Smart Home technologies may improve the comfort and the safety of frail people into their home. To achieve this goal, models of Activities of Daily Living (ADL) are often used to detect dangerous situations or behavioral changes in the habits of these persons. In this paper, an approach is proposed to build a model of ADLs, under the form of Hidden Markov Models (HMMs), from a training database of observed events emitted by binary sensors. The main advantage of our approach is that no knowledge of actions really performed during the learning period is required. Finally, we apply our approach to a real case study and we discuss the quality of the results obtained.

The Health At Home (HAH) is an alternative to the traditional hospital to promote the early discharge and to help patients and elderly people to live autonomously. This paper specifies and models an Integrated System (IS) devoted to the HAH management at the operational level. The IS is designed to monitor the daily living of the apartment inhabitant, detect the possible troubles and accidents, communicate with family, doctors and emergency services. A Petri net model in a modular approach is proposed, in order to effectively describe the actions and the activities of the IS.