A Smartwatch Software Architecture for Health Hazard Handling for Elderly People

A software architecture for smart watches supporting and securing the everyday life for the elderly is presented. The approach starts from the insight that the usage of smart watches for care-giving and handling personal health hazards is a best practice driven, empirical discipline, requiring a declarative, easy to maintain software description for the smart watch devices. The second rationale for the approach presented is the decomposition of the complex interrelations of "health hazards" into definable modules of manageable size, each addressing the handling an individual health hazard. The proposed, hierarchical architecture builds on the automatic detection of the activities and events of daily life (ADLs, EDLs) by the sensors of the smart watch. Handling of health hazards will be modeled by UML state machines, which utilize the detected ADL/EDLs for their state transitions. The architecture has to take account for a plenitude of relevant hazards, managed by a multitude of simultaneously operating state machines, via a central synchronizing control mechanism.