In this paper we investigate the problem of localizing a mobile device based on readings from its sensors utilizing machine learning methodologies. We consider a real-world environment, collect a dense set of 3110 datapoints, and examine the performance of a substantial number of machine learning algorithms. We found algorithms that have a mean error as accurate as 0.76 meters, outperforming other indoor localization systems. We also propose a hybrid instance-based approach that results in a speed increase by a factor of ten with no loss of accuracy in a live deployment over standard instance-based methods. Further, we determine how less dense datasets affect accuracy, important for use in real-world environments. Finally, we demonstrate that these approaches are appropriate for real-world deployment by evaluating their performance in an online, in-motion experiment. The Learning Based Indoor Positioning System (LIPS) Android application source has been made available on the web.
展开▼
