Hellmers, Hendrik and Norrdine, Abdelmoumen and Blankenbach, Jörg and Eichhorn, Andreas (2016):
IMU/ Magnetometer based 3D Indoor Positioning for wheeled Platforms in NLoS scenarios.
In: 2016 International Conference on Indoor Positioning and Indoor Navigation (IPIN) : 4-7 October 2016, Alcalá de Henares, Madrid, Spain, pp. 1-8,
Alcalá de Henares, Madrid, Spain, IEEE, 2016 International Conference on Indoor Positioning and Indoor Navigation (IPIN), Alcalá de Henares, Madrid, Spain, 4-7.10.2016, ISBN 978-1-5090-2425-4,
DOI: 10.1109/IPIN.2016.7743655,
[Conference or Workshop Item]

Abstract

In recent years the research on localization and navigation systems in GNSS-denied environments has been focused from both industry and research. Although many technologies based on e.g. UWB, WLAN, ultrasonic or infrared have been utilized, there is still no final solution for position and orientation determination in indoor areas. The fact, that applied signals in common approaches are influenced by fading and multipath inside buildings leads to restrictions on line-of-sight (LoS) conditions. In contrast, the ability of penetrating any kind of building materials qualifies magnetic fields to realize object positioning in harsh indoor environments. Hence, a DC Magnetic signal based Indoor Local Positioning System (MILPS) has been developed consisting of multiple electrical coils, representing reference stations. Based on the magnetic field intensities of at least three different coils, the corresponding slope distances and therefore the observer’s position can be estimated. Facing kinematic purposes a combination of MILPS and an Inertial Measurement Unit (IMU) has been applied, utilizing methods of sensor fusion. Observed inertial data – in this case three dimensional acceleration and angular rate measurements – lead to the sensor’s relative motion changes, which are processed by kinematic motion models. Based on a discrete integration with respect to the measurement time interval, the sensor’s current state – consisting of position, velocity and orientation – can be predicted. These highfrequency derived predictions are furthermore supported by external MILPS-distances and elevation angles utilizing methods of Kalman Filter. Focus in this contribution lies on the processing of both inertial data and magnetic field measurements for three dimensional applications.

Send an inquiry

Show editorial Details