Dissertation > Excellent graduate degree dissertation topics show

Research and Improvement of Particle Filter Algorithm

Author: ZhaoNa
Tutor: FengChi
School: Harbin Engineering University
Course: Communication and Information System
Keywords: particle filter Markov Chain Monte Carlo resampling algorithm interpolation filter
CLC: TN713
Type: Master's thesis
Year: 2009
Downloads: 196
Quote: 4
Read: Download Dissertation

Abstract


Particle filter is an algorithm based on Monte Carlo and recursive Bayesian estimation. Its basic idea is to use series of weighted samples to approximate posterior probability density distribution in the state space and to replace integral operation by sample mean value. In principle, particle filter can realize any state estimation, especially shows excellent performance in nonlinear and non-Gaussian problems while Kalman filter and extended Kalman filter are disabled. Particle filter has been widely used in the fields such as target tracking, vision tracking, fault diagnosis, navigation position and radio communication etc.This paper studies particle filtering algorithm in depth, introduces realization theory and steps of the algorithm in detail, and discusses the problems of the algorithm. Improved methods are put forward to solve these problems. Firstly, the classical improved algorithm of Markov Chain Monte Carlo particle filter is improved. The improved algorithm brings in the concept of effective particles, abandons the degenerate particles at proper time, and adjusts the number of particle dynamically. As a result, this algorithm can reduce the computational complexity, and improve the running efficiency.Resampling algorithm is an important step and one of important methods to solve degeneracy problem of particle filter, but brings the problem of particle impoverishment. An improved resampling algorithm is proposed, which processes interpolation on particles to produce new particles. The algorithm increases the diversity of particles and improves precision of the algorithm. The efficiency of the algorithm is proved by simulation finally.To enhance the precision of the algorithm further, starting with the nonlinear estimate theory, the paper introduces interpolation filter algorithm to improve the particle filter, and presents Interpolation Particle Filter Algorithm. The improved algorithm expands nonlinear system formula according to Stirling’s interpolation formula, which can reduce the truncated errors of local linearization, and the latest measurements are integrated into the system state transition density so that the approximation to the system posterior is improved. Simulation results indicate that the improved algorithm enhances the accuracy of particle filter obviously.

Related Dissertations

  1. Research on Cooperative Orbit Determination in Satellite Network Based on Multi-Agent System Theory,V474
  2. The Maneuvering Target Tracking Research Based on VRPF,TN957.52
  3. The Research of 3D Human Motion Capture Based on Reference Points,TP391.41
  4. Space Infrared Target Simulation and Application of Target Trace,TP391.41
  5. Detection and Tracking of Moving Object in Complex Background,TP391.41
  6. The Research on the Target Localization and Tracking Based on WSN,TN929.5
  7. Architectural Optimizations for Particle Filters,TN713
  8. Petri net -based network intrusion detection system Research and Implementation,TP393.08
  9. Based on the theory of multi-target tracking FISST Research,TN953
  10. Study on Timing Recovery Technology for All-digital QAM Receivers,TN851
  11. Research on Attitude Determination and Control of Small Satellite Based on Nonlinear Filter,V448.2
  12. Research on TSOA/AOA Positioning Technology in CDMA Network,TN929.533
  13. Research on DOA Estimation and Tracking Approach Based on Monte Carol Method,TN911.7
  14. Outstanding Loss Reserve Estimation Method Comparison Study,F840
  15. Design of Digital Up-converter Based on FPGA,TN773
  16. The Application of Bayesian Method in Actuarial,F224;F840
  17. The Research of Nonlinear Time Series Based on Particle Filter Optimal Estimate,TP391.41
  18. Research on the Theory of Visual Object Tracking Based on Particle Filter for Autonomous Robot,TP242
  19. Detection and Tracking of Underwater Target,TP391.41
  20. Moving Object Detection and Tracking in Video Sequences,TP391.41
  21. Video Vehicle Detection Algorithm Based on Gaussian Model and Condensation Filter,TP391.41

CLC: > Industrial Technology > Radio electronics, telecommunications technology > Basic electronic circuits > Filtering techniques,the filter
© 2012 www.DissertationTopic.Net  Mobile