Room-level Localization of Individuals by Height for Smarthome Applications

Griffiths, Erin, Computer Science - School of Engineering and Applied Science, University of Virginia
Whitehouse, Cameron, Department of Computer Science, University of Virginia

The foundation for many intelligent building applications is the ability to track the movement of occupants for personalized environment conditioning and services. However, today there are no practical tracking solutions that are tolerant to the daily life of an actual home. Most solutions either require occupants to carry wearable devices or perform costly installations that are not practical in a residential home. By monitoring doorways, coarse grained room-level tracking information can be obtained on residents without hindering their daily lives. This thesis presents a doorway tracking system that can track residents in the noisy environment of their own home. Individuals are tracked based on their height, identifying when and in which direction they walk through doorways. The Screen Door System presented in this work outperforms other height based systems in an in-situ environment by 46% and can identify doorway crossings with 93% accuracy.

MS (Master of Science)
All rights reserved (no additional license for public reuse)
Issued Date: