INTELLIGENT SYSTEM FOR ASSISTING ELDERLY PEOPLE
AT HOME
Juan Pascual, Miguel A. Sanz-Bobi and David Contreras
Computer Science Department, Engineering School, Pontificia Comillas University, Alberto Aguilera 23, Madrid, Spain
Keywords: Multi-agent system, risks at the homes of elderly people, intelligent system, human activities monitoring.
Abstract: Nowadays the number of elderly people in our society is increasing thanks to continuous and important
advances related to health care. The ideal situation for these elderly people is to spend as much time as
possible enjoying their life within their family and social environment, without the need to abandon their
homes to go live in specialized centres as long as their physical health permits. This paper describes an
architecture of pro-active intelligent agents in which the main objective is to extend the amount of time as
much as possible that these elderly people can reside in their own homes by means of providing continuous
vigilance of certain parameters concerning daily activities which possibly could be risky by facilitating
reminders to complete specific tasks, by easing communication with the exterior world, and in the case that
it is necessary, by automatically calling for emergency services.
1 INTRODUCTION
At present an important demographic change is
taking place regarding the mean age of our society.
The increase of the population aged 60 years or
older is very significant. This is a world wide
phenomenon as observed by the United Nations
(United Nations, 2006). In Europe the number of
new born children is less than it was 20 years ago
and life expectancy is much longer and in
consequence the ageing effect of the population is
even more important.
When elderly people are healthy, but with typical
problems due to ageing, the best recommendation is
that they spend as much time as possible enjoying
their life within their family and social environment,
without the need to abandon their home to go live in
specialized centres as long as their physical health
permits. In many cases, simply because there do not
exist sufficient means or vigilance time by a
caretaker in their home, these people are forced to
leave their homes to go live in specialized centres
such as nursing homes.
In order to try to solve this problem, several
organizations and companies offer teleassistance
services to elderly people at home. The most part of
this type of services is based on the demand of the
user (elderly person) by simply pushing a button on
a small device that he/she carries on him/her
(Aguilera, 2003). A specialized call centre attends
any request from the user at any time, and also, the
call centre can contact the user periodically in order
to know that everything is going well. All these
services are very helpful, but they require inputs
based on human decisions, the user or people
attending the call centre. It seems that a further step
is needed in order to try to monitor some daily
activities of these elderly people. There are
important advances in the use of new information
technologies for monitoring some activities of
elderly people at home (Jih, 2006), (Fishkin, 2005),
for assistance to find the way, if one is disoriented
(Liu, 2006) and to monitor some important
biological parameters (Pollack, 2005). Also, some
efforts have been developed in the elaboration of
technological platforms able to integrate different
kinds of services of remote assistance to elderly
people (Robocare, 2007), (Hill, 2005), (Attentianet,
2007). This paper is in line with these examples of
research but with the aim of preventing possible
risks for elderly people at home carrying out their
daily tasks.
This paper is organised with the following
sections. The first section presents the objectives of
SIAM, the next section presents its main strategy,
the following two sections describe the agents and
the implementation of SIAM, and finally, an
example is provided.
159
Pascual J., A. Sanz-Bobi M. and Contreras D. (2008).
INTELLIGENT SYSTEM FOR ASSISTING ELDERLY PEOPLE AT HOME.
In Proceedings of the First International Conference on Health Informatics, pages 159-163
Copyright
c
 SciTePress

2 SIAM OBJECTIVES
This paper describes a multi-agent system named
SIAM, which in Spanish stands for intelligent
system for the assistance to elderly people at home.
Its main objective is to contribute to extending the
amount of time as much as possible that elderly
people can reside in their own homes with automatic
and intelligent assistance based on new information
technologies. This is reached by:
• continuous vigilance of certain variables that
could be important for risk detection in daily
activities at home
• facilitating reminders to complete specific tasks
• easing communication with the exterior world
• in the case that it is necessary, automatically
calling for emergency services.
SIAM is based on a set of intelligent agents of a
multi-agent system capable of integrating the
previously mentioned pro-active assistant services
for elderly people as they carry out their usual
activities at home.
The scope of the current version of SIAM is
limited to the agent architecture and their
relationships. It was decided to not install SIAM in a
real environment in the house of a elderly person
without an intensive testing phase of the software
developed in order to prevent possible unnecessary
disturbances. Furthermore it should be noted that
this version of SIAM does not include interaction
with physical sensors. The information coming from
the sensors is obtained from a simulation
environment which represents the main rooms of a
house and the sensors that are installed in each one.
A direct input over a sensor in the simulation
environment can be used to switch its status on/off.
3 SIAM STRATEGY
The SIAM strategy is conceived as a pro-active set
of intelligent services able to help elderly people at
home.
These intelligent services are the following:
a. Detection of possible actions or situations in the
house that could be a risk for the elderly person, and
trying to prevent it. Examples of possible risk
situations are: a sudden fall of the person on to the
dining room floor, unattended gas open in the
kitchen, unattended water running in the bath, etc.
b. Prevention of other types of risks related with
forgetting things such as medications, an
appointment with the doctor, a payment of some
important bill, etc.
c. Facilitation of the communication between the
elderly person and the external world to his/her
home: contact with the caretaker, with family, with
the CMD (Central Monitoring Department), etc.
These services can be accessed by several actors
with different roles. A strategy based on a multi-
agent system (Weiss, 2000), (Wooldridge, 2002)
was chosen in order to fulfil all the requirements of
these services.
The following are the four types of actors: the
user or elderly person, the CMD, the caretaker and
the virtual caretaker or intelligent system.
SIAM has to cover different roles through its
different agents for interaction with the different
actors. The main roles to be covered are the
following:
• Communication to and from the elderly person at
home.
• Communication to and from the caregiver of the
elderly person, if such a person exists.
• Communication to and from the CMD.
• Collection of information coming from sensors.
• Intelligent analysis of the information collected
in order to predict a possible risk for the person
and to issue the corresponding actions.
Intelligent
system
CMD
User
Caretaker
Figure 1: SIAM actors and relationships.
Figure 1 shows the design logic of SIAM based on
the interaction among actors according to the roles
of each one. Usually, SIAM collects information
from sensors installed in the home of the elderly
person in order to detect if a possible risky situation
is produced by an action or event when he/she is at
home carrying out his/her activities. In the case that
some anomaly or risk is detected, SIAM will first
try to contact the elderly person and, if an answer is
not received, this will be notified to the CMD and/or
to the caretaker. Also, from Figure 1, it is possible
to observe that the user can activate a request to the
CMD and the caretaker and vice verse.
HEALTHINF 2008 - International Conference on Health Informatics
160

4 SIAM AGENTS
SIAM contains the following type of agents:
• USER. It is in charge of all the communications
from SIAM to the person and from the person to
SIAM. This agent is located in a mobile device
that the person carries on him or her. At the
moment this agent is installed in a PDA.
• DATA COLLECTOR. This agent is in charge of
the collection of information from key places of
the elderly person’s home in order to know if
some particular activities are occurring which
cause a certain risk for the person. These agents
are located in the house being monitored.
• HOME CONTROLLER. It is in charge of the
integration and pre-analysis of all the
information collected by the Data Collector
agents in order to obtain a global view of
activities in the house. This agent is located in
the house being monitored.
• DIAGNOSTIC. This agent performs a diagnostic
of possible risks in the house of the elderly
person according to the information collected
and specialized knowledge previously stored in a
knowledge base. The structure of this agent is
similar to an expert system. This agent is located
in the Central Monitoring Department
responsible for taking care of a group of elderly
people using the SIAM platform.
• CARETAKER. Its objective is taking care of all
the communications from SIAM to the caretaker
of the elderly person and from the caretaker to
SIAM. This agent is located in a mobile device
that the caretaker carries on him or her. At the
moment this agent is installed in a PDA.
USER
DC 1
DC 2
DC n
HC
DIAGNOSTIC
CARETAKER
CMD
Caretaker
Elderlyperson’shome1
Elderly person’s home 2
Elderly person’s home n
DC Data Collector
HC Home Controller
Figure 2: SIAM architecture.
Figure 2 shows the multi-agent architecture. As
can be observed, the diagnostic agent is physically
located at the CMD. The CMD is conceived as a
specialized centre that could be a nursery home or in
general a company dedicated to taking care of
elderly people. The CMD can monitor several
houses, each one having its own diagnostic agent.
The data collector agents and the home
controller have to be located in a computer in the
house of the elderly person. This computer does not
require attention by the person and no screen
associated to the PC is needed if it is not required by
the user. The maintenance of the SIAM agents must
be done by remote control from the CMD.
The caretaker agent is in a mobile device that
could be supported by another PDA in the
caretaker’s home or by another form of
communication.
5 SIAM IMPLEMENTATION
The implementation of SIAM is based on the multi-
agent system free software tool named JADE (Java
Agent DEvelopment Framework) (JADE, 2007).
SIAM was designed taking into account a low cost
for the resulting product. The JADE architecture is
based on a set of platforms that include containers.
SIAM consists of a central platform that emulates
the CMD. This platform includes a container with
the diagnostic agent.
Each user has his/her own platform consisting of
three containers. The first container includes all the
agents that monitor activities at the house of the
elderly person: data collector agents and home
controller. The second and third containers include
the user and caretaker agents respectively. These
agents are supported by portable devices such as a
PDA, and they have been implemented in JADE
Leap, a special version for theses cases, using J9
from IBM as a virtual machine. The communication
of these agents with the rest is WiFi.
All the SIAM agents have been developed in
JAVA and their interfaces using SWT: The Standard
Widget Toolkit, another free software tool. Finally,
the permanent storage of all the data collected and
other data required by SIAM uses MySQL as
database server.
The diagnostic agent is an intelligent agent
physically running in a computer at the CMD. It has
a graphical interface for each house being
monitored. It receives all the relevant facts from the
corresponding home controller agent of a house and
uses an inference engine to reach conclusions about
possible risks for elderly people in their homes or
attending their demands. The architecture of the
INTELLIGENT SYSTEM FOR ASSISTING ELDERLY PEOPLE AT HOME
161

diagnostic agent is based on an expert system
including a knowledge base where the definitions
are included of risky situations in a house and
actions to take to prevent them. The knowledge base
is particular for each user and can include special
circumstances of each user. The elaboration of the
knowledge base has to be done under the direction
of personnel specialized in the care of elderly
people. The knowledge base has a very simple
architecture based on production rules including
certainty factors. The inference engine that uses the
knowledge is a classical forward-chaining engine.
The diagnostic agent can take autonomous
decisions without waiting for a confirmation from
the CMD personnel which simply will be informed.
This is an important feature of SIAM because it can
react very fast when a risk or demand is coming.
The diagnostic agent exchanges information with
the user through the home controller that is located
at the elderly person’s home. This agent is in charge
of the communication with all the devices that
support SIAM, and also, of keeping the diagnostic
agent updated about the profile of new risk
situations in the house. When a new user is coming
to SIAM, the first agent to start is the home
controller agent. Once it is alive, it tries to make
contact with the diagnostic agent, and if it accepts a
new user, it is monitored by the CMD. At this
moment the other agents associated to the user start
to work.
In this version of SIAM four data collector
agents have been developed in order to monitor
some activities at the elderly person’s home. Each
one is in charge of activities in a room of the house.
They are the following:
- data collector agent monitoring activities in
the bathroom: presence of the person,
closed/opened water valve in the shower and
washbasin and high vibrations.
- data collector agent monitoring activities in
the kitchen: presence of the person,
closed/opened water valve, close/open gas
and high vibrations.
- data collector agent monitoring activities in
the dining room: presence of the person,
and high vibrations.
- data collector agent monitoring activities in
the entrance of the house: presence of the
person and closed/opened door.
As was mentioned previously, this version of
SIAM is focused on the development of the
architecture of the whole multi-agent system. There
are no sensors connected yet, however a simulation
environment to simulate the operation of the real
sensors has been developed. More investigation has
to be done to select and install the appropriate
sensors and data collector according to expert
opinion on elderly people, but this will not change
the current operation of SIAM.
The user agent is running in a mobile device, in
this SIAM version it is in a PDA. It has an interface
which is extremely simple to use with big symbols
to communicate to and from the elderly person. The
user can activate one of three big icons: emergency
situation, communication with the caretaker and
asking some questions to the CMD. Figures 3.a and
3.b show how the user observes these icons.
3.a 3.b
Figure 3: Basic interface for the user agent.
Also, the user can receive warnings using big
icons about the need to review something in the
house that could be a risk for the person. Figure 3.b
shows an example of a warning telling the user to
turn off the gas. The structure and interfaces of the
caretaker agent are similar to those of the user agent,
and also, it is ready for running in a PDA.
6 EXAMPLE
This section describes a simple example of the
operation of SIAM. Let us suppose that the elderly
person is at home and he/she decides to go to the
bathroom to wash his/her hands. In this case a signal
corresponding to the presence of the person switches
from off to on, and a moment later the same occurs
with the signal corresponding to water running in the
washbasin. This information is collected by the data
collector agent that monitors the bathroom and it is
sent to the home controller agent. Once this agent
has pre-processed all the information in the house, it
sends these two new events to the diagnostic agent
in the CMD. Figure 4 shows the interface of the
diagnostic agent corresponding to this user. In this
figure, two new lines of information appear,
representing the two new facts received.
Immediately, the expert system is started and it does
not reach any conclusion about risk situations for the
user and so nothing happens.
HEALTHINF 2008 - International Conference on Health Informatics
162

Person in bathroom
Water running
Figure 4: interface of the diagnostic agent.
However, suppose that the person leaves the
bathroom and he/she is in the kitchen, but the water
is still running. In this case, the data collector in
charge of the bathroom updates the situation and
resends the new events to the home controller agent,
the data collector for the kitchen does the same, and
finally the home collector sends all the information
to the diagnostic agent. It starts the expert system
and it concludes the detection of a risk situation
taking the decision to wait some minutes more,
expecting new information from the home
controller. If the risk situation does not disappear,
the diagnostic agent takes the decision to send a
message to the home controller in order to issue a
warning message to the user agent to turn the water
off in the bathroom. In the case that the required
action is not executed because the diagnostic agent
detects that the problem persists, it automatically
decides to contact the caretaker. In order to do this it
sends a message to the caretaker agent similar to that
received by the user. Finally, if some time passes
without any action, then the CMD personnel is
informed.
Several similar situations were tested and the
general performance of all the processes was
successful.
7 CONCLUSIONS
This paper describes the architecture of the multi-
agent system named SIAM. It has been designed for
an automatic detection of possible risks of elderly
people at home and for assistance in these cases or
on demand of the user. SIAM will contribute to
extending the amount of time as much as possible
that elderly people can reside in their own homes
assisted continuously by an intelligent agent and
using new information technologies.
The resulting application has important
flexibility and incorporates new knowledge and new
features in an easy manner. Simulations of several
situations have been tested and the results are very
promising.
At present, a next phase of development of
SIAM is starting in order to include information
coming from real sensors. In parallel, a deeper
analysis concerning risky conditions in the elderly
person’s home is being developed. After these new
steps are completed in SIAM, it is expected to be
tested in real environments.
REFERENCES
Aguilera, S., Colás, J., Campos, E., 2003. Multi
Functional Call Center for Disabled and Elderly
People. Assistive Technology Conference AAAT,
2003. Dublín
Attentianet. Project web description retrieved June 20,
2007 from: http://www.attentianet.eu
Fishkin, K.P, Philipose, M. Rea, A. 2005. Hands-on
RFID: Wireless wearables for detecting use of objects.
IEEE International Symposium on Wearable
Computers.
JADE. Project web page retrieved June 20, 2007 from:
Robocare. Project web page retrieved June 20, 2007
from: http://robocare.ip.rm.cnr.it
Jih, W., Hsu, J. Y., Tsai, T., 2006. Context-aware service
integration for elderly care in a smart environment.
American Association for Artificial Intelligence
Workshop.
Liu, A. Hile, A., Kautz, H., Borriello, G., Brown, P.,
Harniss, M., Johnson, K, 2006. Indoor wayfinding:
Developing a functional interface for individuals with
cognitive impairment. Proceedings of the 8th
International ACM SIGACCESS Conference on
Computers and Accessibility.
Pollack, M., 2005. Intelligent technology for an aging
population: the use of AI to assists elders with
cognitive impairment. AI Magazine, 26(2)
Robocare. Project web page retrieved June 20, 2007 from:
http://robocare.ip.rm.cnr.it
United Nations, 2006. Department of Economic and
Social Affairs. Population ageing report 2006.
Retrieved June 20, 2207 from:
http://www.un.org/esa/population/publications/ageing/
ageing2006.htm.
Weiss, G. 2000. Multiagent Systems: A Modern Approach
to Distributed Artificial Intelligence, The MIT Press.
Wooldridge, M. 2002. An Introduction to MultiAgent
Systems, John Wiley & Sons.
INTELLIGENT SYSTEM FOR ASSISTING ELDERLY PEOPLE AT HOME
163