No abstract

In our aging society, the elderly's mobility has become critical for our collective prosperity. However, the elderly's mobility is limited in the current built environment due to various types of environmental barriers. Manual surveys have been conducted to detect such environmental barriers, but they are discontinuous, invasive to the elderly's daily lives, and labour-intensive. As such, these methods are not ideal for wider adoption. To continuously, less-invasively, and less-laboriously detect the environmental barriers and advance the elderly's mobility, this study proposes a wearable-based collective sensing approach. This approach measures collective stress, the stress commonly sensed from multiple people on a location, as an indicator of environmental barriers based on people's physiological and location data collected by wearable sensors. To test the feasibility of the proposed approach, a "collective stress metric" is suggested. Then, the values of the collective stress metric on locations of the test site were calculated based on the physiological and location data collected from 10 elderly subjects' daily trips for 2 weeks. Then, every location on the test site was categorized into locations "with environmental barriers" and "without environmental barriers" through site survey. Based on the collected data and results of site survey, the collective stress was statistically compared between locations with environmental barriers and without barriers. The result showed that the collective stress was statistically higher on locations with environmental barriers than without barriers. The results demonstrated that the collective stress has indication of environmental barriers, therefore, the proposed approach is feasible to detect the elderly's environmental barriers.