Kontextabhängige Unterstützung von Menschen mit Demenz und ihren Angehörigen

Konferenz: Technik für ein selbstbestimmtes Leben - 5. Deutscher AAL-Kongress
24.01.2011 - 25.01.2012 in Berlin

Tagungsband: Technik für ein selbstbestimmtes Leben

Seiten: 5Sprache: EnglischTyp: PDF

Persönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt

Mohamad, Yehya; Gappa, Henrike; Pullmann, Jaroslav; Nordbrock, Gaby; Velasco, Carlos (Fraunhofer FIT, 53757 Sankt Augustin, Germany)
Handte, Marcus; Wagner, Stephan (Universität Duisburg-Essen, Duisburg, Germany)
Schweda, Marcel (Assion Electronic GmbH, 53859 Niederkassel, Germany)

Caring for people with dementia in their home environment is a challenging task for family carers. Through web-based services family carers can be supported in fulfilling this task also helping to avoid feelings of being overburdened. Therefore a RFID-based service called Action Planner was developed as part of an AAL-assistant system to provide context-aware assistance for people with dementia and their family carers. The Action Planner allows family carers to describe rules that offer support either for the elderly person who is e.g. reminded to drink or the family carer who receives for example a text message in case a predefined event applies. The service operates context-aware which can be considered as its distinct feature. It is customizable by time- and localization-based rules. For instance, it is possible to configure the service in such a way that a text message is sent when constant wandering between bedroom and corridor is detected at night. The Action Planner relies on a custom indoor localization system that uses active RFID technology. The type of action and the time- and localization-based rules that need to be true in order to initiate an action can be specified using a rule editor. A first prototype of the Action Planner was evaluated in a lab test. Based on the results of the technical tests and the user study investigating usability issues of the rule editor, a second prototype is currently under development. The second prototype will enhance the Action Planer by algorithms for pattern recognition to improve reliability of identified movement patterns in the home environment.