Ambient Assisted Living Buddy

Alexiei Dingli

and Michael Buhagiar

Department of Intelligent Computer Systems, Faculty of ICT, University of Malta, Msida, Malta

Keywords: Ambient Assisted Living, Virtual Companion, Software Agent.

Abstract: The research area of ambient assisted living is concerned with providing required assistance to elderly

and/or disabled people in order for them to get through their everyday tasks and chores in an easy and safe

manner. This usually makes use of hardware and/or software solutions in relation to the virtual / senior

companion field of study. Additionally, ambient assisted living exerts a number of tools and resources

specifically developed as technological aids for an ageing population. This project looks at the possibility of

improving someone’s quality of life through the use of an ambient assisted living prototype solution. By

combining vital research aspects from the fields of: ambient assisted living, technological aids for the aging

population, and senior / virtual companions, this project explore the development and evaluation of such a

software solution, the Ambient Assisted Living Buddy (AALB). This implies that the AALB manages to

work correctly as a concept to ambient assisted living. Furthermore, this project proved that the possibility

of improving someone’s quality of life through an ambient assisted living solution can be effective and has a

high potential.

1 INTRODUCTION

As the human body advances in age it is commonly

observed that the body gets weaker and more fragile

making it more accident prone and having everyday

life tasks gradually harder to complete. Bearing this

in mind, the sole driving force of this project was to

create something that delivers an environment for a

better and longer lasting life. This environment is to

be provided via the fully operational Ambient

Assisted Living Buddy (AALB), which seeks to

research whether an interactive monitoring system

raises hope in delivering an improvement upon the

current style of life.

1.1 Scientific Question

The principal aim of this dissertation was to be able

to come up with a sensible and well supported

answer to the scientific question,

Can an assisted living system (the AALB) improve

someone’s quality of life?

Through this, the concept and implementation of the

AALB started to take shape, with the notion of

managing to supply a sensible scientific solution.

1.2 Aims and Objectives

The primary objective of the AALB project is to

provide the construction for an interactive system

aimed at aiding the elderly and disabled to cope

better with their situation.

The AALB will be installed and integrated with a

user’s personal mobile phone thus allowing the ease

and extended ability of real-time monitoring and

user prompted actions. The most vital aspect of all

of this project is that it is kept simple to use, thus

implying that an old person with no real idea of the

present technological advancements can utilise it

just as well as a geek and an expert. The three main

goals are:

1. The aim was to review the current situation of

the elderly with respect to technological aids.

2. The design and implementations of an AAL,

complemented with a series of vital design

decisions. This was highlighted through the

methodology.

3. An assessment of the output results of this

research was carried out through system

evaluation, while the potential for further

developments was realised under Conclusion &

Future Work.

Dingli A. and Buhagiar M..

Ambient Assisted Living Buddy.

DOI: 10.5220/0005439800530058

In Proceedings of the 1st International Conference on Information and Communication Technologies for Ageing Well and e-Health (ICT4AgeingWell-

2015), pages 53-58

ISBN: 978-989-758-102-1

 2015 SCITEPRESS (Science and Technology Publications, Lda.)

2 RELATED WORK

Ambient assisted living is the concept and idea of

utilising various number of ICT and/or technology

related aspects in order to better the quality of life of

certain specific particular people. This primarily

focuses on aiding the elderly and/or disabled people

in managing to deal with fulfilling their everyday

needs through activities, chores, and entertainment.

Thus, in turn, such technologies and computer

related breakthroughs ease the life and job of said

conventional personnel whilst managing to maintain

a more pleasant experience for the effected elderly

and disabled alike. Finally, this area proves to make

the users lead a more independent life where less

support from the usual caretakers and family

relatives is required. The solutions to this area of

research are designed and implemented with key

aspects such as, a user-interactive feature, and its

integration with the everyday routine. Both of these

are very powerful and take into consideration that in

most cases the user suffers from impairments such

as hearing difficulties and lack of eye sight. By

managing to accomplish all of this, the resulting

solution helps in ensuring the simple yet vital

characteristic of doing so much with so little effort.

2.1 Assistive Technologies

The formal strict definition of an Assistive

Technology (AT) device is given as,

“...an umbrella term for any device or system

that allows an individual to perform a task they

would otherwise be unable to do or increases the

ease and safety with which the task can be

performed.” (Cowan and Turner-Smith, 1999);

(Beech and Roberts, 2008)

This definition is seen to describe what an AT

actually is and not really what it promises to achieve

in its methodologies and deliverables. Perhaps a

more goal oriented definition which focuses on the

AT’s ability to ensure the peek amount of

independence for the elderly, is shown hereunder,

“AT is any product or service designed to enable

independence for disabled and older people.”

(King’s Fund, 2001); (Beech and Roberts, 2008)

Moreover, the term AT goes over a distinct number

of devices that vary from, simple assistances to

move about, to complicated computer oriented

medical devices. Current persisting advances and

breakthroughs in technology bring, to the ATs, a

new amount of promising possibilities that it is now

their best chance of working in achieving and

realising their goal. These ATs, through the

utilisation of aiding assistance and notification

facilities, now have the better chance of allowing the

elderly people to live a longer self-sufficient life

from their own personal home. (Brignell et al., 2007)

(Blaschke et al., 2009) (Williams et al., 2013)

Having said this, the currently available and soon

to be released ATs fall under one of the following

arrangements, according to their specific prominent

task: (Doughty, 2004); (Beech and Roberts, 2008)

 Technologies which offer a sense of assistance to

individuals aiding them in any chores.

 Receptive technologies that aid the individual to

handle hazards and raise an alert.

 Preventative technologies that keep the user

away from hazards, while triggering an alert.

2.2 Technological Aids for the Aging

Population

The main idea of this research area revolves around

any kind of technology-oriented aid that is observed

to have the possibility of leaving an impact on the

quality of life and the life expectancy. It is important

to note that the approach is concerned with the

possibility of an improvement and not actual

certainty; thus the potential enhancements and aids

looked at do not have a definite guarantee of

appearing in the imminent future.

In the light of this project, these technological

aids incorporate new means of providing assistance

and support to elderly and disabled people, which

have been made possible through recent

technological development and breakthroughs. This

assistance and support comes as a methodology of

reducing or completely eliminating the current most

frequent accident types and dangerous situations. In

doing so, it must first be established whether such

accidents and scenarios could be reduced or

eliminated, and how any positive accomplishment

will effect life quality and expectancy.

2.3 Virtual Companions and Software

Agents

Virtual companions and software agents take a lot of

distinct forms, hence properly defining them comes

out as a general statement which collectively

includes every possible form. This general statement

defines an agent to simply be a software component

with some special features, and a virtual companion

to be nothing more than various collaborative agents

ICT4AgeingWell2015-InternationalConferenceonInformationandCommunicationTechnologiesforAgeingWelland

e-Health

working together. This is regarded as being the

prevailing common structure for a virtual

companion, having the diverse software agents

collaborating together and making up for all of the

available system functionalities and capabilities.

(Spyros et al., 2009)

Understanding an entire virtual companion

comes down to proper explanation of the underlying

software agents, which are sub-dived according to

their specific task completion abilities. (Spyros et

al., 2009)

2.4 The Elderly Population Today

As times progress and advances are made in the

development of medical and technological aids,

elderly people are somewhat assured that they are

able to live a longer life. This quickly results in an

increase in the percentage of old persons which

make up the entire population. As a matter of fact,

European countries have undergone a swing in their

population composition, with this percentage

increment taking its full effect (Parker and

Thorslund, August 1991). A logical fact is that,

growing older results in an increase in chronic health

difficulties. These chronically ill patients add up to

three quarters (75%) of all the healthcare’s expenses

and revenues. (Spyros et al., 2009).

Having established this scenario and situation, a

conclusion is now drawn that something must be

done so as to improve the quality of life of an elderly

person and to make him/her feel more secure and

content. In fact, first hand feedback from the elderly

people who already make use of some kind of AT

solutions reported some clear had facts. These

essentially include (i) an improvement in the ability

of taking a decision; (ii) a stronger sense of security;

(iii) more prominent freedom; (iv) an impression of

oversight and governess; (v) enhanced conditions of

life; (vi) preservation of the competence in staying at

home; (vii) less of a handful to the people taking

care of them; (viii) a boost in the aid specifically

towards persons suffering from abiding health

complications; (ix) diminishing in the number of

mishaps, mostly falls, which occur at home. (Beech

and Roberts, 2008)

2.5 Similar Systems

Fall Detection / Sensor Aiding. This branch covers

systems which are observed to make use of some

sort of sensors, with a fundamental focus in either

detecting a fall or tracking a location. These such

similar and relevant systems are namely, Fitbit

(Fitbit Inc., 2014), iFall (Sposaro and Tyson, 2009),

Detecting Human Falls with a 3 axis Digital

Accelerometer (Jia, 2009), and Sensor Aiding of

HSGPS Pedestrian Navigation (Mezentsev, 2005)

Home Care Companions. This division goes over

systems which were built as virtual companions, but

have their main driving force focused on providing

special home care support. These are, HEARTS

(Spyros et al., 2009), Virtual Carer (Sernani et al.,

2013), and HOMIE (Kriglstein and Wallner, 2005).

Dialogue / Conversational. This section explores

solutions which have the capabilities of managing a

dialogue scenario, as well as, closely mimicking

human comparable conversations. With relation to

the AALB, such a system is the Senior Companion

(Pinto et al., 2008).

Other / Miscellaneous. The final category in

comprised of only one system, namely ‘Google

Now’ (Google, 2012) which has a distinctive nature

of both functionality and deliverable aims.

3 METHODOLOGY

When it came to developing a system project, it was

first very important to properly establish the reason

for which the system would be developed. In doing

so, a problematic scenario is initially identified and

then system planning and construction could take

shape around this scenario and thrive to deliver a

proper solution. Following from this, the primary

AALB’s driving force was born and a clear picture

of how the prototype would deal with the problem

was set up.

Figure 1: System Diagram.

AmbientAssistedLivingBuddy

Vitally, strong attention is given to the User

Interface (UI), monitoring sensors and actuators, and

environmental re-actions in order to deliver a simple

to use yet fully functional system.

The AALB is a mobile application available on

the android operating system, which is based upon

the following three core aspects:

1. Presenting information at the right place and at

the right time

2. Offering a clear, understandable and easy to use

interface for certain everyday tasks

3. Collecting information directly and indirectly

from the user, producing the best possible

scenario output.

More specifically, the system utilises an array of

sensors along with the simple voice and touch

commands, in order to harvest the necessary input

data from the user. This captured data is then

processed in accordance to the particular scenario

and situation, and a corresponding user desired

output is produced. Figure 1 below depicts all this as

the system’s internal structure and workings.

With regards to the actual system and interface,

this will take a similar approach to what was

observed in the Google Now system, but with more

focus on elderly and disabled people rather than

everyday users and technology geeks.

A number of sensory inputs will be utilized to

gather additional information to aid a better

understanding of the appropriate and relevant output.

The sensors that are expected to be present are:

 Gravity Sensor: Measures the gravitational

forces.

 Linear Acceleration: Measures the forces of

acceleration, excluding the gravitational forces.

 Location Sensor: Tracks the location through

GPS.

In terms of functionality and the actual information,

this will vary according to the scenario changes

indicated by the numerous sensors.

The living buddy promises to provide the

functionality of, (i) Understanding the user voice

input; (ii) User programmable notifications and

alarms; (iii) Watching over the user and potentially

informing some external entity in the case of an

arising danger; (iv) Brighten or dim the light just by

speaking to the device; and (v) Display a weather

and news report with a click or voice command

The system’s design is considered to be in the

best way in terms of both, code understand-ability,

and system overall performance. This includes, (i)

Threads that enabled simultaneous task running

without hindering system performance; (ii) Stage

variables which allowed different system checks at

different situations; and (iii) Error reporting and

alarm handling including alarm calling routines with

respect to the state variable.

The idea of this design is that every function is

able to perform on its own and only “bothers” the

system UI in the case of an error, or upon a request

for user input data.

4 RESULTS AND EVALUATION

Apart from the usual testing procedures, a public

evaluation was conducted. The respondents for the

public feedback analysis attended a social meeting

known as GRUFAN made up of nurses, elderly and

their relatives.

All of this gave me a total of around 26 people,

with age groups varying from 10-25 to 65+, a

diverse number of capabilities and a wide variety of

distinct living scenarios. For the scope of testing out

the AALB, the gender was considered as irrelevant

and thus was not even recorded.

The user stories are a couple of thought out

scenarios that the AALB is expected to be used in.

This inspects the specific situation the AALB will

mostly be useful at, whilst also identifying any

lacking features with respect to the situation.

Primarily, the system is aimed to be used on a

mobile device and thus different mobile

environments are compared to highlight some of the

limitations with respect to this criteria. Secondly,

exploration is done with regards to a person’s need

for support or living habitat. There are a total of nine

(9) user stories, divided up into two categories,

mobile oriented (first 3), and living environment and

support (the rest); as seen hereunder:

1. Prohibited mobile device use

2. Use of mobile device without carrying it around

3. Use of a mobile device and carrying it around

4. Living alone at home

5. Living with relatives but can manage alone

6. Living with relatives and requiring their

assistance

7. Living at home with someone’s assistance

8. Not living at home but can manage alone

9. Not living at home and requiring assistance

For each of the user story scenarios, the AALB was

evaluated and findings were made with respect to

the most effected functions.

 Scenario 1: The AALB faired very well in this

scenario with all of the vital features functioning.

ICT4AgeingWell2015-InternationalConferenceonInformationandCommunicationTechnologiesforAgeingWelland

e-Health

In fact, more than 80% of the respondents rated

the effected functions a 4 or a 5 (out of 5).

 Scenario 2: The major concern of this scenario

was the notifications functionality. Positively,

over 90% of the results rated this feature to be

near perfect with a score of either a 4 or a 5 (out

of 5).

 Scenario 3: This utilises all of the functions of

the AALB, and thus its evaluation comes down

to the analysis of entire system. Overall the

system generated a successful feedback and so

this scenario has been entirely fulfilled.

 Scenarios 4 - 9: These scenarios utilise the same

features, thus requiring AALB testing in its full

potential. This performed very well, around 90%

of the respondents gave a high rating in the

overall evaluation and above 80% gave a high

rating in the in depth functionality evaluation.

An overall evaluation questionnaire was used in

order to gather basic user information detail, along

with the overall system experience. This was divided

into three parts, (i) Personal Information; (ii) Daily

Activities and Scenario Capabilities and (iii) General

Overall System Usability.

With respect to the overall performance of the

AALB, the system was greeted very well and

positively managed to reach what it was set out to

do. Additionally, 95% of the respondents felt an

added sense of safety through fall detection and

location tacking. On a positive conclusion, just short

of 95% of the respondents indicated interest in

future similar systems, and the concept of AAL. An

in depth functionality questionnaire was used to

focus more on sole performance of four system

features, (i) Fall Detection; (ii) Location Tracking;

(iii) Notification Alerts and (iv) Voice Command

Recognition.

Analysing the respondents’ feedback, it was

observed that the system functionalities performed

very well. The great majority of the users gave a

score of 4 and above. Moreover, the greater portion

of the users agreed that the system features did not

seem to lack anything in particular, while others

pointed out that the AALB could use something

more. With respect to the fall detection, suggestions

were mostly related to notifying/ calling a pre-set

relative number. Some of these suggestions also

mentioned incorporating the GPS location with an

SMS alert in order to improve the response time of

the alerted parties.

The location tracking feature did not favour one

or two specific improvements. In fact, this generated

the distinct enhancements of, (i) Tracking past

locations; (ii) Audio alarms / notifications; (iii)

Stating the units for the input parameters; and (iv)

Accuracy. The notifications functionality followed a

similar approach with four distinct areas which the

users pointed out as requiring further working on.

Namely these are: (i) Repetitions; (ii) Location

orientation; (iii) Voice message / Read back; and

(iv) Important updates / relevant information.

It is also worth mentioning that both the location

tracking and the notifications functionalities found

users that reported that the specific feature worked

fine and did not require any kind of change and/or

improvement.

In the end, attention was directed at the future of

the system and what this might hold for it, with

respect to any additional features that the users feel

such an AAL system should have. A prominent 80%

of the participants pointed out that some additional

features such as voice commands, notifications, etc

would make the system better.

Once again, apart from all this, it is worth

mentioning that a couple of the respondents simply

reported that the system does not require any

additional features.

5 CONCLUSION

This project involved research which mainly

revolved around the areas of ambient assisted living

and senior companions. Both of these areas were

discovered to be very vast and wide spread, thus the

most relevant of the two was put together to get the

basic idea for the development of the project’s

prototype. In fact, the area of AAL provided for the

concept of the project in terms of its required

functionality, whilst the area of senior companions

provided for the feel of the system and its interaction

with the target audience.

The AALB brought with it a number of valuable

achievements as well as beneficial and helpful

lessons. Both of these are regarded to be among the

most crucial aspects of the entire project. The

valuable achievements accomplished through this

project can be derived from the results obtained

through the evaluation of the AALB against user

feedback. In fact, (i) more than 80% of the

respondents were pleased with the system

performance, resulting in an overall average of 4.5

out of 5; (ii) just under 95% of the users indicated an

added sense of safety through the fall detection and

location tracking abilities; and (iii) almost 95% of

the users showed a tendency to use a similar system

in the future. All of these positive results proved to

AmbientAssistedLivingBuddy

be vital with respect to the scientific question as they

all managed to show the great potential of AAL. In

fact, the AALB was found to be a valid candidate to

suit the needs of a future AAL system.

REFERENCES

Beech, R., & Roberts, D. (2008). Assistive technology and

older people. Social Care Institue for Excellence

(SCIE).

Blaschke, C. M., Freddolino, P. P., & Mullen, E. E.

(2009). Ageing and technology: a review of the

research literature. The British Journal of Social Work,

Vol. 39 No. 4, 641-656.

Brignell, M., Wootton, R., & Gray, L. (2007, July). The

application of telemedicine to geriatric medicine, Vol.

36 Issue 4. Age & Ageing, 369-374.

Cowan, D. D., & Turner-Smith, A. (1999). The Role of

Assistive Technology in Alternative Models of Care

for Older People. Royal Commission on Long Term

Care, 2, 325-346.

Doughty, K. (2004). Supporting independence: the

emerging role of technology. Housing, Care and

Support, Vol. 7 Issue 1, 11-17.

Fitbit Inc. (2014). Fitbit Official Site. Retrieved from

fitbit: http://www.fitbit.com/uk/home

Google. (2012, June). Introducing Google Now. Retrieved

from Google now:

http://www.google.com/landing/now/#whatisit

Jia, N. (2009, July). Detecting Human Falls with a 3-Axis

Digital Accelerometer. Analog Dialogue, Vol. 43

Number 7.

King’s Fund. (2001). Consultation Meeting on Assistive

Technology. London, United Kingdom: King's Fund.

Kriglstein, S., & Wallner, G. (2005). HOMIE: an artificial

companion for elderly people. CHI '05 Extended

Abstracts on Human Factors in Computing Systems,

2094-2098.

Mezentsev, O. (2005). Sensor Aiding of HSGPS

Pedestrian Navigation . Calgary, Alberta, Canada:

University of Calgary.

Parker G., M., & Thorslund, M. (August 1991). The Use

of Technical Aids Among Community-Based Elderly.

American Journal of Occupational Therapy, 712-718.

Pinto, H., Wilks, Y., Catizone, R., & Dingli, A. (2008).

The Senior Companion Multiagent Dialogue System

(Short Paper). 7th International Conference on

Autonomous Agents and Multiagent Systems (AAMAS

200) (pp. 1245-1248). Estoril, Portugal: Padgham,

Parkes, Müller and Parsons (eds.).

Sernani, P., Claudi, A., Palazzo, L., Dolcini, G., &

Dragoni, A. F. (2013). Home Care Expert Systems for

Ambient Assisted Living: A Multi-Agent Approach.

In C. Workshops (Ed.), The challenge of Ageing

Society: technological roles and opportunities for

Artificial Intelligence (pp. 1-16). Torino, Italy:

Universita Politecnica delle Marche.

Sposaro, F., & Tyson, G. (2009). iFall: An android

application for fall monitoring and response.

Engineering in Medicine and Biology Society, 2009.

EMBC 2009. Annual International Conference of the

IEEE (pp. 6119–6122). Tallahassee, Florida, United

States of America: Florida State University.

Spyros, R., Pirros, T., Aimilios, C., & Sotiris, K. (2009).

User Interaction Design for a Home-Based Telecare

System. In A. Holzinger, & K. Miesenberger, Human-

computer interaction and Usability for eInclusion (pp.

333-344). Berlin & Heidelberg, Germany: Springer.

Williams, V., Victor, C. R., & McCrindle, R. (2013). It Is

Always on Your Mind: Experiences and Perceptions

of Falling of Older People and Their Carers and the

Potential of a Mobile Falls Detection Device. Current

Gerontology and Geriatrics Research, vol. 2013,

Article ID 295073. Retrieved from

doi:10.1155/2013/295073

ICT4AgeingWell2015-InternationalConferenceonInformationandCommunicationTechnologiesforAgeingWelland

e-Health