Capturing activities of daily living for elderly at home based on environment change and speech dialog

Abstract

The ICT-based elderly monitoring systems attract great attention as a promising technology for home elderly care. However, the conventional systems have limitations of deployment cost and invasiveness, the effort of activity labeling, and a lack of communication. To cope with the limitations, we propose a system that captures activities of daily living (ADL) of the elderly, based on speech dialogue triggered by environment changes. Specifically, we deploy Autonomous Sensor Boxes, developed in our previous study, within a house of the elderly. The boxes gather and send house environmental data to the cloud. Then, the Change Finder algorithm is applied to the time-series data, to detect changes in the house online. On detecting a change, the Virtual Agent (VA) in the house asks the elderly what he/she is doing now. The elderly speaks to the VA, by which an ADL is recorded in the system. The proposed system can capture ADL with non-invasive sensing and create an opportunity for communication.