Dynamic state estimation using particle filter and adaptive vector quantizer

• Published in: 2009 IEEE International Symposium on Computational Intelligence in Robotics and Automation pp. 429 - 434
• Main Authors: Nishida, T., Kogushi, W., Takagi, N., Kurogi, S.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.12.2009
• Subjects: Bayesian methods; Data mining; Distortion measurement; Information filtering; Information filters; Particle filters; Robustness; Shape; State estimation; Vector quantization
• ISBN: 1424448085; 9781424448081
• DOI: 10.1109/CIRA.2009.5423166

Particle filter (PF) is a method for discrete approximation of dynamic and non-Gaussian probability distribution by using numerous particles, and its procedure can execute at high speed and is suitable for on-line applications. However, in conventional methods, a weighted average value or a maximum weighted value of particles is used as a filter output, and information on most particles is disregarded. On the other hand, an adaptive vector quantization (AVQ) algorithm called competitive reinitialization learning (CRL) that can achieve high-speed adaptation without depending on initial conditions has been proposed. Then, in this research, a method for extracting information on shape of probability density distributions by combining PF with CRL is proposed. Moreover, a rapid adaptation performance and the robustness of the proposed method are shown by the simulations.