Modeling and Learning for LaCasa, the Location And Context-Aware Safety Assistant

Wandering is a common behavior among people with dementia (PwD). It is also one of the main concerns of caregivers since it can cause the person to get lost and injured. The frequency and manner in which a person wanders is highly influenced by the person’s background and contextual factors specific to the situation. In this paper we investigate some of the properties of wandering behaviours, particularly related to our ability to sense them with mobile devices. We then present a novel decision-theoretic model that estimates the risk faced by the PwD and decides on the appropriate action to take, such as prompting the PwD or calling the caregiver. The model can be tailored to the user needs (e.g. known locations, level of cognitive decline) takes into account uncertainty, seamlessly merges data from different sensors, reasons about the relative worth of sensor readings (and so can be used for active sensing), and learns about a person’s patterns of behaviour. Contextual information gathered from sensors is seamlessly factored into the model, including current location, battery power, and proximity to the caregiver. The system can reason decision theoretically about the costs of sensors (e.g. battery charge) and the relative costs of different types of assistance. A preliminary version of the system has been instantiated in a wandering assistance application for mobile devices running on an Android platform.