My current research activities focus on industrial strength software technologies for Assisted Independent Living (aka Independent Assisted Living, aka Ambient-Assisted Living), leading to the development of distributed systems composed of sensors, actuators, computers, and other ICT devices. These are being used for a wide range of applications, which span from enabling elderly and disabled people to enjoy living in their own homes to monitoring security and safety of specific environments and contexts.
I am happy to discuss any kind of project in this application area; however, I do not supervise self-proposed projects that do not fit with my research & development interests and with my expertise.
Video analysis - a core topic: analysing video
streams from one or more wide-angle webcams positioned in a
room to infer the likely Activities of Daily Living performed
by the room occupier.
Environment and human monitoring can be effectively done using cheap,
standard off-the shelf WebCams in conjunction with smart image
analysis software that can quickly recognise movement and
efficiently categorise it.
There is substantial research
and development going on at the moment on tracking movements in
enclosed areas to understand exactly what is going on there. The
problem can be separated in two intertwined issues, namely (i)
understanding how somebody is moving in an area by processing a
video stream and (ii) understanding what kind of likely Actions
of Daily Living those movement are part of.
However, the situation becomes slightly more complicated
if movements must be tracked in a room where there is more than
one human performing actions, or where there is a human and a
pet. In this case, either extra devices (e.g., RFID tags) are
used to disambiguate between the different moving subjects, or
finer image analysis must be devised to discriminate between
them.
This project aims at developing real-time image
analysis method to track movements in an ordinary room for more
than one single subject and to learn customary movements, so that
their eventual instances can be easily recognised.
Note that this is a real project to be done in partnership with a software company that will require any and every generated IPR to be assigned to them. A video acquisition infrastructure is already in place, so a project of this kind would build on top of existing company technology. IP associated to any outcome to be owned by the company.
Image recognition and disambiguation: analysing video
streams from a wide-angle webcam positioned in a
room ceiling to recognise a person moving in a room or to disambiguate between two different people moving in the room.
This project derives from the above core topic and focuses on an unusual aspect: recognising people from the top rather than from the face.
Note that this is a real project to be done in partnership with a software company that will require any and every generated IPR to be assigned to them. A video acquisition infrastructure is already in place, so a project of this kind would build on top of existing company technology. IP associated to any outcome to be owned by the company.
The smart house on the cheap:
developing an infrastructure for controlling a wide range of
(electric, gas, and water) devices around the house to save money,
energy, and resources and to have a more comfortable ambient where
to live.
It is well known that Java was originally
introduced in the mid nineties with the idea of ambient (e.g.,
house) automation in mind. Hence, the language has a series of
features that make it easy to interact with a whole range of
suitably connected devices around the house. The idea of a
computer that organises and controls the whole house, from
heating to lighting to cleaning to food warming thus rapidly
comes to the mind.
Unfortunately, the point made at
the time was that smart appliances were needed to enable such a
vision. For a number of reasons (first and foremost the cost
and thus the market size for such sophisticated appliances),
smart appliances never materialised and smart house are a
reality only for a tiny minority of very wealthy people.
However, devising a smart house management system that
allows existing appliances and devices to be controlled and
operated through cheap, easily programmable dongles would make
the smart house concept a reality for everybody. Most
importantly, were such a system tailored towards the control of
energy- and resource- consuming devices around the house, thus
leading to sizeable savings in running costs, it would quite
certainly become a commercially interesting product.
This project aims at developing (i) a widely applicable
software solution (architecture and proof of concept) to the
problem of controlling a wide range of house appliance and
devices (ii) a method that uses cheap, off-the-shelf
programmable dongles to control and operate such
appliances/devices.
Note that this is a real project
to be done in partnership with a software company that will
require any and every generated IPR to be assigned to them.
Ambient Scenario Modelling and Analysis Mark III:
refining and engineering an intelligent system for modelling and
analysing behaviours, personal lifestyles, and ambient scenarios.
Assisted Independent Living systems aim to enable elderly and
disabled people to keep on living in their own homes for as long as
possible by deploying a range of sensors that monitor the safety of
the living environment and the personal well being of the resident,
seamlessly building at the same time a lifestyle profile of the
resident based on individual behavioural patterns.
To do so, behavioural patterns composing lifestyle profiles and
also possible living scenarios have to be captured, modelled and
analysed by matching them with a knowledge base of existing 'safe'
scenarios for every particular person in a specific living
environment. A prototype of such a system already exists but needs
to be extended, further industrialised, and made smarter by
assessing the impact of technologies such as Jena and its possible
current limitations.
This project aims at (i) extending (ii) reengineering and (iii)
creating a rich ontology of behavioural patterns and scenarios for
elderly and disabled people living alone.
Note that this is a real project to be done in partnership with a
software company that will require any and every generated IPR to
be assigned to them.
Some projects are more methodological in nature, others are more implementation-oriented. However, to foster the professional development of students, we require each project to include the following three equally weighted components:
This requirement obviously excludes some potentially interesting projects that lack at least one of the listed components. Moreover, I do not propose projects whose preliminary risk analysis highlights major uncertainties that could affect both the quality and the size of the project outcome. The above project list is by no means an exhaustive one, and I do encourage students to proposed modified or additional project topics in my interest area other than those explicitly listed.
Back to Ernesto Compatangelo's personal page
Top of this page | Computing Science | University of Aberdeen
Page last updated 21 October 2009
Optimised for
