Topological Enviromental Localization Based on Particle Filters

王道顺,王謇

DOI: https://doi.org/10.3969/j.issn.1673-9590.2010.04.019

2010-01-01

Abstract:An improved particle filter is introduced for mobile robot indoor localization based on a large-scale topological environmental model.For localization,the traditional particle filters use large number of samples when applied in large-scale environment,which greatly increase the computational cost and the memory space.To solve this problem,the particle filters are improved by selective distributing.Estimation of the possible nodes where the robot is adjacent to in the large-scale topological environment model is conducred by laser scan-matching,and the samples are distribute around the candidate nodes so that the number will decrease remarkably,comparing with the traditional particle filters.Experiments result proves the validity and practicability of this method.