Low Level Navigation System for a POMDP based on WiFi and Ultrasound Observations

Abstract

In this paper we present the low level navigation system carried out in a partially observable Markov decision process (POMDP) based on WiFi and ultrasound observations. We use an H-shape model for the corridor, obtained from ultrasound range sensor. This system leads the robot to follow a corridor while it's detecting transitions for each door in autonomous mode. We demonstrate that this system is useful as low level navigator in a POMDP for indoor environments with a real robot. Some experimental results are shown. Finally, the conclusions and future works are presented.