2. Context: Aging population
• According to Eurostat, the percentage of
the European population aged 65 years and
over will grow from 16% in 2010 up to
29,3% in 2060 (EUROSTAT, 2011).
Demographic changes in Europe (source: Birks, 2007) #
2#
3. Def: A Companion is an agent that
a) makes itself 'useful', i.e. is able to carry out a variety of tasks in order
to assist humans
b) behaves socially, i.e. possesses social skills in order to be able to
interact with people in a socially acceptable manner.
Note, a) and b) imply a long-term perspective:
• repeated interactions
• need for learning and adaptation
• potential for relationships, but does not assume that people “bond”
with robots (friend vs. assistant/butler)
Idea:
Being useful & making a positive contribution to people’s quality of life
The Robot as a Companion
Paradigm
4. Project:
ACCOMPANY
• Acceptable robotiCs COMPanions for AgeiNg Years
• The Accompany project addresses target outcome (a) service and
social robotics systems for “Ageing well” under objective
ICT-2011.5.4 ICT for Ageing and Wellbeing under EU FP7.
• A major challenge concerning the development of service robots for
the elderly is to enhance existing state of the art technologies and
interfaces in order to facilitate independent living at home.
• Focus on re-ablement
• User-centred approach
• Start 1 October 2011
• 3.6 M Euros EC contribution
• Coordinated by University of Hertfordshire
• (Farshid Amirabdollahian)
• 9 partners in 5 EU countries
5. Consortium
• The University of Hertfordshire (UH), United
Kingdom
• Hogeschool Zuyd (HZ), The Netherlands
• Fraunhofer (Fraunhofer), Germany
• University of Amsterdam (UVA), University
of Twente (UT), The Netherlands
• University of Siena (UNISI), Italy
• Maintien en Autonomie à Domicile des
Personnes Agées (MADOPA), France
• University of Birmingham (UB), University of
Warwick (UW), United Kingdom
5#
6. Involving users
• 6 rounds of user panels throughout the
project
• Evaluation in formative and summative
studies:
– Primary users (elderly)
– Secondary users (informal carers)
– Health care providers and local authorities
(formal carers)
6#
7. Focus on existing platform
Care-O-Bot 3 - redesigned
http://www.care-o-bot.de/english/Care-O-bot_3.php#
8. ACCOMPANY Objectives &
Results I
• Social and empathic interaction design
– Novel interfaces and interaction modalities
• Robot learning and adaptability
– Cognitive architecture for robot integrated in
smart home, human-robot co-learning and
teaching
• Environment and activity monitoring
– Tracking and activity recognition (data set)
8#
9. ACCOMPANY Objectives &
Results II
• Acceptability of robots for promoting
independence
– Studies in summative evaluation with users in
three countries
• Ethical issues surrounding companion
robots
– New frameworks for ethical guidelines for
robot technology and elderly people
9#
10. Some lessons learnt/challenges
• Activity recognition for multiple persons in
realistic home setting
• Safe and socially interactive platform
• Combining learning & personalization/
human-user teaching
• Easy to understand interfaces for users