MagRef4PIL - Magnetic Reference for Accurate Indoor Tracking

Indoor navigation is an important enabling technology for applications such as navigation aid, location-based information and advertisement, as well as to provide enhanced emergency response.

Several approaches have been proposed in recent years but none of them was able to achieve meter-level accuracy. More recently, a new approach has emerged as a solution for the accuracy issue. This approach relies on inertial sensors to track a user by continuously estimating displacement from a known position. Most recent smartphones are equipped with several sensors (e.g.: accelerometer, gyroscope, magnetometer) that provide the necessary hardware for this indoor navigation systems based on inertial sensors.

As previously stated, these indoor navigation systems, based on inertial sensor, define the actual position estimating displacement from a previously known position. On the other hand, these systems rely on low-cost sensors that can be extremely noisy. Noisy sensors will introduce position errors due noise itself but also drift (from gyroscope) and magnetic interference (from magnetometer). Eventually, these cumulative errors will lead to an erroneous estimation current position.

In order to correct positional errors, several solutions for aided indoor navigation have been proposed, e.g.: GPS (Global Positioning System) signals (when available), Wi-Fi fingerprinting, light patterns, structural magnetic fingerprinting, RFID (Radio Frequency Identification) check-points.

This MSc topic goal was to develop magnetic position references for an accurate indoor tracking system.

The main ideas was to study and develop a set of ultra-low frequency antennas that can interact with compass enabled phones in order to give them accurate references ,with less than one meter precision.

This project was a follow-up from the previous work developed under Ultra Low Frequency Magnetic Communication (ULF-MC), a wireless communication solution based on artificial magnetic references.

Two important results were expected from this work. On the one hand, theoretical models for magnetic elements should be investigated in order to develop a systematic procedure to project and build new types of magnetic antennas (different shape and size). On the other hand, an algorithm that is able to provide direction and orientation on human movement based on artificial magnetic fields should also be developed.

The aforementioned results were partially achieved and the core concepts are being considered in the on-going work developed under Fraunhofer AICOS’ Indoor Location related projects. Include the outcomes on other Indoor Location related project was also one of the most important outcomes for this MSc topic.

Finally, a proof of concept was implemented. The developed algorithms resulted in a new solution that was able to provide information on the heading, in order to correct Indoor Location system’s misplacing errors, despite the fact that the results were not so accurate as expected.

 

For any additional information regarding this project, please contact us using the inquiries form.