074] So the problem is transformed into how to minimize KL(Ψ||g). Obviously, if you want to minimize KL(Ψ||g), you only need to maximize L(Ψ). Assuming that the hidden variables are independent of each other, then L(Ψ) can be expressed as

[0075]

[0076] Let -KL(Ψ(z j )||Ψ * (z j ))=0, then...

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 PHD multi-target tracking method based on variational Bayesian T distribution Kalman filtering, belongs to the technical field of guidance and intelligent information processing, and mainly solves the problem that the precision of a multi-target tracking algorithm is reduced under the conditions of nonlinearity and abnormal measurement values. The method is based on a variational Bayesian inference framework. The target state is updated by adopting the T distributed Kalman filtering technology, and the posterior probability density of the target is deduced based on the variational Bayesian framework and the T distributed Kalman filtering, so that the robustness and the overall tracking precision of the PHD tracking algorithm are improved, the design requirements of an actual engineering system can be met, and the engineering application value is high.

Description

technical field [0001] The invention belongs to the technical field of intelligent information processing and relates to a multi-target tracking method. Specifically, it is a robust PHD tracking method based on variational Bayesian and T-distributed Kalman filtering techniques, which can be used for target tracking in radar systems such as unmanned vehicles, air control and precision guidance. Background technique [0002] Traditional target tracking technology is mostly suitable for linear measurement scenarios, but the measurement information in real tracking scenarios has a multi-dimensional nonlinear model, and the measurement error will fluctuate greatly in the case of sudden interference, resulting in measurement outliers. Decreases the target tracking accuracy. The current classic multi-target tracking frameworks include Probability Hypothesis Density (PHD), Potential PHD (CPHD), etc. These methods are mainly based on Bayesian inference and Kalman filtering technolog...

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 Applications(China)
IPC IPC(8): G06T7/277G06N5/04G06N7/00
CPCG06T7/277G06N5/041G06T2207/20076G06N7/01
Inventor 李鹏王文慧舒振球邱骏达由从哲李嘉伟徐宏鹏
Owner JIANGSU UNIV 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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com