AMBIENT INTELLIGENCE
“Ambient intelligence (AmI) refers to electronic environments that are sensitive and responsive to the presence of people. In an ambient intelligence world, devices work in concert to support people in carrying out their everyday life activities, tasks and rituals in easy, natural way using information and intelligence that is hidden in the network connecting these devices. As these devices grow smaller, more connected and more integrated into our environment, the technology disappears into our surroundings until only the user interface remains perceivable by users.” (Wikipedia)
In this context, my research is focused on analyzing and modeling human behavior in order to reduce the awareness gap between devices and human. The research is focused on three aspects of human behavior: micro-behavior – the way a person performs movement, e.g. walking, standing up; meso-behavior – the way a person performs some activity, e.g. preparing lunch, morning routine; and macro-behavior – the way a person behaves during a day, week, month, e.g. a usual working day. The results are applicable in security applications, remote health care, sport analysis, personalization, etc.
The Confidence Project
The Confidence project is an EU FP-7 project. The main objective is development of an ubiquitous care system to support independent living of elderly people. The goal is to develop and integrate innovative technologies for the detection of abnormal events (such as falls) or unexpected behaviors that may be related to a health problem.
We are developing intelligent methods for monitoring elderly and detecting abnormal events. A video of the first system prototype is available on YouTube:
CIVaBiS
CIVaBiS was founded by Slovenian Ministry of Defense and Slovenian Research Agency, and carried out by Department of Intelligent Systems, company ŠPICA International d.o.o. and Faculty of Electrical Engineering, ULJ. The project aimed to improve security at an access point and reduce the risk of possible intrusion, for example, robbery, terrorist attack, etc.
We developed an intelligent access control system heavily relaying on intelligent methods. It upgrades classic biometric control with an arbitrary number of intelligent modules. The system is capable to take into account various access points with different sensors. A short demonstration is presented in the attached video below.
SEARCH PATHOLOGY
Computer games based on the minimax principle usually produce better results when searching deeper. When researchers attempted to explain this formally using mathematical models, they found that under seemingly reasonable conditions minimax behaves unexpectedly: deeper search produced worse results – just the opposite of what happens in practice. This phenomenon was termed minimax pathology.
In my BSc thesis, supervised by Ivan Bratko, Matjaž Gams, and Mitja Luštrek, I investigated conditions using synthetic game trees as well as real game.
