Journal article
A Genetic Algorithm Approach to Motion Sensor Placement in Smart Environments
Journal of reliable intelligent environments, Vol.2(1), pp.3-16
04/2016
Handle:
https://hdl.handle.net/2376/109339
PMCID: PMC4957659
PMID: 27453810
Abstract
Smart environments and ubiquitous computing technologies hold great promise for a wide range of real world applications. The medical community is particularly interested in high quality measurement of activities of daily living. With accurate computer modeling of older adults, decision support tools may be built to assist care providers. One aspect of effectively deploying these technologies is determining where the sensors should be placed in the home to effectively support these end goals. This work introduces and evaluates a set of approaches for generating sensor layouts in the home. These approaches range from the gold standard of human intuition-based placement to more advanced search algorithms, including Hill Climbing and Genetic Algorithms. The generated layouts are evaluated based on their ability to detect activities while minimizing the number of needed sensors. Sensor-rich environments can provide valuable insights about adults as they go about their lives. These sensors, once in place, provide information on daily behavior that can facilitate an aging-in-place approach to health care.
Metrics
2 File views/ downloads
18 Record Views
Details
- Title
- A Genetic Algorithm Approach to Motion Sensor Placement in Smart Environments
- Creators
- Brian L ThomasAaron S CrandallDiane J Cook
- Publication Details
- Journal of reliable intelligent environments, Vol.2(1), pp.3-16
- Academic Unit
- Electrical Engineering and Computer Science, School of
- Grant note
- This material is based upon work supported by the National Science Foundation under Grants 1262814 and 0900781.
- Identifiers
- 99900547488401842
- Language
- English
- Resource Type
- Journal article