2012 | OriginalPaper | Buchkapitel
Smart Meter: Detect and Individualize ADLs
verfasst von : Jana Clement, Joern Ploennigs, Klaus Kabitzsch
Erschienen in: Ambient Assisted Living
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Smart meters offer a low-cost, simple, and unobtrusive method to design monitoring systems for the detection of emergency situations in homes. The aim of this paper is to automatically detect the activity (Activity of Daily Living, ADL) of occupants based on such data. The approach trains a Semi-Markov-Model that describes the daily use of appliances, such as domestic appliances, sanitation, and heating system. Human habits are detected within the Semi Markov Model and rated by a similarity measure that calculates the probabilities for the ADL executed by the occupant based on the appliances he uses. This rating influences the probability for the ADL and permits model adaptation to individual user behavior. Considering the timeline of appliance usage, the probability estimations of each model state allow individualized and situation depending conclusions about the ongoing activity of the occupant.