Wearable Computing to Support Activities of Daily Living

This paper proposes an approach to determining an occupant’s indoor location through the use of machine vision techniques combined with wearable computing. Based on “off-the-shelf” machine vision tools a system is introduced to obtain a user’s indoor location through the detection of “reference” objects in their immediate environment. This information is subsequently cross-referenced with a knowledge base containing details of which rooms reference markers are located in. Details of the architecture required to realize the solution are presented which also accommodates for the fusion of information sources overcoming the heterogeneous nature of data gathered from multiple sources within the environment. The solution can be used to provide context aware assistance with Activities of Daily Living to those who may normally require assistance in their day-to-day life hence allowing them to live independently at home for longer.