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

Global optimum particle filtering method and global optimum particle filter

A global optimal particle and particle technology, applied in the field of signal processing, can solve the problems of particle diversity and insufficient guiding ability of the optimization process, inability to effectively handle nonlinear and non-Gaussian signals, and increase the complexity of particle filtering and the amount of calculation. , to avoid particle degradation, ensure real-time and rapidity, and improve particle utilization.

Inactive Publication Date: 2017-07-04
李琳 +1
View PDF3 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to solve the problem that the existing particle filter algorithm will cause a large deviation between the sampling sample and the real posterior probability density sample, and the lack of guidance ability in controlling the particle diversity and the optimization process will increase the complexity of the particle filter And the amount of calculation, as well as the shortcomings of the existing particle degradation and particle impoverishment that cannot effectively deal with nonlinear and non-Gaussian signals, and construct a global optimal particle filter method and a global optimal particle filter

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
  • Global optimum particle filtering method and global optimum particle filter
  • Global optimum particle filtering method and global optimum particle filter
  • Global optimum particle filtering method and global optimum particle filter

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0036] The global optimal particle filter method provided by the present invention uses particles to describe the state space of the dynamic system, and the state space model of the nonlinear dynamic system is:

[0037] x k = f k-1 (x k-1 , u k-1 )

[0038] z k = h k (x k , v k )

[0039] where x k ∈R n is the n-dimensional system state vector at time k, z k ∈R m is the m-dimensional measurement vector at time k; the system state transition map and measurement map are f k-1 (×): R n × R n →R n and h k ( ): R m × R m →R m ; The process noise and measurement noise of the system are respectively u k-1 ∈ R n and v k ∈ R m .

[0040] It should be noted that the expression form of the state space model of the nonlinear system is equivalent to the above formula, that is, those skilled in the art can think that the formula expression of the nonlinear system is as shown in the above formula.

[0041] First, the insensitive Kalman filter algorithm is used to gener...

specific Embodiment approach 2

[0050] Specific implementation mode two: the difference between this implementation mode and specific implementation mode one is:

[0051] The initial particle set in step 1 is It is characterized in that the step 2 is specifically:

[0052] Step 2.1: Calculate the initial particle set mean of and variance Get UKF's suggested distribution Particles of which Satisfy

[0053] Step 2.2: Calculate the sampled particles weight of And normalize to get the normalized weight which is

[0054]

[0055] Step 2.3: According to the particles and its weight Import to Sampled Particles

[0056] Other steps and parameters are the same as those in Embodiment 1.

specific Embodiment approach 3

[0057] Specific embodiment three: the difference between this embodiment and specific embodiment one or two is:

[0058] Step 3 is specifically: use the floating-point number format with a fixed effective number l to convert the particle Expressed as The encoded particle set is obtained as in Indicates the value of the lth significant digit of the Nth particle.

[0059] the first digit of the floating point value Represents the sign bit, "1" represents a positive number, and "0" represents a negative number. The fixed effective number of digits l is set through the pre-filtering range. It should be noted here that the precision in Matlab is 4 digits after the decimal point, and if the number of digits is less than l, the highest digit is filled with 0. For example, the state value of the i-th particle at time k is 15.6745, and l=7, then its floating-point number format is as follows figure 2 shown.

[0060] Other steps and parameters are the same as those in Embod...

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 relates to a global optimum particle filtering method and a global optimum particle filter and belongs to the field of signal processing. The defect that according to an existing particle filter, relatively high deviation exists between samples and true posterior probability density samples is overcome, and the problem of processing nonlinearity and non-Gaussian signal through particle filtering is effectively solved. The main technical way is establishing the global optimum particle filter through utilization of a Lamarch genetic natural law. The global optimum particle filtering method comprises the steps of generating an initial particle set; carrying out importance sampling on the initial particle set through unscented Kalman filter, thereby obtaining sample particles; carrying out float-point encoding on each sample particle, thereby obtaining an encoded particle set; setting an initial population; taking the initial population as an original test initial and carrying out Lamarch rewrite operation, real number decoding operation and elitism reservation operation in sequence; and taking real number form optimum candidate particles as prediction samples of the next moment, thereby obtaining a state estimation value of a system. The method and the filter are applicable to machine learning.

Description

technical field [0001] The invention relates to a global optimal particle filter method and a global optimal particle filter, belonging to the field of signal processing. Background technique [0002] The state estimation of dynamic systems involves many fields, especially signal processing, artificial intelligence and image processing, and it also has important application value in navigation and guidance, information fusion, automatic control, financial analysis, intelligent monitoring and other fields. The traditional Kalman filter is only suitable for linear Gaussian systems, and the extended Kalman filter can only deal with the weak nonlinearity of the system. Therefore, the particle filter, which is not limited by the system model characteristics and noise distribution, has attracted much attention in the filtering problems of nonlinear and non-Gaussian dynamic systems. [0003] Particle filter is a filtering method based on Monte Carlo simulation and recursive Bayesi...

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): H03H17/02G06N3/00G06N3/12
CPCH03H17/0257G06N3/006G06N3/126G06N7/01
Inventor 李琳李耘
Owner 李琳
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