Indoor localization using particle filter and map-based NLOS ranging model

User localization is one of the key technologies for mobile robots to successfully interact with humans. Among various localization methods using radio frequency (RF) signals, time of arrival (TOA) based localization is popular since the target coordinates can be directly calculated from the accurate range measurements. In complex indoor environment, however, RF ranging-based localization is quite challenging since the range measurements suffer not only from signal noise but also from signal blockages and reflections. A set of range measurements taken in complex indoor environment verifies that almost all measurements are non-line-of-sight (NLOS) ranges which have striking difference to the line-of-sight (LOS) distances. These NLOS range measurements make severe degradation in the accuracy of trilateration based localizations if used without any compensation. In this paper we propose a particle filter-based localization algorithm which utilizes indoor geometry from a given map to estimate the NLOS signal path and compensates for the range measurements. The algorithm is verified with experiments performed in real indoor environments.