Full citation is:
Robotherapy: Definition, Assessment, and Case Study
Elena
LIBIN and
Alexander
LIBIN
Institute of Robotherapy at the Robotic Psychology, CyberAnthropology Research Inc., U.S.A.
Department of Psychology, Georgetown University, Washington DC
libina@georgetown.edu
Abstract. Robotherapy is
defined as a framework of person
– robot interactions aimed at the reconstruction of one’s negative
experiences through the development of new coping skills mediated by
technological tools. Robotherapy
as a new research area focuses on the analysis of person - robot communication,
viewed as a complex interactive system, with the emphasis on psychological
evaluation, diagnosis, prognosis, and principles of non-pharmacological
treatment. Effectiveness of
robotherapy is influenced
by both (1) one’s past experiences, current needs and individual preferences,
and (2) an artificial partner’s non-transitive physical features and
behavioral configurations defined through the intensity of simulations and
responses. In
the proposed article we discuss the construction of a new scale, which includes
(1) an assessment of a person’s
individual style of communication with the robotic cat NeCoRo; (2)
participant’s evaluation of his/her new experiences with the robotic cat; (3)
a person’s past experiences with live pets and modern technology; and (4) a
participant’s evaluation of the robotic cat features, and also NeCoRo’s
advantages and disadvantages. Finally, in the context of robotherapy we assume
that human-robot interactions have symbolic meaning.Therefore each
pattern evaluated via the newly developed scale, is assigned certain
psychological value derived from a person's life experiences, likes and
dislikes, and current emotional, cognitive, and behavioral states.
1.
World of robotics
There was a time when humanity faced
the
universe
alone and without a friend. Now he has
creatures to help him…Mankind is no longer alone.
Isaac
Asimov (I, Robot).
As technology expands its influence on human society, new unexpected
opportunities challenge the human mind. The merge of artificial and human
worlds, as long predicted by Issac Asimov [1], is happening before our very
eyes. An optimistic view builds up productive expectations among the lay public,
researchers, and practitioners who all see different but positive consequences
of human beings interacting with artificial partners [2]. A great variety of
artificial creatures already exists in the contemporary world of robotics. This
variety provides the possibility for a primary classification of robots that
have come into being in recent decades. Such criteria as the purpose of a robot,
its relation to human activity, and the robot’s behavioral configuration
defined by the degree of freedom, robo-IQ (complexity of artificial
intelligence), and the robot’s physical properties serve as the basics for
the following classification:
1. The
purpose of
robot’s creation:
· Industrial robots
· Research robots (i.e., used in space and marine
exploration, etc.)
· Military robots
· Medical robots
· Recreational robots (i.e., Disney World’s
dinosaurs, the humanoid Pino, etc.)
· Interactive simulation robots
with therapeutic potentials (Omron’s robotic cat NeCoRo,
Sony’s dog AIBO, the automated doll Amazing Amy, etc.)
2.
Robot’s relation to human activity:
· Substitute for human activity (i.e., rescue and
research robots)
· Addition to human activity (i.e. medical, military
and industrial robots)
· Stimulation of human activity (i.e., recreational
and simulation robots)
3.
Type of the degree of freedom in combination with robot’s physical properties
and robo-IQ:
· Closed loop (i.e., industrial robots with low robo-IQ)
·
Open loop (i.e., simulation-therapeutic robots with high robo-IQ provided by
enhanced
artificial intelligence)
The class of interactive
simulation robots was found as the most appropriate for a pilot study
at the initial stage of robotherapy.
2. Artificial partners as human companions
The above described class of interactive simulation robots has two
characteristics that clearly place
it in its proposed classification. First, a general
feature of the class
of interactive simulation robots is that those creatures
are created for the purpose of communication with a person. This is why we call
this type of robot an artificial partner. The concept of artificial partners [2]
placed the relationships between humans and robots into a psychological, rather
than technological, context. A few characteristics depict artifical partners
(ART) as a good human companion:
a. It
simulates a real life (human- or animal-like) behavior;
b. It
is based on modeling emotional, cognitive, motor and other
mental traits and states
normally experienced by humans;
c. It
gives a person an opportunity to communicate with robots on various levels such
as tactile-
kinesthetic, sensory, emotional, cognitive, and socially – behavioral.
A second major distinction that differentiates the class of interactive
and simulative artificial creatures from other groups is that it includes
several subgroups that reflect a structure of a living world:
a.
Anthropomorphic robots or humanoids;
b.
Robots imitating living beings other than humans or fictitious creatures.
Robots of this class recently became the subject of a new field of study,
which emphasized the importance of human – robot interaction analysis into
psychological and social contexts [3] [4] [5]. The robotic
cat NeCoRo (see Schema 1), invented by Dr. Shibata [6], was a primary subject in
our pilot study [7].
In October 2001 Omron Corporation (Japan), a world-known leader in manufacturing
medical equipment, released its new product, a cat type communication robot.
NeCoRo,which means ‘companion’ in Japanese, alternatively called a mental
health robot,
Ó
Omron Corporation, 1999
Schema 1. Robotic cat NeCoRo – a
primary tool for robotherapy
was
designed as a first of its kind real-life-looking emotional creature. Besides
being a sophisticated robot with an artificial intelligence system and multiple
buit-in sensors (see – Schema 1) which provide a self-organizing behavior, an
artificial cat creates playful, natural communication with humans by mimicking a
real cat’s reactions. NeCoRo streches its body and paws, moves its tail, opens
and closes its eyes, meows, and cuddles when being touched. Therefore we agreed
that this robotic creature has all the potentials for our first robotherapy
study. NeCoRo was sponsored for this project by the founder of the VSMM Society,
Professor Takeo Ojika, who also suggested the term ‘robotherapy’ for our
approach. Below we are developing for the first time some basic definitions and
principles of robotherapy as an innovative approach for non-pharmacological
treatment of psychological problems.
3.
Robotherapy: exploring technological tools for psychological treatments
Robotherapy
is defined as a framework of person – robot interactions aimed at the
reconstruction of one’s negative experiences through the development of coping
skills mediated by technological tools in order to provide a platform for
building new positive experiences. Robotherapy
as a new research area focuses on the analysis of person - robot communications,
viewed as a complex interactive system [2], with the emphasis on psychological
evaluation, diagnosis, prognosis, and principles of non-pharmacological
treatment. The effectiveness of robotherapy influenced by both
(1) one’s past experiences, current needs and individual preferences, and (2)
an artificial partner’s non-transitive physical features and behavioral
configurations defined through the intensity of simulations and responses. In
this article we discuss the construction and procedures of a new scale (see
section 4–5) – the Person – Robot Complex Interactive Scale (PRCIS). The
scale includes (1) an assesment of a person’s individual style of
communication with the robotic cat NeCoRo; (2) a participant’s evaluation of
his/her new experiences with the robotic cat; (3) a person’s past experiences
with live pets and modern technology; and (4) a participant’s evaluation of
the robotic cat features, and also NeCoRo’s advantages and disadvantages.
Finally, in the context of robotherapy we assume that human-robot interactions
have symbolic meaning. Therefore each pattern evaluated via the newly developed
scale is assigned certain psychological value derived from a person's life
experiences, likes and dislikes, and current emotional, cognitive, and
behavioral states.
4.
Scale design and descriptions
An analysis of the simulation robots’ behavior initiated the design of
a Person – Robot Complex Interactive Scale (PRCIS). The next sections describe
in detail the concept of PRCIS, scale design, and description, as well as some
preliminary results.
4.1.Scale
design
The PRCIS was developed with the purpose of assessing of interactive
patterns between a person and robotic creatures – in our case, the cat NeCoRo.
The items were developed as a result of direct observations and literature
analysis on human – robot
communication.
Observations were recorded on video and content analysis was conducted by two
independent evaluators. After summarizing observational data and experimental
facts derived from the robotics studies coherent patterns of the robot’s
actions and responses were established. Person – robot interactions were
grouped into several frames and included: verbal, nonverbal, and emotional modes
(see – Schema 2). The overall participant’s involvement with the
interactions was specified as a separate category. Each mode was also
characterized in terms of intensity of interactions on a Lickert-type scale from
1 (lowest score) to 5 (highest score).
The general principles of human – robot interactions were modified for
the purpose of our study with the robotic cat NeCoRo, incorporating analyses of
a person’s experiences with a real pet. Pet-therapy is a well-known
non-pharmacological approach to the treatment of mood disorders, negative
emotional condition, loneliness, and depression [8].
4.2.
PRCIS structure
Part 1 and 2 are designed to assess participant's answers:
·
Part 2 serves for
examining person’s interests and experiences with pets and modern technology.
Parts 1
and 2 are to be filled out by the participant after completion of a session.
Parts 3 and 4 have to be filled out by the instructor who performs the
observational assessment of human–robot communication. Part 3 presents an
instrument for recording observations by the instructor, who monitors a
participant’s behavior during a 15-minute session. This part also contains a
standard instruction for the participant. Part 4 includes the instructor’s
overall assessment of person –
robot interactions, and is to be filled out after a session is completed.
4.3.
PRCIS description
A major part of PRCIS is its observational part that needs to be filled
in during the session. Along with four universal dimensions (see Section 4.2),
four scales and four subscales were developed:
A-2.Manipulation
subscale
B.
Verbal communication scale
C.
Animated interactions scale
D.
Emotional display scale
D-1.Negative
display subscale
D-2.Positive
display subscale
Non-verbal
communication scale describes
tactile and manipulative characterisics of person’s interactions with the
robotic pet.
Tactile
subscale
takes into account the specifics of a
participant’s tactile behavior such as touching a cat’s body with one –
two fingers or with the palm open; stroking cat’s paws, tail and ears or
playing with them. Subscale also indicates if participant touches the cat
nose-to-nose or does not touch a robot at all.
Manipulation subscale captures
specific components of participant’s manipulative behavior, for instance,
whether participant shakes cat's paw, holds robo-toy or picks it up, puts
robotic cat on the lap, hugs it, cradles or kisses.
Verbal communication scale
measures
whether a participant interacts with the robot through voice. Also it
differentiates between types of verbal behavior such as direct orders
(i.e., Sit down! Don’t do it!) or friendly engagement commands (i.e., Come to
me, please!). The scale indicates whether a participant expresses verbal
approval (i.e., Good boy! Very good!) or disapproval (i.e., You are bad! Shame
on you!); talks with robot in personalized way (i.e., What a good boy you are!)
and calls robo-toy by its name (i.e.,
Max, look at me!); keeps up with a conversation (i.e., Do you like to play with
me? – Yes, you do, don’t you?!); or reacts verbally only when the robot
become active.
Animated interactions scale
assesses how a participant perceives the robotic pet and treats it. This scale
specifies whether a participant interacts with NeCoRo in personalized way and
treats it like a real pet, but not a technological device. The following
behaviors are identified as animated interactions: participant looks directly
into robo-toy eyes or uses different objects to trigger response of the robotic
cat; performs as if activities such as feeding, toileting or putting robotic pet
asleep. A scale takes into account whether the participant creates gaming
situations by developing new ways of interacting with robo-toy (i.e., by using
paper butterfly, a feather, a ball etc.), or gives cute names such as
‘Sleepy-head’, ‘Tommy-boy’ and uses during interactions with the robotic
cat greetings (i.e., Hello!), excuses (i.e., Sorry! I didn’t mean it!),
farewells (i.e., Good bye! See you soon!), expressions of his/her feelings
toward NeCoRo (i.e., Don’t be angry with me! I like you!). The animated
interaction scale also measures negative manifestations, i.e., hits or punches.
Emotional display scale
deals with the participant’s positive and
negative emotional responses to the robot’s behavior.
Negative display subscale shows
whether the participant expresses negative feelings toward the robotic partner
or not; i.e., is
the participant disturbed by the robot presence, or not. Other parameters
identify if a participant is nervous, afraid of the robot or frustrated by its
presence, ignores the robo-toy, expresses anger toward it, or remains aloof.
Positive
display subscale
focuses on constructive emotional reactions of the
participant.
This
subscale describes a wide range of positive emotions: from curiosity, joyfulness,
and excitement to tenderness, pleasure, and playfulness. Emotions included
in this subscale differ by the intensity and specifics of their manifestation.
Curiosity
has
a low intensity in the display of positive emotions and focuses on exploration
rather than feelings.
A
curious participant examines the robot with interest, cautiously touches it, but
is not as excited about his/her interactions with NeCoRo.
Joyfulness
expresses in
participant’s behavior when he/ she is obviously glad to communicate with the
robot.
Excitement
manifests itself through such behaviors as clapping before the robot’s face,
lifting it in the air, or expressing intensive positive emotions in some other
way.
Tenderness
can be defined
as warm care provided for the robotic cat, by tenderly stroking the robotic toy
with an open palm; gently touching its paws, tail and ears; or rocking it to
sleep.
Pleasure
can be noticed in the participant’s behavior when he/ she puts NeCoRo on their
lap, hugs it, cradles or kisses.
Playfulness
has a high intensity and is recorded when the participant triggers robotic cat
responses by playing with it while using other objects, creating all kinds of
gaming situations.
5.
Procedure for performing an assessment with the Person – Robot Complex
Interactive Scale
Before the session, an informed consent should be obtained from each
participant. A session starts with the introduction of the robotic cat to the
participant. Standard instructions dictated that the researcher introduce the
robotic cat NeCoRo by its name (Max in our case), explain the duration of a
session (15 minutes) and invite the participant to interact with the robot in
any way he/she likes. After administering an introduction, instructor performs
the observational assessment twice:
·
First time –
within the first 3 minutes of a session (Beginning Evaluation)
·
Second time –
within the last 3 minutes of a session (Ending Evaluation)
If the participant asks questions before or during the session, i.e.,
"How can I interact with a robotic cat?" or "What am I allowed to
do with it?", the evaluator should try to avoid direct answers so as not to
give any clues about how to operate a robotic cat. An answer that allows a
participant to choose his/her interactive style independently would sound like
this: " You can interact with the robotic cat in any way you like", "You
can do whatever you think is appropriate", or " You may address Max in
any way that pleases you." Also, the instructor is not supposed to give any
remarks about the usage of robotic toys.
During the session, instructor uses Part 3 of the PRCIS for performing
observations of person – robot communications and, when there is a video
release as part of a standard informed consent, record the session via a
camcorder. It is important for an instructor to make notes during the session
and mark down all of the participant’s comments concerning his\her
interactions with the robotic toy (i.e., the nature of exclamations, approving
and disapproving phrases, and specific manner of manipulation that is not
captured by a PRCIS).
After a session is completed, the participant provides his assessment
(Part 1) followed by the instructor’s directive: "Now, please share with
us your impressions about the interactions with the robotic toy! What did you
like or dislike about Max?" In addition, it is also desirable to receive
from the participant his/her explanations about the choices made while working
with the PRCIS (Part 1, E – 8). The instructor also asks the participant’s
to describe their past experiences with pets and modern technologies (Part 2).
5.1
Some examples of the notes about participant’s behavior
Verbal behavior.
Participant says phrases like: " This cat is so silly! ", " This
is just a brainless peace of junk! "," All is clear with you!
"," What a pretty cat! He looks just like a real cat! ","Do
you bite? Do you get angry? Dogs and cats always bite me...", "He is
so nice! I would like to kiss him again and again!"
Participants talk to the robot in the way adults talk to newborn babies,
by using tone of voice and wording appropriate to ‘baby-talk’, i.e.,
"Max is such a good boy!"
Nonverbal
behavior.
Specifics of nonverbal behavior, manipulations, and emotions should be noted,
too. For example, if a participant:
6.
Illustrations: Robotherapy case studies
Robotherapy
has
a wide
spectrum of
psychological applications.
Developed with the
purpose of investigating diagnostic and therapeutic potentials of person–robot
interactions, PRCIS was tested on individuals of both sexes with ages ranging
from 6-year old children to 89-year old adults. Our preliminary data show that
robotherapy might be a useful tool for both research and psychological practice
[7] [9].
Case 1. Alice and Nancy: two
eight-year old girls
Alice and
Nancy are girls of the same age. Their interactions with Max (the robotic cat
NeCoRo) reveal individual differences in the manner of self-expression. During
15-minute session with Max, Alice started by cautiously touching and stroking
the cat. However, soon she began to interact with the cat in a very friendly and
caring way by creating gaming situations and expressing love and pleasure for
their new partnership. Alice tried all possible ways of communicating with the
robotic cat, except for verbal. At the end of the session she learned enough
about the robot’s skills to provoke the cat’s responses at any time. Alice
initiated interactions with the robot and actively involved the robotic cat in
her games. At the same time she was very sensitive and responsive to the
robot’s reactions and never tried to force the cat into any activities beyond
its abilities. They got along very well. At the end of a session, Alice became
so attached to the robotic cat, that she began to rub her nose against Max’s
nose while looking into its eyes, lifted him in the air, and lovingly hugged Max
and kissed him. Alice was an initiator and creator of the gaming situations and
she obviously felt pleasure and joy from interacting with the robot. However, a
distinctive feature of Alice’s interactions with the robotic cat was an
absolute absence of verbal communication on one hand, and a
display of a wide range of nonverbal behaviors with a rich variety of
positive emotions on the other. Alice demonstrated high tactile, as well as
manipulative scores, but poor levels of verbal communication. Alice determined the robot’s character as a friendly
one and provided following explanation of her choice: “Max is friendly because
he responds to my touch. I like to play with him, stroke him and watch how he
responds to me.”
Nancy chose a different strategy of interactions with the robotic cat.
She was skeptical at the beginning and started her session with Max by rubbing
behind his ears with two fingers. However, when Max responded to her touch by
slightly moving his head, Nancy backed up and said: “It’s weird! It’s
scary.” Then for a long time she just stared at the robotic cat without any
intention to touch him. She expressed nervousness by moving restlessly her legs.
She was detached from interaction and provided the following explanation by
expressing verbal criticism (while
trying to hide her fear of uncertainty in robot’s behavior): “He is so
boring! He probably does not like me. What am I supposed to do with him?”
After a while she repeated her attempt to establish a communication with Max,
this time with success. She picked the cat up, held him on her lap, stroked him
with one or two fingers, and sometimes with an open palm. But her attention was
shifting back and forth and she tried to talk about different subjects with the
instructor avoiding direct communication with the cat. Nancy demonstrated an
ambivalent style of interaction with the robotic cat, which may indicate a fear
of uncertainty and rejection.
Case
2. A 45-year old Emma and her 12 year-old son Andy with the diagnosis of
attention deficit
disorders
Andy
seemed to be very detached when he first approached Max. He
pretended that he is a researcher who studies the robot's behavior but does not
play with it. Andy turned the cat over in his hands and formally examined the
robot as a technical novelty without showing any of his emotions. Observing the
robot's behavior Andy made remarks to its every movement or sound:
“Hm-m…”,
“Uhu…” (with the connotation: “Everything is clear with you!”, “It
is understandable”, “This is what you do, don’t you!” Compared to Alice
(Case 1), who tried to amuse the robot itself, Andy expected activity and
initiative from the robotic cat. Nevertheless, at the end of his session Andy
showed emotions of higher intensity toward Max and even expressed his
compassion.
However, his
activity was still intended to provoke the robot’s initiative. Andy gave the
robotic cat his cheek and wanted Max to lick it with the “expression of
love”. Andy was annoyed with robot's indifference: “Why can't he lick me as
my dog does?!” The more emotions Andy felt toward the robot, the greater
activity and love he demanded from it, even trying
forceful manipulations with the robot's paws. Andy wanted expression
of love,
care and attention
toward himself both through nonverbal and voice channels of communication.
Andy's mother’s communication with the robot was limited to mostly
verbal behavior: “What a nice kitty! Max,
you are such an adorable kitty! Yes, you are!” From the very beginning, Emma
interacted with the robotic cat in a personalized way. She carried on
‘a personalized dialogue’ with the robot, calling Max by his name and
expressing pleasure from having a ‘conversation’ with the robotic cat.
During her session, Emma showed compassion to the robot by using baby talk. The
closer and more intimate her relationships with Max became, the more she used
baby talk in her communication.
Mother’s and son’s communicational styles with the robotic cat have a
distinct diagnostic meaning. Analysis of their interactions shows active
behavior of the mother and passive behavior of the son. The absence of tactile
behavior in their interactive profiles may reflect difficulties in their
relationship stemming from an inability to express warm feelings toward each
other (the latter was confirmed by the analysis of a therapeutic session).
Therefore, their specific way of establishing communication with the robotic cat
was a projection of the existing problems in their real life mother – son
relationships.
Case
3. Rick: a 70 year-old men
Rick felt awkward
in the presence
of the
robotic cat, as he
was trying to act aloof.
He started dialogue
with the robotic cat with the following remarks: “I am not a cat-person! But
it does not matter for the research, right?” The instructor had to re-assure
him by saying that each person communicates with the robotic cat in one's own
way. After this ‘introduction’, Rick carried on a dialogue with the robot as
if it were a real cat by asking questions: “You do not bite, do you? Are you
aggressive? Do you have sharp teeth? Well! Dogs and cats constantly bite and
scratch me! You would not bite me, would you?”
During the session, Rick never took the cat in his hands, and touched it
only with one or two fingers. Rick constantly distracted himself by talking
about or looking at other objects. He did not express any pleasure or joy while
interacting with the robotic cat, but just “participated in research.”
Rick’s slight initial interest manifested at the beginning of the session was
quickly exhausted and in five minutes was replaced with indifference and
boredom.
While indicating his interests after the session, Rick called himself an
active technology user and emphasized that he never had mutual relationships
with any pets.
Case 4. David: a 76 year-old men
From
the first minutes of the interactive session David was amazed by Max: “This
is a robotic cat ?! Well, I should admit that this is a piece of art!” He
passionately grabbed the cat with two hands and started to stroke it with love
and compassion. David was extremely curious about Max and enjoyed interactions
with the robot very much. His communication with the artificial creature from
the very beginning was personalized. David treated Max as if he were a real pet:
”I am not afraid of you! You do not have teeth and I am aware of it!” He
accompanied the cat's meows by saying: “See, I can talk in your language!
Meow! Meow!” David was exited when the cat responded to his touch: ”Oh, you
like this ! Good
boy!”
David was trying to catch Max's attention by singing him songs, calling him:
”Kitty,
Kitty!”
and rubbing under the cat's neck. David
was under the impression that Max will interact with him like a real cat. David
tried to pick the robot up by its front paws, but when Max refused to respond,
David realised it, so he would not force the robotic cat to do something against
its “will” or “ability”. David created gaming situations for robotic pet
by clapping above
the table’s surface while trying to provoke some new reactions, or by giving
the cat cute names such as “pussy cat”. Max was very interactive with David; the robotic cat
was trying to stand up on his four legs and even made an attempt to stay on two
back paws, but unfortunately almost
fall sideways.
In
sum, David tried a wide repertoire of verbal and non-verbal communications with
the cat, expressed
love and compassion to the robot, and enjoyed playing with it. Only a few
exceptions that probably require further psychological analysis occurred: David
did not call Max by his own name and never tried to turn Max over, pick up or
hold the robot in the hands, stroke its whiskers, ears or tail.
The effectiveness of robotherapy is associated with the productive
compatibility between human beings and their artificial partners, which can be
achieved through:
o friendly
design (appearance) of the robotic creatures
o adjustable
behavioral configuration of a robot
o multilevel
(tactile, sensory-motor, cognitive, emotional, and social) person - robot
interactions based on the open-loop principle.
As a new way of helping people, robotherapy is aimed at the improvement
of personal well-being through the development of various coping skills mediated
by technological tools.
8.
Acknowledgments
This
article was supported in part by the Gifu University, where the authors were
invited by Professor Ojika as Visiting Researchers at the Virtual System
Laboratory in February - March 2002. Dr. Jiska Cohen-Mansfield, Director at the
Research Institute on Aging and Professor of Health Sciences at the George
Washington University (USA), made valuable contributions to this project. Some
of our ideas were influenced by Dr. Cohen-Mansfield's work for the last 15 years
in the field of non-pharmacological treatments for elderly people with cognitive
impairments. We would also like to thank Yevgeny and Alex Glasamitsky for their
generous and continuing support of our projects in the USA.
References
[1] Asimov, I. (1963). I, Robot. New
York: Doubleday & Company, Inc., p.17
[2]
Libin, A. (2001). Virtual reality as a complex interactive system: A
multidimensional model of a person –
artificial partner co-relations. In: H. Thwaites, & L. Addison
(Eds.), Proceedings of the Seventh International
Conference on Virtual Systems and Multimedia. Enhanced realities:
Augmented and unplugged, Los Alamitos,
CA: IEEE Computer Society, pp. 652-657.
[3] Brooks,
R.A. (2002). Flesh and Machines. New York: Pantheon Books. 315 p.
[4] Shibata, T. & Tanie, K (2000).
Influence of A Priori Knowledge in Subjective Interpretation and Evaluation by
Short-term Interaction with Mental Commit Robot, IEEE/RSJ International
Conference on Intelligent Robots
and Systems, pp. 169-172.
[5] Plaisant., C., Druin, A., Lathan,
C., Dakhane, K., Edwards,K., and Vice, J.M. (2000). A Storytelling Robot for
Pediatric Rehabilitation. [Online]. Available at: http://www.cs.umd.edu/hcil/anthro
[6]
“Is this a real cat?”- A robot cat you can bond with like a real pet –
NeCoRo is born. Omron Press
&nb