A Guardian Angel For The Extended Home Environment

(GUARANTEE)
G. Papadourakis1, J. Lin Manzaneda2 and M. Sinadinakis1
1

Department of Applied Informatics and Multimedia, Technological

Educational Institute of Crete, Heraklion, Greece, e-mail: papadour@cs.teicrete.gr
2

Polytechnic School of Crdoba, University of Crdoba,

Crdoba, Spain, e-mail: jlinan@uco.es

Abstract
The main purpose of this paper is to introduce the European Project entitled A Guardian Angel For The
Extended Home Environment (GUARANTEE) which is supported by the Information Technology for
European Advancement (ITEA2). The goal of GUARANTEE is to research software products and services that
provide personal safety in the residential environment. These products and services provide direct support and
advice to people in unsafe situations, or they connect people and enable the support of others. The project will
investigate the consumer needs and the commercial opportunity for this type of product. Further, the project will
develop the software components and personal safety services as well as the system and software architecture.
GUARANTEE will develop 3 demonstrators: for childrens safety, elderly safety, and home security. The
demonstrators integrate components for sensing, decision making, and alerting of users. The demonstrators
comprise both an in-home and networked system for commercial or community services. This paper will be
organized as follows: Initially, an introduction to Smart Homes and background information will be presented
followed by a detailed description of the project; the various possbile senarios to be implemented as well as the
various available sensors for these senarios will then be outlined. Finally, possible decision support techniques
will be discussed.

Keywords
1. Introduction
Statistics show that many personal accidents happen in the residential environment.
As much as 6% of the European population requires hospital treatment after a home or leisure
accident each year. These accidents lead to great personal grief, loss of productivity and
medical cost. Many adults are responsible for caring for their young children or elderly
parents, an extremely demanding situation when combined with a busy working and social
life. Humans in general feel a strong need to connect to loved ones and to be able to provide
support when needed; connectedness provides peace-of-mind.
In the World Health Organization (WHO) European region report of 2008, unintentional
injuries are the leading cause of childhood death for children over 11 years of age, resulting in
830,000 deaths per year and millions of injuries for this specific population. The WHO report
on World Report on Child Injury Prevention, provides he following top five causes of
unintentional injuries: RTI (Road traffic injuries), drowning, burns, falls and poisoning [1].
The share of older people in all fatal injuries in the EU has been steadily rising by

approximately 1% every two years (from 40% in 1997 to 49% in 2006). If the rate of nonfatal injuries for elderly people does not decline, falls will be a significant contributor to the
increasing dependency on care in the near future. Every year 100,000 elderly (>65) citizens of
the European Union (EU-25) die as a result of the most preventable cause of death, namely
injury. Despite the fact that they constitute about 20% of the population, they contribute to
almost half of the total number of deaths due to unintentional injuries [2].
Inexpensive support systems for elderly people and children staying alone at home, allowing
care and health centers to remotely observe and help them, are becoming feasible with today's
technology.
In the 1990s, Gann and Venables (Gann, D., Barlow, J., & Venables, T.) had been centrally
involved in the provision of demonstration Smart Homes in Edinburgh and York in
conjunction with Edinvar Housing Association and The Joseph Rowntree Trust. The major
difficulty identified in the development of Smart Homes in the UK appeared to be the lack of
completed projects where by data could be acquired and analyzed. Stemming from Gann's
work, the author identified a number of networks and demonstration projects claiming to meet
the needs of Smart Home developers. This study aspires to provide additional data to facilitate
and expedite the much-needed development of Smart Homes, particularly for disabled
persons.
In 1991, Ad van Berlo established an advice bureau for assistive technology in the care sector.
He switched from the world of medical technology (cure) to that of care technology (care).
The field of 'Gerontechnology' was strongly emerging. The latter is technology aimed at
enhancing the lives of the elderly through improved comfort and ease. Eindhoven University
of Technology piloted gerontechnology in the 90's as a new research area, with living or home
technology as one of the main sub-disciplines. Eventually, this became widely referred to as
home automation. In the early 90's, most people could not yet imagine what home automation
would entail. However, in the following years, various projects were initiatied which would
enable the elderly to experience the benefits home automation could offer.
In 1993 Smart Homes started its efforts to promote the development and use of home
automation. Both professionals and interest groups of consumers can take advantage of Smart
Homes services. Most of the activities aim at the collection, exchange and distribution of
knowledge and experience.
In 1998, Corien van Berlo, Ad's partner, setup Smart Homes together with her husband. The
aim was primarily to further the promotion of home automation, execute demonstration
projects and start experiments. To this end, Smart Homes provided support to all Brabant
pilots as commissioned by the North Brabant province. These demonstration projects were
delivered in 2000 and 2001, yet these projects remain significant today because the necessity
of a system integrator's participation was clearly demonstrated. Consequently, the supply of
technology offering more efficient and affordable solutions has emerged as a prolific area of
research and development.
In the future, it is likely that people will be required to work outside the home more frequently
than today. Research has showed that demographical changes are taking place in most
European countries, in addition to U.S.A and Japan. Technologies such as sensors, ambience

technologies, etc., will help provide the autonomy to elderly persons to remain for a longer
period of time in their homes. This autonomy will therefore meet many of their day to day
demands, thereby enhancing their quality of life and psychological wellness.
The goal of GUARANTEE is to research software products and services that provide personal
safety in the residential environment. These products and services provide direct support and
advice in unsafe situations, or they connect people and enable the support of others. The
project will investigate the consumer needs and the commercial opportunity for this type of
product. Further, the project will develop the software components and services for personal
safety as well as the system and software architecture.
Existing home safety products address only basic needs. The introduction of software-based
home safety solutions may open the door to many novel methods of delivering real
innovations to consumers. Software home safety products leverage the capabilities of existing
sensor components, connectivity, and communications infrastructure. The project will
produce the signal processing and decision making algorithms for specific home safety
situations. Further, the project will develop the software architecture for in-home safety
systems and for commercial and community-based home safety services, in which user
interaction acts as a key element.
GUARANTEE will develop 3 demonstrators: for childrens safety, elderly safety, and home
security. The demonstrators integrate components for sensing, decision making, and alerting
of users. The demonstrators comprise both an in-home and a networked system for
commercial or community services.

2. The GUARANTEE Project

PHILIPS leads the GUARANTEE consortium, which is composed of leading and
complementary partners originating from several European countries. The GUARANTEE
consortium consists of large companies (e.g. Forthnet and Philips) and several universities
(e.g. the universities of Madrid and Kaunas) as well as research centres (VTT and ESI). This
consortium includes world-class complementary partners with the required expertise in the
entire added value chain, from research to manufacturing, business roles and market coverage.
GUARANTEE will develop technology to address the specific personal safety needs of
individuals in the residential environment. GUARANTEE is far more advanced than existing
home safety techniques. The goal of GUARANTEE is to significantly enhance the personal
safety of people in the home environment by introducing an active safety product (or safety
system) that is functionally enhanced by a networked service. In a typical set-up as depicted
in the Figure 1, a central station evaluates specific elements of the safety situation at a certain
location or of certain people. The central station builds an awareness of the safety situation on
the basis of sensor input provided by several multi-modal monitoring devices. When the
station detects a safety event, it will induce a local action. Networked service may then
support the station.

Figure 1 General idea of GUARANTEE

The GUARANTEE home safety system comprises sensors and signal processing (e.g.
audiovisual, ECG, and body temperature), decision making processes, user interface, and
external services. The diagram of the Figure 2 provides a technical overview of the
GUARANTEE system:

Figure 2 GUARANTEE Architecture

3. Use cases and business interests

The use-cases defined by a process of merging and refinement fit with the domains of young
children and elderly persons, both in and around the home, as well as those nomadic

situations that pose great risk for injury. For elderly the GUARANTEE Project is focused in
the following three primary use-cases:
Visit of unexpected swindlers
Related to common behavior of elderly persons
Falling on Stairs
Family monitoring
Related to security risks
Fire risk and Gas leakage risk in the cooker
Water leakage risk in the house
Checking the door of the fridge and the state of the food inside
Power outages
Privacy enhanced tracking
For young children, the use-cases are classifieds depending of the age range and the
existence of special needs:
Infant
Baby Camera
Baby monitor in order to prevent Sudden Infant Death Syndrome (SIDS)
Toddler
Unaccompanied toddler (protects children of running on the a street alone, ...)
Home safety for toddler in risk areas at home (fireplace, kitchen stove, ...)
Child safety tutored by elderly people
School age
Illumination for school safety and energy saving
Playing field monitoring
Child in traffic
Watching unattended children
Therapy-wellness devices for disabilities

4. Data acquisition
To meet the needs of a multifaceted and diverse environment such as the home, the
GUARANTEE project includes a variety of essential sensors. In particular, requirements for
the format of the data and the description of certain events, control of sensor and the network,
privacy, reliability and response in real-time, needs for collecting information, price, etc., are
addressed:
Tools related to home Automation:
Fire Sensor
Water Sensor
Home Temperature Sensor
Home Security System Sensors: PIR sensor (Passive Infra Red), glass
break detection, access control (contactless reader), finger print readers, ...
IP-based Display devices: IPTV, IP Phone, etc.
Cameras (Visual Tools)
Telemedicine sensors:
Heart Rate Monitor

Oximetric
Infection Control POC Terminal
Activity Monitor
Body Pressure sensor
Fallen Sensor
Aggresion Detection Devices: (Sound Intellegence)
Energy Monitoring
Indoor / Outdoor Localization Devices: GPS and UWB
Vision-based Sensors:
People Tracking and Behaviour Analysis
Vision-based Identification
Noise Sensor (Sound Intelligence)

5. Decision support
For each sensor, one or more algorithms are used to interpret the sensors outputs, providing a
semantic interpretation. The input function is intended to enable a consideration of events,
status or situations independent from technical constraint, allowing further common
reasoning. One method to achieve this objective is the development of an event ontology that
is related to the GUARANTEE scenarios.
In GUARANTEE the output of several sensors will be combined to make decisions.
Therefore, a way to combine sensor outputs (aka sensor fusion) is needed. Data fusion
techniques combine data from multiple sensors, as well as related information from associated
databases, to achieve improved accuracies and more specific inferences than possible via a
single sensor alone [3][4][5]. There are several methods to achieve the data fusion, the most
popular of which include: Linear Evidence Combination, Classical inference and Bayesian
inference method, Dempster-Shafer Theory, Voting Fusion Method, Fuzzy Logic Method or
Neural Network Method.
Sense making functions are used in the context of home safety. GUARANTEE covers the use
of expert systems, neural networks, pattern recognition, evolutionary systems, fuzzy logic,
ambient intelligence and dimensionality reduction. [6][7]

6. References
[1] World Health Organization web site (2010), World report on child injury prevention,
http://www.who.int/violence_injury_prevention/child/en/, (Accessed 8 June 2010).
[2] Leiden University web repository (2009), Physical Activity and Falls in Older Persons,
https://openaccess.leidenuniv.nl/bitstream/1887/13471/1/proefschriftWijlhuizencompleet.pdf,
(Accessed 8 June 2010).
[3] Waltz, E. (1986). Data Fusion for C31: A Tutorial. Palo Alto: EW Communications.
[4] Llinas, J., Waltz, E. (1990). Multisensory Data Fusion. Boston: Artech House.

[5] Hall, D. (1992). Mathematical Techniques in Multisensory Data. Boston: Artech House.
[6] Arbib, M. (2003). The Handbook of Brain Theory and Neural Network. Massachusetts
Institute of Technology.
[7] Rabunal, J. R., Dorado, J. (2006). Artificial Neural Networks in Real-Life Applications.
Spain: Idea Group Inc