University of Twente Student Theses


A software solution for absolute position estimation using WLAN for robotics

Deen, Bart (2008) A software solution for absolute position estimation using WLAN for robotics.

[img] PDF
Abstract:This project is divided in two parts; Software frameworking for robots and absolute position estimation using the WLAN infrastructure. With increasing demands in functionality and complexity of a robot, the software complexity increases also. This increase in complexity stresses the limits of conventional software writing used today. To release this stress and give structure to the complexity of the software, a robust framework using a modular approach is needed. After a literature analyse of seven software architectures for robotics, Orocos matched best with the requirements defined in this project. Tests of Orocos’ implementation on the JIWY setup shows Orocos’ ability for hard real-time execution of tasks and communication between non real-time and hard real-time task. This without significant performance loss. Drawback of using Orocos are the resources needed. Orocos is a promising framework and should be implemented on amore complex system than JIWY to explore the real power of this framework Position estimation is a vital part in developing an autonomous mobile robot. Before a mobile robot can reach its destination it must know its current position. Some dedicated absolute position systems are developed, like GPS and Galileo. These systems work great when in sight of the satellites but are of no use indoors. This project aims to fill this gap and create amethod which uses existing WLAN infrastructure for indoor absolute positioning. Position estimation using propagation models is chosen instead of empirical models. Propagation models are better suitable on an embedded system where recourses are limited. Developing aWLAN positioningmethod results in a evaluation of four methods; (two radio propagation methods and two trilateration methods) in two different situations; moving from point to point for navigation purposes and stationary on a fixed point for orientation purposes. In the situation “stationary on a fixed point” the best result was obtained with the ITU path loss model and the linear trilateration method. Thismethod achieved mean error Ç 3m and can be used for orientation purposes. Orientation means in this case “Am I in the right room”, “Which direction is the nearest recharger” etc. Positioning when moving from point to point results in a position dispersion too great to be of any use. Navigation using the WLAN infrastructure in this setting is therefore not possible. Dedicated hardware, for measuring the signal strength, should result in more reliable measurement with less dispersion. This should result in better position estimations.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:53 electrotechnology
Programme:Electrical Engineering MSc (60353)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page