Conference PaperPDF Available

ARI: the Social Assistive Robot and Companion



ARI: the Social Assistive Robot and Companion
Sara Cooper1, Alessandro Di Fava2, Carlos Vivas3, Luca Marchionni4, Francesco Ferro5
Abstract With the world population aging and the number
of healthcare users with multiple chronic diseases increasing,
healthcare is becoming more costly, and as such, the need
to optimise both hospital and in-home care is of paramount
importance. This paper reviews the challenges that the older
people, people with mobility constraints, hospital patients and
isolated healthcare users face, and how socially assistive robots
can be used to help them. Related promising areas and
limitations are highlighted. The main focus is placed on the
newest PAL Robotics’ robot: ARI, a high-performance social
robot and companion designed for a wide range of multi-modal
expressive gestures, gaze and personalised behaviour, with great
potential to become part of the healthcare community by
applying powerful AI algorithms. ARI can be used to help
administer first-care attention, providing emotional support to
people who live in isolation, including the elderly population
or healthcare users who are confined because of infectious
diseases such as Covid-19. The ARI robot technical features
and potential applications are introduced in this paper.
World population is aging, and projections from Europe,
USA, and Asia foresee that the number of people older than
64 will increase by more than 70% because of an increase
in average lifespan. The World Health Organization (WHO6)
estimates that by 2050 the number will increase to 2 billion
from the 900 million in 2015.
Healthcare spending has in the last few decades grown
steadily, from 4-6% of Gross Domestic Product (GDP) in
1970 to an average 8.8% in 2019 (statistics from OECD-
countries7). In the coming years, this spending is expected to
keep increasing, due to a number of factors, including the fact
that the proportion of older people among the population will
be increasing, escalating costs per work hour in the health-
care sector[1] and investment in healthcare technology[2],
affecting even more those with lower socioeconomic status
and with multiple chronic conditions.
This also means that there will be fewer hands to take
care of those requiring healthcare [3], [4], increasing the
healthcare burden. Efficient technological solutions have
the potential to improve the quality of healthcare services
1Sara Cooper is a robotics software engineer at PAL Robotics
2Dr. Alessandro Di Fava is a robotics software engineer at PAL Robotics
3Carlos Vivas is the head of social robotics and products for research at
PAL Robotics
4Luca Marchionni is CTO at PAL Robotics
5Francesco Ferro is PAL Robotics CEO
at the same time as keeping costs and staff requirements
In the last few years the healthcare sector has been
attempting to shift hospital care to homes in order to reduce
hospital stays and their cost [3]. For patients it is more
familiar, convenient and cost-effective to stay at their homes
while they can receive regular care, remote monitoring and
support, delivered by community nurses or caregivers. Care-
givers can coordinate with patients’ clinicians and exchange
data through eHealth, becoming part of an integrated care
system. By shifting care to the home, re-admission to nursing
homes or hospitals may be prevented, with consequent cost-
saving, in particular for rural areas where healthcare centres
may be too far away.
Loneliness and lack of engagement with the community
can be experienced by older people that are living alone,
but that still want to live as independently as possible in
their homes [5], even among older patients with good overall
health. Furthermore, there is a high incidence of strokes
among the older population [6] and it is important that they
receive intensive and frequent rehabilitation therapy, help in
other physical activities at home and general entertainment
to improve their mood, which can be difficult due to lack of
human resources.
Robots have significantly impacted on the healthcare sec-
tor by means of surgical, rehabilitation or assistance robots
[7]. Healthcare robots, in particular Socially Assistive Robots
(SAR) can help in reducing emergency visits and healthcare
costs, and encourage independent living, while also reduc-
ing caregiver burden [3]. In a hospital context, healthcare
robots can become a part of the “Smart hospital of the
future”, consisting of an environment integrated with diverse
technologies and functions with the goal of empowering
patients, optimizing patient cases, and processing automation
and alerts 8.
The rest of the paper is organized as follows. Section
II provides an overview of healthcare robots focusing on
SAR, their features and possible uses, and highlighting the
needs during contagious diseases such as Covid-19. Then
possible limitations and challenges of SAR are examined in
Section III. Section IV presents the ARI robot, the new robot
companion and social assistive robot, focusing on its features
and applications. Finally, a brief conclusion is presented in
Section V.
As healthcare robotics has only started growing signifi-
cantly in the past few years, there is no standardised method
for classifying them as yet. Considering the classification
of several related works [8], [3], [9] a possible general
classification might consist of the following categories:
Surgical robots
Mobile logistics robots
Robots for mobility and rehabilitation
Personal assistant robots
Aside from robotic surgery, which is the main area in
healthcare where robots have been used, some of the opportu-
nities arising now include administering of hospital logistics:
smart inventory management systems, efficient transport of
hospital materials such as blood samples, diagnostic tests and
food delivery, and medicine delivery at the request of a med-
ical professional to the patient’s rooms by autonomous robot,
to name some examples. Right now the main challenges in
this field are ensuring safety in crowded areas by suitable
sensing abilities as well navigation from floor to floor and
through narrow areas [9].
Rehabilitation robots adapt and interact with patients phys-
ically with the aim of restoring functionality, especially for
those that have suffered strokes or spinal cord injuries and
need post-stroke therapy. They may provide support to a
patient’s weight as they walk (SoloWalk, by McCormick et
al. [10]). Related robots are exoskeletons such as ReWalk
[11], used among people with paralysis, which help users
walk and climb stairs to compensate for their lack of regular
physical abilities with positive outcomes. Brain-computer in-
terface controlled robots such as wheelchair-mounted robotic
arms [12] or robot prosthetics can help such users or those
with limb loss carry out manipulation tasks.
However, recently there has been a focus in robotics
research on the use of personal assistant robots, “robots de-
signed for living together with and assisting human beings”,
by using robots that can collaborate and communicate with
people. Personal assistant robots are still not used much in
real applications, but are expected to grow in demand. These
can be of classified into two types:
Physically assistive robots
Socially assistive robots: therapy, companion, entertain-
ment and telepresence robots
Such assistive robots could be used at homes to assist in
daily activities like cooking, cleaning, eating, and especially
handovers, where the robot brings an object requested by the
end-user. Two examples of such robots are shown in Figure
1. Personalised dressing assistance is another interesting area
of research, as it involves daily living activity that entails
the greatest burden on caregivers, with some work carried
out by Zhang et al. [13], using a dual-armed Baxter robot.
However, this is complex due to many older people having
limited upper-body movements and the need for a real-time
system that adapts to quick unexpected behaviour.
There are also many novel robots that are being designed
and tested to target specific tasks, robots like Robear can
Fig. 1: Two versions of older Robotics’ TIAGo for object
fetching (left) and interacting with older adults for enRichMe
project (right)
aid in lifting and transferring older people and those with
paralysis from bed to wheelchair [14], which is especially
physically demanding and risky for nurses.
A. Socially Assistive Robots (SAR)
A special group of Healthcare robots is the Socially As-
sistive Robots [15], [6], [16]. SAR focus on communication
with the aim of enhancing users’ health and psychological
well-being through diagnosis, therapy, or offering compan-
ionship [17] to those who feel lonely. One of the major
advantages of therapy robots is that users can have access
to them whenever they need them, rather than depending on
therapist availability.
Table I summarises how SARs can be of help for different
types of users. The table has been developed based on
PAL Robotics’ experience in healthcare related works such
as EnrichMe9, SOCRATES10, SPRING11 and SHAPES12
H2020 projects, not to mention the user-centered design and
market analysis carried out for all PAL Robotics’ robots.
SAR are “robots that assist users by providing social
interactions such as appropriate emotional or social cues” [9].
Positive outcomes have been obtained in the health sector as
they increase motivation and quality of life [15], [6], [16].
In the context of health, they can offer personalised care
and encouragements, interact in many languages, call users
attention, make appointments through speech or reminding
users to take their medicine, as well as early diagnosis and
continuous health assessment and monitoring.
SAR can be pet-like, such as Paro, Aibo or iCat, who
can serve as stress relief [19] and encourage interaction
between older people. Research has demonstrated that robots
like Paro can reduce stress and anxiety among older people
with dementia [20] and may also serve as user-adapted
telepresence robots. Telepresence robots “enable an operator
to be virtually present at a remote location and to provide
actuators that enable the operator to interact with the remote
environment” [9]. Some commercial humanoid robots like
TABLE I: Potential use of SAR robots for different types of users
Potential user Needs SAR benefits
Older people: frail, those with severe
chronic conditions, with neurodegenerative
Isolation and boredom
Cognitive and physical decline
Forgets to do tasks, does not know where
some things are
Monitoring the progression of illness
Engage users with community (telepres-
ence), share news, social events
Play cognitive games, physical exercises
and other in-home activities customised
to their interests and capabilities
Remind users to take medications, how
to do tasks
Monitoring of vital functions, risks (gas,
fire hazards, users falling), cognitive
Users with mobility or physical difficulties
(adults / older adults) due to paralysis, loss
of limb, general physical weakness.
Cannot walk independently
Need help in physical manipulation tasks
Need of carrying out rehabilitation exer-
cises regularly and frequent monitoring
Want to be up to date with their illness
Guide and support in navigation
Fetch objects, including teleoperation
Offer physical rehabilitation games
Remote monitoring
Emergency support, fall detection and
Isolated patients or users with special needs
that require remote monitoring.
Isolation and boredom
Communication with professionals
Monitoring the progression of illness
Engage users with community and pro-
fessionals (telepresence), share news, so-
cial events
For people isolated due to an infection, it
also reduce risk of spreading the disease
Hospital staff can monitor and make re-
quests of robot actions through internet
Encourage children with autism to de-
velop and use social skills and engage
them in conversation [18]
Patients in waiting rooms or entrance of
hospitals and care centers [5].
Boredom while waiting
Patient logistics
Monitoring the progression of illness
Can serve as companions and provide
entertainment, including cognitive games
Access to users personal health records in
order to provide assessment, enable smart
appointments scheduling
Give information to patients, serve as
a receptionist to welcome, check-in and
alert the medical personnel of appoint-
ments, and guide new incomers to the
hospitals, triage tasks
Hospital staff can monitor and make re-
quests of robot actions through internet
Kompai 13, Pepper 14, Care-o-bot 15 , Sanbot 16 or TIAGo
17 [21] can be used as SAR for assistive tasks, for instance
to facilitate communication with family and doctors, and to
ensure that frail, isolated or vulnerable people, may engage
in other social activities. Thanks to this, doctors can thus
interact with patients remotely, and provide education and
instructions, without a face-to-face interaction. Users that can
benefit most of such robots include older people, individuals
with cognitive impairments, those recovering from stroke or
related injuries, and children with autism [18], [22]. In fact,
SAR used in groups of older individuals have been shown to
reduce stress and loneliness, with an overall positive health
impact [4], [8] and, in general, acceptability regarding a
robot’s long-term stay at home [15], [16].
Multiple research projects have focused on SAR. HU-
MAVIPS (Humanoids with Audio-visual Abilities in Pop-
ulated Spaces) 18, FP7 CompanionAble (Companionable
research project delivers robotic assistance for the older
individuals) 19 aimed to provide a companion robot for
older people with Mild Cognitive Impairment that helped
engage with community. H2020 GrowMeUp 20 highlighted
on encouraged independent living while engaging with the
community by teleconference. H2020 EnrichMe has tested
their robot TIAGo (Figure 1) for monitoring, physical, social
and cognitive assistance to older people with mild cognitive
impairment, by playing games and giving personalised re-
minders [23], [24]. By working closely with the target users
and their carers the robot design and validation results proved
to be highly positive.
B. Healthcare robots and Covid-19
Many of the applications mentioned so far may be of direct
use in the situation the world is facing right now - Covid-
19 - and some have already been adapted with the aim of
reducing risk of infection and optimising health management.
New technologies are becoming of paramount importance
in fighting the pandemic, including robots, which can help
promote remote human-to-human interactions in order to
reduce risk of transmission, lighten the burden on health care
providers, reduce loneliness and improve overall health [25].
The SAR role in these situations can be multiple:
Use of drones by CloudMinds 21 to enforce quarantine
restrictions, alerting individuals to return to their homes,
deliver medicine to patients with Covid-19 at Wuhan as
well as transferring test samples.
Mobile robots can be used for hospital logistics 22 23.
Mobile robots that can sterilize surfaces with UV light
such as [26] or Blue Ocean’s UV disinfection robot 24.
Automatic temperature-taking with thermal-sensor
equipped robots in public places to quickly screen
several people at the same time in large areas, helping
to cover screening.
Social robots that make people less lonely in the
presence of widespread quarantine, monitoring them,
encourage treatment follow-up and offer reminders,
as social distancing can have a negative impact on
mental health. Such robots can also help with hospital
While the above mentioned robotic solutions offer excit-
ing opportunities, their deployment in complex real-world
scenarios and hospitals is not without its limitations [9].
Andrade et al. [7] identified the main concern to be the
high cost of robots, which limits most robot use in research
and reduces large-scale robot acceptability studies both with
those who manage the robot and end-users. Data privacy
issues, which is also a concern for other AI systems such
as smart-speakers as well as telepresence robots, may result
in individuals feeling uncomfortable about being recorded
without their consent. In the context of ethics, robots should
be compliant with regulation, and doing a proper risk assess-
ment can be a time-consuming and complex procedure. Aside
from these points, human-robot interaction poses several
challenges for healthcare [27]:
Ethical challenges
Positive user experience
Cultural differences
User social acceptance and attitude toward robotic tech-
nologies, especially among the older population
Robot morphology, associated with robot design
In general, a more user-centred design approach would
be needed to solve these challenges. Firstly, social robots
must be efficient and robust enough to achieve specified
goals, for instance [23], by speeding up physical tasks e.g.
bringing objects for domestic use. Secondly, social robots in
general are still lacking in interaction capabilities. One main
drawback in speech interaction with SAR is that so far they
have been focused on one-on-one interaction, with limited
words or sentences that they can recognise, rather than multi-
modal/multi-party scenarios outside the established field [9].
Robots ought to improve their human emotion and activity
recognition skills (emotionally intelligent robots) in order
to achieve a more expressive interaction and adapt to the
needs of each individual, instead of the other way around;
especially so that robots can help in psychotherapeutic and
home support settings [27]. Safety of all users involved
(robot, patient and healthcare workers) must be taken into
account during HRI design, by reducing sharp edges on
the robot, speed limits, safety distances [27], anticipating
potential hazardous situations and responding quickly to new
situations via real-time perception of human activity. This is
especially important for physical HRI [28]. To improve user
experience and acceptance user interfaces should be easy and
intuitive for users with different types of needs, by using a
combination of verbal and non-verbal communication.
Lastly, it should not be forgotten that robots are just one
part of healthcare services, and that for best outcomes they
should be well integrated with other medical devices, mobile
apps, health records and sensors, and take communities of
pharmacists, labs, doctors, and so on into consideration.
The newest PAL Robotics’ social robot ARI25 was con-
ceived taking into account the previous challenges. ARI’s
user-centered design (Figure 2) has been focused on several
key considerations: mobility, lightweightness, safety, sim-
plicity and modernity. The major goal has been to improve
user acceptability of social robots. both by operators and
end-users, by making it more human-like when it comes
to both visual appearance and behaviour features such as
voice and movement. For this reason ARI has been designed
to resemble the human body, on the one hand considering
a suitable body-head proportion and degree of iconicity.
Secondly, its height has been set to 1.65 cm, the average
height of a female adult, adding to it two arms, human-
like face and body form. In contrast to related robots in
the market it is the one that most accurately mirror humans
appearance [29]. Its covers are 3D printed with PA12 to
obtain smooth surfaced to reduce risk of injury to people,
with a low centre of gravity and low mass of the upper limbs
to minimize impact and fall risks.
ARI is provided with a mobile base, torso with an inte-
grated 10.1” Linux-based touch-screen, two arms and a head
with expressive gaze thanks to its 2 LCD eyes. It works
on Ubuntu LTS, with open source ROS (Robotic Operating
System) 26, with an ARI simulator available in Gazebo 27. It
can be configured according to customer needs, facilitating
integration of external software components. Some of its
technical specifications are detailed in Table II.
Fig. 2: PAL Robotics ARI robot
TABLE II: ARI robot specifications
Arm payload 0.5 kg
Hardware max speed 1.5 m/s
Battery life 8-12 h
Computing power Intel i5/i7, up to 32GB RAM
Interfaces 2 x LCD screen eyes with custom ani-
2 x 16 GB LED rings in the ears, and
a 40 RGB LED back ring
10.1” 1200x800 projected capacitive
touch screen
Connectivity Bluetooth, WiFi, Ethernet
With a high processing NVIDIA GPU and its ROS and
REST API, it can be used to develop and deploy powerful
AI algorithms, such as deep learning applied to natural
language processing, object or face recognition, as well as
using reinforcement learning the robot may learn from the
interaction, and adapt its behaviour to each user.
ARI’s cameras, such as a Sony 8 MegaPixel RGB camera
on the head, and Intel RealSense RGB-D cameras on the
front and back of the torso, make it possible to use advanced
3D Perception algorithms and understand the surrounding
environment. It is already equipped with some perception
packages for ARUCO marker detection, face and people
detection as well as planar object detection.
Emphasis has been paid on the design of the audio
architecture of the robot, strengthening the robot’s speech
interaction capabilities (Figure 3). It has 2 speakers facing
to the front and a ReSpeaker MicArray V2.0 28 microphone
array for audio input/output in order to converse using natural
language. Audio tests were conducted to verify audio quality,
and the covers were modified to include a hole on the front
side of the torso, where the array is added with a protective
case, which has reduced the reverberation initially detected
when the robot recorded its own voice. The ReSpeaker has
4 microphones and was selected due to its feasibility with
28 Mic Array v2.0/
integration in ROS through the respeaker ros package 29 and
usage of pulseaudio 30 for audio play/recording. For speech
interaction, ARI is currently equipped with ACAPELA text-
to-speech solutions 31.
Fig. 3: ARI’s audio architecture.
PAL Robotics has implemented ARI with a Visual-SLAM
system for mapping and localization in indoor environments,
based on ORB SLAM [30], which uses ARI’s front torso In-
tel Realsense RGB-D camera to detect features and perform
loop closures when it returns to a previously seen location
to update the map of the environment. Through this, after
mapping, an occupancy grid map is produced, where ARI can
localize in. For autonomous navigation it makes use of ROS’s
move base thanks to which it can avoid obstacles and plan a
path to the desired goal. ARI’s SLAM and navigation system
may be adapted further by researchers to guide hospital
newcomers to their appointments or help the older people
with mobility issues around the house. For this purpose, it
has 2 differential drive wheels and 2 caster wheels, with a
maximum speed of 0.5 m/s respect the hardware maximum
speed, limited by obstacle avoidance frequency that needs to
slow down autonomous max speed to prevent collision.
It is demonstrated that people are able to interpret hu-
manlike (affective) nonverbal behaviour (HNB) in artificial
entities. To express emotion/ empathy a few body cues are
used simultaneously mainly: facial displays, body movement
and posture and vocal cues.
Ruhland et al. [31] suggests that robots that exhibit such
human-like gaze attention, by generating automatic robot
facial and body gestures related to the prosodic content, may
create stronger feelings from user, improve their understand
of the robot [20], making it more predictable, increasing its
acceptability and trust. Such gestures can also emphasize
what the robot says through words. ARI’s strong point is its
ability in providing this multi-modal behavior.
Its LCD eyes with animations, supported by the movement
of the head (Figure 4) , enrich the non-verbal interaction
29 ros/
by facilitating the establishment of joint attention between
human robot and convey intention, interest and emotions
to initiate and respond to the human partner. The design
of ARI’s slender arms offer the option to show expressive
and human-like gestures combined with facial and prosodic
features to enrich the interaction. Other cues that ARI can
display are changing of the LED colours of the ears or the
back torso, to inform about its battery or when it has heard
someone speak to it, etc. The LED manager enables robot
users to adjust colour, brightness and effects.
Fig. 4: ARI robot’s gaze behaviour
A. ARI applications as a healthcare assistant and companion
In the healthcare context, ARI’s capabilities could easily
adapt to carry out many of the tasks previously mentioned
by SAR including hospital reception, patient registration,
provide health assessment and also entertain users at hos-
pital waiting rooms or at home. ARI may also be used to
remind users to do different tasks, based on monitoring of
physiological or behavioural parameters (mood, food intake,
sleep), e.g. to take their medication, drink water or eat,
suggest social tasks or a specific diet. These are the major
applications for which the ARI robot was designed and some
of them are in progress even through collaboration projects.
The touchscreen that is fixed on the torso enables ARI
to function as a telepresence robot which users can use
to communicate with family or doctors without requiring
physical contact. It can be used to show entertainment
content such as videos, or games, including cognitive games
for people with cognitive decline, or to show content that
is aligned with what ARI is saying, thus achieving a user-
friendly interface for those that may have hearing problems.
ARI is IoT friendly and enables interconnection with smart
devices, wearable sensors, mobile phones and other Ambient
Assisted Living applications. In the healthcare context, it
is suitable for remote monitoring of older patients and
post-stroke users at home, and collection of physiological
data from wireless temperature or pulse oximeter sensors
[32]. The data from these devices can be collected by ARI
via Bluetooth or WiFi, and, alongside other data collected
through interaction (game progress, etc) resend to a cloud
server in order to apply deep learning algorithms that may
be used to detect abnormalities and improve early disease
diagnosis. Through this connection, ARI can update doctors
or caregivers of the patients health and alert them if nec-
essary. Similarly, by deploying human activity recognition
algorithms and thanks to its cameras, ARI can be adapted to
detect falls or other house hazards (gas leaks, etc. intruders)
and trigger alarm.
Face tracking and recognition systems enable ARI to
recognise new users that come to a hospital, or identify
uniquely each patient in order to provide customised care -
suggestions, games, reminders - by accessing their electronic
medical records, or identify the role of each person caregiver,
doctor, patient. Through emotion recognition algorithms,
ARI could detect when a person is in need of attention or
bored, in order to initiate contact or adjust the interaction ac-
cordingly. Gesture recognition algorithms can be employed,
combined with learning by imitation, to deliver a physical
therapy session to people with upper-limb mobility problems,
where ARI can show a series of arm movements that have
been previously taught by a physical therapist and that users
need to repeat.
In the context of Covid-19 in particular, ARI can help
optimise first hospital care and reduce excessive exposure by
medical professionals to ill patients. It can also be added with
a thermal camera to take a group of people’s temperature
from a distance, and provide initial health assessments, thus
offering a first triage of possible cases an reducing hospital
staff burden. At the same time, it can entertain and encourage
social interaction of quarantined users, monitor users both at
hospitals and home, or offer telepresence communication.
Some of these applications are already study subjects in
the two EU projects where ARI is currently involved, H2020
SHAPES and H2020 SPRING. In SPRING project ARI’s
skills will be further advanced to assist patients at a day-
care hospitals, working on improving the navigation, multi-
modal speech interaction, and human behaviour understand-
ing. Audio evaluation and improvement is a core topic in
SPRING, where the goal is that the robot can engage in
multi-party conversations at a day-care hospital. In SHAPES
ARI will serve as a companion robot to support healthy living
of older individuals. The robot will work delivering cognitive
games as it is enriched by digital assistant, facial and emotion
recognition systems provided by partners.
This paper has presented an overview of ARI, the new PAL
Robotics’ SAR and companion. The work started by high-
lighting the motivation behind the need for healthcare robots
and reviewing the main identified needs of the older people,
those with physical constraints and people in isolation due
to infectious diseases, and also how SAR can benefit them,
to improve hospital care and promote independent living.
The ARI robot was conceived and designed to address these
demands including a wide range of multi-modal expressive
gestures, gaze and personalised behaviour, with great poten-
tial to become part of the healthcare community by applying
powerful AI algorithms. ARI’s expressive eyes enrich its
multi-modal behaviour even more. ARI is flexible in design,
and its purpose is to create a positive user experience, being
a companion offering emotional support. Future works with
ARI will be validating it with real users especially to improve
the human-robot interface. This effort is already in progress
in some healthcare-related collaborative projects (SPRING,
We have seen that although SAR robots have many ad-
vantages such as helping reduce emergency visits, healthcare
costs, and promoting independent living, while also reducing
caregiver burden, there are still some barriers however. These
include robot acceptability; the need to redesign hospital
management, infrastructure, IT, information flow between
care providers and patients; constraints in speech interaction;
lack of empathy (emotion recognition). Up until now, real-
life applications of SAR for healthcare are emerging and
being validated for their deployment, and are still behind
logistics and surgery in quantity of deployments. However,
over the next few years the adoption of SAR robots will be
streamlined with an urgent need, in response to situations
such as the COVID-19 crisis, maturing these solutions as
they become more relevant. Importance in robot design,
being end-user centered, conveying social cues to under-
stand and be understood by humans, and ensuring the robot
remains easy to use and deploy, will become increasingly
important for SAR in order to turn them into real partners,
not just tools - and the ARI robot is moving in this direction.
[1] W. J. Baumol, The cost disease: Why computers get cheaper and
health care doesn’t. Yale university press, 2012.
[2] B. Przywara et al., “Projecting future health care expenditure at
european level: drivers, methodology and main results,” Directorate
General Economic and Financial Affairs (DG ECFIN), European ... ,
Tech. Rep., 2010.
[3] H. Robinson, B. MacDonald, and E. Broadbent, “The role of health-
care robots for older people at home: A review,” International Journal
of Social Robotics, vol. 6, no. 4, pp. 575–591, 2014.
[4] J. Broekens, M. Heerink, H. Rosendal, et al., “Assistive social robots
in elderly care: a review,” Gerontechnology, vol. 8, no. 2, pp. 94–103,
[5] M. Gombolay, X. J. Yang, B. Hayes, N. Seo, Z. Liu, S. Wadhwania,
T. Yu, N. Shah, T. Golen, and J. Shah, “Robotic assistance in the
coordination of patient care,” The International Journal of Robotics
Research, vol. 37, no. 10, pp. 1300–1316, 2018.
[6] M. J. Matari ´
c, J. Eriksson, D. J. Feil-Seifer, and C. J. Winstein,
“Socially assistive robotics for post-stroke rehabilitation,Journal of
NeuroEngineering and Rehabilitation, vol. 4, no. 1, p. 5, 2007.
[7] A. O. Andrade, A. A. Pereira, S. Walter, R. Almeida, R. Loureiro,
D. Compagna, and P. J. Kyberd, “Bridging the gap between robotic
technology and health care,” Biomedical Signal Processing and Con-
trol, vol. 10, pp. 65–78, 2014.
[8] J. Abdi, A. Al-Hindawi, T. Ng, and M. P. Vizcaychipi, “Scoping review
on the use of socially assistive robot technology in elderly care,BMJ
open, vol. 8, no. 2, 2018.
[9] L. Bodenhagen, S.-D. Suvei, W. K. Juel, E. Brander, and N. Kr¨
“Robot technology for future welfare: meeting upcoming societal
challenges–an outlook with offset in the development in scandinavia,
Health and Technology, vol. 9, no. 3, pp. 197–218, 2019.
[10] A. McCormick, H. Alazem, A. Morbi, R. Beranek, R. Adler, G. Tibi,
and E. Vil´
e, “Power walker helps a child with cerebral palsy,” in
International Conference on Control, Dynamic Systems, and Robotics,
[11] G. Zeilig, H. Weingarden, M. Zwecker, I. Dudkiewicz, A. Bloch, and
A. Esquenazi, “Safety and tolerance of the rewalkTM exoskeleton suit
for ambulation by people with complete spinal cord injury: a pilot
study,The journal of spinal cord medicine, vol. 35, no. 2, pp. 96–
101, 2012.
[12] M. Palankar, K. J. De Laurentis, R. Alqasemi, E. Veras, R. Dubey,
Y. Arbel, and E. Donchin, “Control of a 9-dof wheelchair-mounted
robotic arm system using a p300 brain computer interface: Initial
experiments,” in 2008 IEEE International Conference on Robotics and
Biomimetics. IEEE, 2009, pp. 348–353.
[13] F. Zhang, A. Cully, and Y. Demiris, “Probabilistic real-time user
posture tracking for personalized robot-assisted dressing,” IEEE Trans-
actions on Robotics, vol. 35, no. 4, pp. 873–888, 2019.
[14] D. Szondy, “Robear robot care bear designed to serve japan’s aging
population,” Retrieved July, vol. 19, 2018.
[15] Y.-h. Wu, J. Wrobel, M. Cornuet, H. Kerherv´
e, S. Damn´
ee, and A.-S.
Rigaud, “Acceptance of an assistive robot in older adults: a mixed-
method study of human–robot interaction over a 1-month period in
the living lab setting,Clinical interventions in aging, vol. 9, p. 801,
[16] M. Pino, M. Boulay, F. Jouen, and A. S. Rigaud, ““are we ready for
robots that care for us?” attitudes and opinions of older adults toward
socially assistive robots,Frontiers in aging neuroscience, vol. 7, p.
141, 2015.
[17] J.-J. Cabibihan, H. Javed, M. Ang, and S. M. Aljunied, “Why robots?
a survey on the roles and benefits of social robots in the therapy of
children with autism,” International journal of social robotics, vol. 5,
no. 4, pp. 593–618, 2013.
[18] D. Feil-Seifer and M. J. Matari´
c, “Toward socially assistive robotics for
augmenting interventions for children with autism spectrum disorders,”
in Experimental robotics. Springer, 2009, pp. 201–210.
[19] K. Wada and T. Shibata, “Living with seal robots—its sociopsycho-
logical and physiological influences on the elderly at a care house,”
IEEE transactions on robotics, vol. 23, no. 5, pp. 972–980, 2007.
[20] E. B. Onyeulo and V. Gandhi, “What makes a social robot good at
interacting with humans?” Information, vol. 11, no. 1, p. 43, 2020.
[21] J. Pages, L. Marchionni, and F. Ferro, “Tiago: the modular robot that
adapts to different research needs,” in International workshop on robot
modularity, IROS, 2016.
[22] B. Scassellati, H. Admoni, and M. Matari´
c, “Robots for use in autism
research,” Annual review of biomedical engineering, vol. 14, pp. 275–
294, 2012.
[23] S. Cos¸ar, M. Fernandez-Carmona, R. Agrigoroaie, J. Pages, F. Ferland,
F. Zhao, S. Yue, N. Bellotto, and A. Tapus, “Enrichme: Perception and
interaction of an assistive robot for the elderly at home,International
Journal of Social Robotics, pp. 1–27, 2020.
[24] C. Salatino, L. Pigini, M. M. E. Van Kol, V. Gower, R. Andrich,
G. Munaro, R. Rosso, A. P. Castellani, and E. Farina, “A robotic
solution for assisting people with mci at home: Preliminary tests of
the enrichme system.” in AAATE Conf., 2017, pp. 484–491.
[25] G.-Z. Yang, B. J. Nelson, R. R. Murphy, H. Choset, H. Christensen,
S. H. Collins, P. Dario, K. Goldberg, K. Ikuta, N. Jacobstein, et al.,
“Combating covid-19—the role of robotics in managing public health
and infectious diseases,” 2020.
[26] P. Chanprakon, T. Sae-Oung, T. Treebupachatsakul, P. Hannanta-
Anan, and W. Piyawattanametha, “An ultra-violet sterilization robot for
disinfection,” in 2019 5th International Conference on Engineering,
Applied Sciences and Technology (ICEAST). IEEE, 2019, pp. 1–4.
[27] I. Olaronke, O. Oluwaseun, and I. Rhoda, “State of the art: a study
of human-robot interaction in healthcare,” International Journal of
Information Engineering and Electronic Business, vol. 9, no. 3, p. 43,
[28] I. Maurtua, A. Ibarguren, J. Kildal, L. Susperregi, and B. Sierra,
“Human–robot collaboration in industrial applications: Safety, inter-
action and trust,” International Journal of Advanced Robotic Systems,
vol. 14, no. 4, p. 1729881417716010, 2017.
[29] E. Broadbent, V. Kumar, X. Li, J. Sollers 3rd, R. Q. Stafford, B. A.
MacDonald, and D. M. Wegner, “Robots with display screens: a robot
with a more humanlike face display is perceived to have more mind
and a better personality,PloS one, vol. 8, no. 8, p. e72589, 2013.
[30] R. Mur-Artal, J. M. M. Montiel, and J. D. Tard´
os, “ORB-SLAM:
a versatile and accurate monocular SLAM system,” CoRR, vol.
abs/1502.00956, 2015. [Online]. Available:
[31] K. Ruhland, C. E. Peters, S. Andrist, J. B. Badler, N. I. Badler,
M. Gleicher, B. Mutlu, and R. McDonnell, “A review of eye gaze
in virtual agents, social robotics and hci: Behaviour generation, user
interaction and perception,” in Computer graphics forum, vol. 34,
no. 6. Wiley Online Library, 2015, pp. 299–326.
[32] M. Al-khafajiy, T. Baker, C. Chalmers, M. Asim, H. Kolivand,
M. Fahim, and A. Waraich, “Remote health monitoring of el-
derly through wearable sensors,” Multimedia Tools and Applications,
vol. 78, no. 17, pp. 24 681–24 706, 2019.
... TIAGo is a mobile service robot whose notable features allow it to perform a wide range of tasks such as navigation, manipulation, perception, interaction and motion planning [136]. In addition, ARI is a mobile humanoid service robot whose primary design goals are enhancing user acceptability regarding social robots [137] and adopting AI algorithms for caring purposes [11]. Figure 14 shows the latest versions of both robots. ...
... Figure 14 shows the latest versions of both robots. [136] and ARI (right) [137] from PAL Robotics. ...
Full-text available
Global demographics trend toward an aging population. Hence, there will be an increased social demand for elderly care. Recently, assistive technologies such as service robots have emerged and can help older adults to live independently. This paper reports a review starting from 1999 of the existing mobile service robots used for older adults to grow old at home. We describe each robot from the viewpoint of applications, platforms, and empirical studies. Studies reported that mobile social robots could assist older adults throughout their daily activities such as reminding, household tasks, safety, or health monitoring. Moreover, some of the reported studies indicate that mobile service robots can enhance the well-being of older adults and decrease the workload for their caregivers.
... Subsequently, in [10] we combined elements of this earlier research; integrating visually-grounded dialogue with social conversation and task-oriented domain-specific conversations in a prototype multimodal receptionist for a hospital waiting room. That demonstration was via a webbased interface; here, we present our current progress in integrating the multimodal conversational AI system in a Social Robot, the ARI robot [11], and report on a preliminary experimental validation with a small group of volunteers. ...
... For the preliminary validation of the system, the Conversational Manager and Task Planner modules were integrated with the ARI robot [11] as ROS nodes, and run on an external PC. The robot was provided with a female Text-to-Speech (TTS) voice (Acapela's UK English voice 'Rachel') set at 50% of the maximum volume. ...
... Social robots have been deployed in healthcare settings during the pandemic, for: 1) reducing social isolation and loneliness through telepresence [12], [13]; 2) triaging of incoming patients [2], and 3) disinfecting surfaces and temperature monitoring [14]. ...
... The expressive ARI robot was introduced in [13] with the potential to help with tasks during COVID-19, in hospitals and home-care settings, including engagement in cognitive games, providing reminders, and initiating video calls with family. ...
Conference Paper
The rapid spread of COVID-19 around the globe has increased the need to adopt autonomous social robots within our healthcare systems. In particular, socially assistive robots can help to improve the day-to-day functioning of our healthcare facilities including long-term care, while keeping residents and staff safe by performing repetitive tasks such as health screening. In this paper, we present the first human-robot interaction study with an autonomous multi-task socially assistive robot used for non-contact screening in long-term care homes. The robot monitors temperature, checks for face masks, and asks screening questions to minimize human-to-human contact. We investigated staff perceptions of 7 attributes: screening experience without and with the robot, efficiency, cognitive attitude, freeing up staff, safety, affective attitude, and intent to use the robot. Furthermore, we investigated the influence of demographics on these attributes. Study results show that, overall, staff rated these attributes high for the screening robot, with a statistically significant increase in cognitive attitude and safety after interacting with the robot. Differences between gender and occupation were also determined. Our study highlights the potential application of an autonomous screening robot for long-term care homes.
... Hence, our society is on the verge of brimming with loneliness, and mental health is sure to deteriorate if nothing is done to remedy the situation. Socially assistive robots can be useful in dealing with loneliness as they can also function as companions [6,7]. ...
Full-text available
Emotion monitoring can play a vital role in investigating mental health disorders that contribute to 14% of global diseases. Currently, the mental healthcare system is struggling to cope with the increasing demand. Robot-assisted mental health monitoring tools can take the enormous strain off the system. The current study explored existing state-of-art machine learning (ML) models and signal data from different bio-sensors assessed the suitability of robotic devices for surveilling different physiological and physical traits related to human emotions and discussed their potential applicability for mental health monitoring. Among the selected 80 articles, we subdivided our findings in terms of two different emotional categories, namely—discrete and valence-arousal (VA). By examining two different types of signals (physical and physiological) from 10 different signal sources, we found that RGB images and CNN models outperformed all other data sources and models, respectively, in both categories. Out of the 27 investigated discrete imaging signals, 25 reached higher than 80% accuracy, while the highest accuracy was observed from facial imaging signals (99.90%). Besides imaging signals, brain signals showed better potentiality than other data sources in both emotional categories, with accuracies of 99.40% and 96.88%. For both discrete and valence-arousal categories, neural network-based models illustrated superior performances. The majority of the neural network models achieved accuracies of over 80%, ranging from 80.14% to 99.90% in discrete, 83.79% to 96.88% in arousal, and 83.79% to 99.40% in valence. We also found that the performances of fusion signals (a combination of two or more signals) surpassed that of the individual ones in most cases, showing the importance of combining different signals for future model development. Overall, the potential implications of the survey are discussed, considering both human computing and mental health monitoring. The current study will definitely serve as the base for research in the field of human emotion recognition, with a particular focus on developing different robotic tools for mental health monitoring.
... Including healthcare and elderly assistance, contributing socially and commercially at marketplaces and in transportation, as well as providing assistance at schools, offices and even at homes, covering both indoor and outdoor environments. Roomba, TurtleBots, Pioneer3-AT, Pepper robot are few examples [1] [2]. ...
Full-text available
To ensure the steady navigation for robot stable controls are one of the basic requirements. Control values selection is highly environment dependent. To ensure reusability of control parameter system needs to generalize over the environment. Adding adaptability in robots to perform effectively in the environments with no prior knowledge reinforcement leaning is a promising approach. However, tuning hyper parameters and attaining correlation between state space and reward function to train a stable reinforcement learning agent is a challenge. In this paper we designed a continuous reward function to minimizing the sparsity and stabilizes the policy convergence, to attain control generalization for differential drive robot. We Implemented Twin Delayed Deep Deterministic Policy Gradient on Open-AI Gym Race Car. System was trained to achieve smart primitive control policy, moving forward in the direction of goal by maintaining an appropriate distance from walls to avoid collisions. Resulting policy was tested on unseen environments including dynamic goal environment, boundary free environment and continuous path environment on which it outperformed Deep Deterministic Policy Gradient.
... Les robots d'assistance sociale peuvent opérer dans différents lieux (e.g., centres de santé spécialisés ou chez l'habitant, Portugal leur permettre une meilleure qualité de vie, une plus grande autonomie ainsi qu'un maintien à domicile prolongé (Abdi et al., 2018;Di Nuovo et al., 2018;Portugal et al., 2019). Les robots d'assistance sociale, en étant capables de communiquer avec les individus, peuvent apporter une aide importante aux professionnels de santé surchargés notamment en proposant de les soutenir dans la réalisation de certaines tâches (Cooper et al., 2020;Mihoub et al., 2013a). ...
Cette thèse répond à la demande issue d’un projet ANR (Agence Nationale de la Recherche) qui vise à doter un Robot d’Assistance Sociale (RAS) de compétences le rendant capable de procéder à un dépistage précoce de troubles neurocognitifs. Ce travail doctoral a deux objectifs. Le premier correspond à l’identification, la catégorisation et l’opérationnalisation des compétences que le psychologue mobilise lors de l’évaluation des capacités cognitives de personnes âgées. Le deuxième objectif vise à objectiver la qualité de l’alliance de travail entre le psychologue et la personne âgée dans le contexte de la passation évaluative. Dans les deux cas, une analyse du processus de leur mise en œuvre a été menée.À cette fin, un corpus multimodal a été créé à partir de l’enregistrement audio-visuel de 11 psychologues filmés dans un living-lab pendant qu’ils évaluaient les capacités cognitives de 64 personnes âgées à l’aide de deux tests évaluatifs (i.e., MMSE et RL/RI-16). Basée sur le relevé des actions verbales, une grille d’analyse des compétences du psychologue en contexte évaluatif a été élaborée à partir de ce corpus audio-visuel, selon une approche inductive en trois étapes. Cette grille inventorie 15 compétences, dont 9 compétences centrées test et 6 compétences centrées relation, nécessaires aux psychologues dans la réalisation de la tâche d’évaluation des troubles neurocognitifs.Les résultats montrent que les psychologues verbalisent davantage lorsque les personnes âgées présentent des capacités cognitives faibles, sans pour autant être en mesure de préciser à quel type de compétences le psychologue a recours (i.e., compétences centrées test ou centrées relation). La qualité de l’alliance de travail de la dyade a été analysée en mesurant la synchronie interactionnelle non verbale (SINV). Les résultats montrent que la SINV est significativement prédite par le taux de compétences centrées relation mobilisées par le psychologue.Ce travail doctoral apporte des éléments de réponse sur les déterminants de l’interaction psychologue – personne âgée en contexte évaluatif. Par ailleurs, les résultats concernant l’identification, la catégorisation et l’opérationnalisation des compétences du psychologue en contexte évaluatif tentent de pallier certains problèmes théoriques liés aux compétences. De plus, dans ce contexte spécifique, la SINV semble être une mesure prometteuse de la qualité de l’alliance de travail de la dyade. Pour finir, cette thèse apporte un éclairage théorique et méthodologique sur la conception ergonomique d’un RAS dont l’objectif est de dépister précocement les troubles neurocognitifs de personnes âgées.
Full-text available
To ensure the steady navigation for robot stable controls are one of the basic requirements. Control values selection is highly environment dependent. To ensure reusability of control parameter system needs to generalize over the environment. Adding adaptability in robots to perform effectively in the environments with no prior knowledge reinforcement leaning is a promising approach. However, tuning hyper parameters and attaining correlation between state space and reward function to train a stable reinforcement learning agent is a challenge. In this paper we designed a continuous reward function to minimizing the sparsity and stabilizes the policy convergence, to attain control generalization for differential drive robot. We Implemented Twin Delayed Deep Deterministic Policy Gradient on Open-AI Gym Race Car. System was trained to achieve smart primitive control policy, moving forward in the direction of goal by maintaining an appropriate distance from walls to avoid collisions. Resulting policy was tested on unseen environments including dynamic goal environment, boundary free environment and continuous path environment on which it outperformed Deep Deterministic Policy Gradient.
Robotic task instructions often involve a referred object that the robot must locate (ground) within the environment. While task intent understanding is an essential part of natural language understanding, less effort is made to resolve ambiguity that may arise while grounding the task. Existing works use vision-based task grounding and ambiguity detection, suitable for a fixed view and a static robot. However, the problem magnifies for a mobile robot, where the ideal view is not known beforehand. Moreover, a single view may not be sufficient to locate all the object instances in the given area, which leads to inaccurate ambiguity detection. Human intervention is helpful only if the robot can convey the kind of ambiguity it is facing. In this article, we present DoRO ( D isambiguation o f R eferred O bject ), a system that can help an embodied agent to disambiguate the referred object by raising a suitable query whenever required. Given an area where the intended object is, DoRO finds all the instances of the object by aggregating observations from multiple views while exploring & scanning the area. It then raises a suitable query using the information from the grounded object instances. Experiments conducted with the AI2Thor simulator show that DoRO not only detects the ambiguity more accurately but also raises verbose queries with more accurate information from the visual-language grounding.
Socially active humanoid robots (SAHRs) are designed to communicate and interact with humans in humancentric environment using speech, movements, gestures, or facial expressions to communicate with their users following some set of social behavior while providing their assistance. Just like humans interact in an adaptive manner with others by changing their speech, tone, and body language intuitively, such type of adaptive behavior can be developed in SAHRs to get a human-like rich interaction capabilities. Therefore, a lot of research work and studies are going on to replicate various behavioral aspects of humans into SAHRs, so that human-robot interaction can be improved further. Besides interacting with humans, humanoid robot should be able to perform the assigned tasks remotely and also in real time with better accuracy. Thus, these social robots designed can be used in a diversified field of applications like education, healthcare, entertainment, communication, constructions, medical, collaborations, hazard management systems, etc.
Full-text available
Recent technological advances enabled modern robots to become part of our daily life. In particular, assistive robotics emerged as an exciting research topic that can provide solutions to improve the quality of life of elderly and vulnerable people. This paper introduces the robotic platform developed in the ENRICHME project, with particular focus on its innovative perception and interaction capabilities. The project’s main goal is to enrich the day-to-day experience of elderly people at home with technologies that enable health monitoring, complementary care, and social support. The paper presents several modules created to provide cognitive stimulation services for elderly users with mild cognitive impairments. The ENRICHME robot was tested in three pilot sites around Europe (Poland, Greece, and UK) and proven to be an effective assistant for the elderly at home.
Full-text available
This paper discusses the nuances of a social robot, how and why social robots are becoming increasingly significant, and what they are currently being used for. This paper also reflects on the current design of social robots as a means of interaction with humans and also reports potential solutions about several important questions around the futuristic design of these robots. The specific questions explored in this paper are: “Do social robots need to look like living creatures that already exist in the world for humans to interact well with them?”; “Do social robots need to have animated faces for humans to interact well with them?”; “Do social robots need to have the ability to speak a coherent human language for humans to interact well with them?” and “Do social robots need to have the capability to make physical gestures for humans to interact well with them?”. This paper reviews both verbal as well as nonverbal social and conversational cues that could be incorporated into the design of social robots, and also briefly discusses the emotional bonds that may be built between humans and robots. Facets surrounding acceptance of social robots by humans and also ethical/moral concerns have also been discussed.
Full-text available
Robotic solutions to dressing assistance have the potential to provide tremendous support for elderly and disabled people. However, unexpected user movements may lead to dressing failures or even pose a risk to the user. Tracking such user movements with vision sensors is challenging due to severe visual occlusions created by the robot and clothes. In this paper, we propose a probabilistic tracking method using Bayesian networks in latent spaces, which fuses robot end-effector positions and force information to enable cameraless and real-time estimation of the user postures during dressing. The latent spaces are created before dressing by modeling the user movements with a Gaussian process latent variable model, taking the user’s movement limitations into account. We introduce a robot-assisted dressing system that combines our tracking method with hierarchical multitask control to minimize the force between the user and the robot. The experimental results demonstrate the robustness and accuracy of our tracking method. The proposed method enables the Baxter robot to provide personalized dressing assistance in putting on a sleeveless jacket for users with (simulated) upper-body impairments.
Full-text available
In many countries, a demographic change has been recognized and is subject to public discussions, either directly or indirectly by social systems being challenged with the growing demands. However, with the increase of life expectancy, also the type of needs change due to an increase of co-morbidity and multi-chronic conditions asking for an increased focus on the patient as a whole rather than the individual diseases. Recent technological advances provide new opportunities for technical solutions that interact with end users and the utilization of robots is considered one potential mean for addressing this challenge. This article outlines the changes in the demands, with particular examples taken from the Danish health care system as an example, together with the technological achievements within the robotics domain. We identify where technologies that to a large degree are existing already today can be utilized to support the social systems in the near future. We show that several of the challenges related to the demographic change can be addressed with technology that is already available and that for some cases have reached the mass market already. We also outline the to be expected opportunities and challenges in the development of future robots in the health-care domain.
Full-text available
Due to a rapidly increasing aging population and its associated challenges in health and social care, Ambient Assistive Living has become the focal point for both researchers and industry alike. The need to manage or even reduce healthcare costs while improving the quality of service is high government agendas. Although, technology has a major role to play in achieving these aspirations, any solution must be designed, implemented and validated using appropriate domain knowledge. In order to overcome these challenges, the remote real-time monitoring of a person’s health can be used to identify relapses in conditions, therefore, enabling early intervention. Thus, the development of a smart healthcare monitoring system, which is capable of observing elderly people remotely, is the focus of the research presented in this paper. The technology outlined in this paper focuses on the ability to track a person’s physiological data to detect specific disorders which can aid in Early Intervention Practices. This is achieved by accurately processing and analysing the acquired sensory data while transmitting the detection of a disorder to an appropriate career. The finding reveals that the proposed system can improve clinical decision supports while facilitating Early Intervention Practices. Our extensive simulation results indicate a superior performance of the proposed system: low latency (96% of the packets are received with less than 1 millisecond) and low packets-lost (only 2.2% of total packets are dropped). Thus, the system runs efficiently and is cost-effective in terms of data acquisition and manipulation.
Full-text available
Objective With an elderly population that is set to more than double by 2050 worldwide, there will be an increased demand for elderly care. This poses several impediments in the delivery of high-quality health and social care. Socially assistive robot (SAR) technology could assume new roles in health and social care to meet this higher demand. This review qualitatively examines the literature on the use of SAR in elderly care and aims to establish the roles this technology may play in the future. Design Scoping review. Data sources Search of CINAHL, Cochrane Library, Embase, MEDLINE, PsychINFO and Scopus databases was conducted, complemented with a free search using Google Scholar and reference harvesting. All publications went through a selection process, which involved sequentially reviewing the title, abstract and full text of the publication. No limitations regarding date of publication were imposed, and only English publications were taken into account. The main search was conducted in March 2016, and the latest search was conducted in September 2017. Eligibility criteria The inclusion criteria consist of elderly participants, any elderly healthcare facility, humanoid and pet robots and all social interaction types with the robot. Exclusions were acceptability studies, technical reports of robots and publications surrounding physically or surgically assistive robots. Results In total, 61 final publications were included in the review, describing 33 studies and including 1574 participants and 11 robots. 28 of the 33 papers report positive findings. Five roles of SAR were identified: affective therapy, cognitive training, social facilitator, companionship and physiological therapy. Conclusions Although many positive outcomes were reported, a large proportion of the studies have methodological issues, which limit the utility of the results. Nonetheless, the reported value of SAR in elderly care does warrant further investigation. Future studies should endeavour to validate the roles demonstrated in this review. Systematic review registration NIHR 58672.
COVID-19 may drive sustained research in robotics to address risks of infectious diseases.
We conducted a study to investigate trust in and dependence upon robotic decision support among nurses and doctors on a labor and delivery floor. There is evidence that suggestions provided by embodied agents engender inappropriate degrees of trust and reliance among humans. This concern represents a critical barrier that must be addressed before fielding intelligent hospital service robots that take initiative to coordinate patient care. We conducted our experiment with nurses and physicians, and evaluated the subjects’ levels of trust in and dependence upon high- and low-quality recommendations issued by robotic versus computer-based decision support. The decision support, generated through action-driven learning from expert demonstration, produced high-quality recommendations that were accepted by nurses and physicians at a compliance rate of 90%. Rates of Type I and Type II errors were comparable between robotic and computer-based decision support. Furthermore, embodiment appeared to benefit performance, as indicated by a higher degree of appropriate dependence after the quality of recommendations changed over the course of the experiment. These results support the notion that a robotic assistant may be able to safely and effectively assist with patient care. Finally, we conducted a pilot demonstration in which a robot-assisted resource nurses on a labor and delivery floor at a tertiary care center.
Robots have the potential to support care and independence of older adults. The ENRICHME project is developing an integrated system composed of a robot, sensors and a networking care platform, aiming at assisting older adults with MCI in their home environment. This paper reports findings of the tests performed on a sample of MCI users and their caregivers, with the first version of the ENRICHME system, in a controlled environment.