Experimental Evaluation of the Forkbeard Ultrasonic Indoor Positioning System

Abstract

Indoor positioning systems are crucial to provide location-based services in areas that are not covered by a Global Navigation Satellite System. Among the many technologies applied to this field, ultrasound has emerged as a potential low-cost, high-accuracy approach to positioning based on trilateration. Several ultrasonic systems have been proposed over the years. Of these, academic systems are typically prototypes that are unavailable to the public, whereas commercial systems generally do not provide characterization test results. In this work, we have conducted a detailed characterization study of a commercial indoor positioning system for smart devices developed by Forkbeard Technologies AS. We tested the system under static and dynamic conditions in a motion capture lab of approximately 150 m2. We considered different room occupancies, beacon configurations, and device positions. The results, given in terms of 2-D absolute errors at different confidence levels, show great variation depending on the aforementioned conditions. The worst case scenario corresponds to a pedestrian in motion in an office setup with four beacons, with the phone placed in the pocket. In this case, 80% of the errors were below 143 cm. The best results were obtained under static conditions using ten beacons, for which 80% of the errors were below 44 cm.