Assisted cognition: compensatory activity assistance technology

Abstract

The predicted increase in the number of elderly members of the industrialized world suggests an associated increase in the number of people who are going to be diagnosed with various forms of cognitively disabling dementias, including Alzheimer's disease.A possible solution to the cost of caring for these members of society is to augment our care network with sophisticated cognitive aids which can compensate for simple cognitive errors.In this thesis I present the design and implementation of two cognitive aids. The first is an outdoor navigation assistant and the second is an indoor household activity monitor. Both systems are characterized by a fusion of sensor data with background knowledge and are interpreted in a probabilistic framework.The result of this work is a demonstration of the feasibility of developing cognitive aids based on real-time streaming sensor data.