Lokalizační systém pro mobilní robotické zařízení

Abstract

This thesis deals with the integration of localization module for existing autonomous mobile exploration vehicle. The vehicle is equipped with many sensors that are suitable for localization purposes, such as laser distance sensor, ultrasonic rangefinders, GPS and also WiFi module. The vehicle currently announces its location using GPS, which is usually inaccurate and unusable in buildings. A comprehensive localization system suitable for autonomous control, is still missing. The output of this work is probabilistic localization system based on Monte-Carlo method, which aggregates the data provided by the sensors, odometry, GPS modules, WiFi and calculates the vehicle's position in space. The developed module is modular and extensible robust system that provides location also in situation, that data from some sensors are not available. The work also deals with the use of WiFi access points for localization purposes and implements it as a one of sources for localization.