Dissertation
Behaviometrics for Multiple Residents in a Smart Environment
Doctor of Philosophy (PhD), Washington State University
01/2011
Handle:
https://hdl.handle.net/2376/2855
Abstract
Smart homes and ambient intelligence show great promise in the fields of medical monitoring, energy efficiency and ubiquitous computing applications. Their ability to adapt and react to the people relying on them positions these systems to be invaluable tools for our aging populations. This work introduces and explores solutions for issues surrounding real world multiple inhabitant smart home situations. Dealing with multiple residents without requiring wireless tracking devices, while paying heed to privacy concerns, is a difficult proposition at best.
The Center for Advanced Studies in Adaptive Systems research group has developed and tested a number of novel technologies to address the issues of multiple inhabitants within a smart home context using inexpensive, low profile, privacy sensitive sensors. These smart home implementations, when combined with artificial intelligence tools, are designed to provide localization, tracking, and identification through behaviometric approaches that are useful and deployable in real world situations. They have been evaluated using unscripted living spaces with multiple residents, and their capabilities explored as a means of benefiting other modeling tools, such as detecting the Activities of Daily Living.
Given the complex nature and diverse needs of smart home technologies, the tools presented here are by no means definitive solutions to handling multiple resident smart environment situations. However, they do provide a strong working base for the continued development of smart environments with demonstrable benefits on real-world implementations.
Metrics
5 File views/ downloads
52 Record Views
Details
- Title
- Behaviometrics for Multiple Residents in a Smart Environment
- Creators
- Aaron S. Crandall
- Contributors
- Diane J Cook (Advisor)Lawrence B Holder (Committee Member)Behrooz Shirazi (Committee Member)
- Awarding Institution
- Washington State University
- Academic Unit
- Electrical Engineering and Computer Science, School of
- Theses and Dissertations
- Doctor of Philosophy (PhD), Washington State University
- Number of pages
- 271
- Identifiers
- 99900581661301842
- Language
- English
- Resource Type
- Dissertation