A multi-agent-based activity recognition approach for ambient assisted living

Among the central challenges of Ambient Assisted Living (AAL) systems are autonomous and reliable recognition of the assisted person's situation and the proactive offering of adequate assistance services. In the context of emergency support, such situations can be acute emergencies or inconspicuous critical trends over a longer period of time. To enable the latter, reliable recognition of characteristic activities is necessary. In this paper, we explain our approach to activity recognition, based on a multi-agent design. In addition to the basic functionality of each agent, we describe the first experiences we made by integrating this reasoning approach into an exemplary Assisted Living laboratory.