See http://smartcondo.org/ for additional information and data sets.
People:
Julia Vlasenko, Masoud Vatanpour, Parisa Mohebbi, Noelannah Neubauer, Lili Liu, Dillam Diaz Romero, Eleni Stroulia, Ioanis Nikolaidis
Overview:
Recognizing the movements and activities of the home occupants is a basic functionality, underlying a variety of smart-home services and assistive-living services for seniors and people with disabilities.
In our work on the Smart-Condo project, we have, for a long time, been interested in inexpensive, non-intrusive, easy-to-deploy methods for observing and analyzing the activities of people at home and we have collected a rich data set from our recent study – see smart-condo.ca.
Objectives:
Our aim is to use these observations for recognizing interesting patterns and events, so that we may develop personalized supports for seniors, people with chronic conditions, and people experiencing mild to moderate cognitive decline.
Methodology:
This project is motivated by two methodological assumptions. First, we believe that accurate information about a person’s daily-living activities can provide rich evidence on the person’s abilities and functions; in particular, significant exceptions and changes in one’s daily-living routines may alert care givers of potentially risky events, or even the need for increasing long-term support. Second, we believe that such information can be obtained by unobtrusive ambient sensors deployed inside users’ homes, including infrared motion sensors, switches, pressure and temperature sensors, and, now, BLE beacons.
The design of this new release of the Smart-Condo platform includes different kind of sensors attached to different objects and surfaces in the house, collecting as much data as they can and sending them to the smart condo main server. Different sensors have different characteristics when it comes to receiving evidences from the environment. Some of them like motion sensors and switches can recognize that someone anonymous is within their sensing area, on the other hand, BLE beacons can differentiate the person. Some of them can just tell the existence of a person around while others can determine what the occupant is doing like cooking, watching TV or reading a book, at that specific time. All the data, coming from different sensors, are collected to a backend server to be used to infer each person’s location and activity. As mentioned before, and since different sensors have particular confidence on what they sense, inferences from various sensors can be combined and help each other for the system to come up with a more reliable picture of the environment where each inhabitant’s action and location is identified correctly. Finally, the Smart-Condo server generates textual reports and spatial visualizations for the movements and inferred activities of every occupant in any time interval, and warnings for special incidents which can be accessible by the person’s doctor or caregiver or anyone of his/her choice.