Fast Bayesian tracking for location estimation by grid-based method

With increasing demand on location estimation, Bayesian tracking algorithm for mobile devices that requires low computing power is needed. By nonlinear/non-Gaussian Bayesian methods, grid-based methods are considered inefficient compared with the particle filters. We propose fast grid-based Markov l...

Full description

Saved in:
Bibliographic Details
Main Authors: Wanhee Kim, Sook-Jin Lee
Format: Conference Proceeding
Language:English
Subjects:
Bayesian
Bayesian methods
Dead reckoning
Estimation
Global Positioning System
grid-based
location estimation
Markov localization
Markov processes
Mobile handsets
Particle filters
sensor fusion
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:With increasing demand on location estimation, Bayesian tracking algorithm for mobile devices that requires low computing power is needed. By nonlinear/non-Gaussian Bayesian methods, grid-based methods are considered inefficient compared with the particle filters. We propose fast grid-based Markov localization algorithm by reducing the computational complexity of grid-based method. The fusion of dead reckoning and GPS location simulation shows that fast grid-based algorithm shows better performance on both location accuracy and time requirement.
ISSN:2162-1233
DOI:10.1109/ICTC.2012.6386792