Rule-based Location and Activity Recognition Based on Environmental Sensors
Name
Andreas Baum
Abstract
It is useful to extend the health care system to home environment in order to reduce the healthcare costs. One solution for that would be to create a sensor system that provides helthcare professionals with necessary information for assisting patient's treatment. SPHERE project has developed an experimental building for mentioned purposes that is equipped with sensors. The aim of this bachelor’s thesis is to develop an automatic system that is capable of recognising human locations and activity using rule-based approach. The developed system is able to detect successfully all the rooms, where the patient has been with error less than two seconds most of the time. Room sensors don't give many opportunities for recognising activities, hence the system is capable of recognising a few of them while being mistaken less than ten seconds.
Graduation Thesis language
Estonian
Graduation Thesis type
Bachelor - Computer Science
Supervisor(s)
Meelis Kull
Defence year
2018