53home is where the robot is
nd their way into households. ese days there is so much research
into smarter household robots, but thus far robotic developers always
come up against the fact that household tasks are still harder than we
originally thought.
Due to the complexity of the household, the household robots
are monomaniacal: they can only take over a portion of one specic
household task. For a successful launch in the market, these robots
will have to be able to fully take over and complete certain tasks, in
order to actually make household work lighter. e non-monomaniacal
weekly cleaner will probably remain more ecient for far longer than
a household robot.
2.2.3 Liability of Home Robots
e use of household robots raises no urgent ethical issues. e
main ethical issue concerns safety in the deployment of these
devices in the household—developing robust devices that do not
walk in people’s way and have a much-reduced risk of accidents.
With regards to the safety of using a household robot, the degree
of its autonomy aects the outcome of any act performed by the
robot and the possible legal consequences. Asaro (2012) sees no
legal issues relating to liability as long as the robot does not act of
its own accord—either unexpectedly or according to decisions it
reaches independently of any person. e potential harms posed by
such a robot will be covered by civil laws governing product liabil-
ity, since the robot, available for the commercial market nowadays
and in the short term, can be treated as a tool, such as a toy, weapon,
or car and so on.
This completely changes when the robot performs independent
actions or decisions, and although there is a pressing need to deal
with questions of liability, there has been little discussion about
the legal aspects of human–robot interaction in the literature. No
legislation is yet available to deal with questions of liability in
accidents involving these robots. Schaerer, Kelley, and Nicolescu
(2009) propose a creative framework for liability in human–
robot interaction based on the principle “robots as animals” for
semiautonomous robots programmed to act within a predictable
range of behavior.
54 Just ordinAry robots
On this basis, our negligence rationale becomes clear. e owner of a
semi-autonomous machine should be held liable for the negligent super-
vision of that machine, much like the owner of a domesticated animal is
held liable for the negligent supervision of that animal. Semi-autonomous
machines, like domesticated animals, are more predictable than wild
animals, but remain occasionally prone to sporadic behaviour – even in
the absence of manufacture or design defects, and even despite adequate
warnings, e.g., “Do not leave this product unattended.”
Schaerer etal. (2009, p. 75)
is framework involves a two-step analysis: (1) making a decision
about whether the product was defective. If so, then the manufacturer
is held liable and the analysis ends. If not, then the second stepis
(2)to decide whether the owner was negligent. If so, the owner is
liable. If not, the victim alone bears the cost of the accident.
For example, if a vacuum-cleaning robot knocks over and seriously
injures a baby, the manufacturer might be held liable because of a
defect of the vacuum-cleaning robot. If the robot was not defective,
we must ask whether the owner of this robot was negligent in caus-
ing the accident because the baby was a victim within the foreseeable
zone of danger. In other words, that a reasonable person would believe
that the negligent supervision caused the accident and that the acci-
dent was therefore foreseeable.
e proposed framework of Schaere, Kelly, and Nicolescu has been
developed for semiautonomous robots with a level of autonomy well
below that of humans. In the future, we may be able to create robots
that are capable of moral autonomy or even legal responsibility (see
Wallach & Allen, 2009). en many questions about legal respon-
sibility, legal liability, and legal action become relevant, because the
robot may operate beyond the control of a user. Robots, however, have
not reached this state, and it seems likely that it will be decades before
this is realized (on this subject, see Richards, 2002). erefore, there
is little point in talking about law or judicial rulings, because such
statements can only be based on expectations about the development
of robot technology and not on specic legal issues.
In 2013, the International Electrotechnical Commission (IEC)
prepared an International Standard that covers the safety aspects of
55home is where the robot is
vacuum-cleaning robots and their washing peers, IEC 60335-2-2,
since the Commission states that the safety of cleaning robots, like that
of all household appliances, is essential.* e International Standard
provides reliable guidance for ensuring safety in the design and con-
struction of vacuum-cleaning robots as well as their integration, instal-
lation, and use throughout their life cycle.
2.3 Humanoid Robots: e Companion
In this section, the focus is on companion robots. ese robots interact
with humans, which requires social intelligence based on deep models
of human cognition and social competence. e companion robot is
designed with anthropomorphism in mind, which means attributing
human characteristics and behaviors to nonhuman subjects, in our
case socially interactive robots. In robotic social interaction, an activity
shared between human and machine is the goal. e crucial purpose
of a companion robot is to work with the user to activate, entertain,
and engage the user. Companion robots thus enter into social inter-
actions with people at home. A major design goal for producers of
companion robots is that they will also be able to communicate on a
nonverbal level and will be able to transfer emotions because they are
embodied (Breazeal, Takanski, & Kobayashi, 2008). Because of the
design of the robot’s body, its bodily communication can be greatly
broadened; body messages can run parallel with spoken language, for
example, through gestures and touch, or warmth and texture, or by
exchanging facial expressions. e role of embodiment will be dis-
cussed in Section 2.3.1. Due to their mechanical outer body, robots
can also combine communicative functions to perform other tasks,
such as household chores, keeping track of calendars and shopping
lists, and reading news from the newspaper and so on. But the robot’s
behavior can also have a purely social purpose, such as companionship
or entertainment. According to the creators, the private living space is
the habitat of choice for these new social robots.
Although the companion robots that are currently available com-
municate on a very primitive level and the predictability of their
*
http://www.iec.ch/etech/2013/etech_0113/tech-2.htm.
56 Just ordinAry robots
response is high, this also motivates researchers to proceed in order
to reach a more ecient and eective interaction (Breazeal, 2003;
Heerink, Kröse, Wielinga, & Evers, 2009). Not only is a lot of knowl-
edge missing about the mechanisms that encourage communication
between humans and robots, but knowledge about how behavior occurs
between humans and robots, and even how the interaction between
people actually works, is also absent. is knowledge is critical to the
design of robots, because the success of the social robot depends on
successful interaction (Breazeal etal., 2008; Dautenhahn, 1995). is
research is still in its infancy and is being studied in the discipline of
human–robot interaction. In Section 2.3.2, we will briey describe
the state-of-the-art human–robot interaction and the focal points of
this discipline today.
Since users are strongly inclined toward anthropomorphism, these
robots quickly generate feelings (Duy, 2003). e mechanization
(rationalization) of the interaction with a human being therefore
evokes ethical questions with respect to dehumanization. is issue
will be detailed in Section 2.3.3.
2.3.1 e Robot Body
In a study by Ray, Mondada, and Siegwart (2008), it was found that
people do not like the idea that a household robot looks like a human
being, and that people prefer the robot to look like a small machine.
is machine should also match the interior of the house, so that
householders feel more comfortable adopting it (Sung, Christensen,
& Grinter, 2008). For companion robots, however, the humanoid
shape enables the robot to perform its task, that is, mainly the inter-
action with a human being, in a better way. If the companion robot
has, for example, a face, so that it is able to use his eyes, mouth, ears,
and eyebrows to express all kinds of emotions, it will be capable of
intuitive communications with human beings. Companion robots
have, however, one remarkable thing in common and that is they
do not resemble humans in appearance in the slightest. ey usu-
ally express childish or abstract life forms (see Boxes 2.1 and 2.2).
For most designers of companion robots, the key is about some kind
of communication via verbal andnonverbal means, but the appear-
ance of the robot matters less. e nonverbalmeans,however, can
57home is where the robot is
BOX 2.1 FURBY
e Furby (Figure 2.2) was designed in Japan by Tiger
Electronics and was introduced in 1998. Since its introduction,
there has been a lot of hype generated for this intelligent stued
animal. Within a year, 1.8 million items were sold, and the fol-
lowing year 14 million were sold; in total over 40 million Furbies
were sold.* In 2005, yet another improved Furby by Hasbro was
introduced, but 2years later, the Furby was taken out of pro-
duction. As usual with a craze, interest slowly ebbed and disap-
peared, leaving Furbies in the closet or even in the garbage bin.
In 2012, there was a revival. Hasbro released a new Furby, Furby
2012, with more expressive liquid crystal display (LCD) eyes, a
wider range of emotions, its own iOS and Android app and the
ability to adapt its personality in reaction to user behavior.
†
A
year later, a new Furby was released: Furby Boom that “has more
than twice as many responses as the previous Furby, remembers
the name you give it, has 5 new personalities to discover and,
on top of all that, it’s hatching a new generation!”
‡
Furby Boom
*
www.time.com/time/specials/packages/article/0,28804,2049243_2048661_2049232,
00.html.
†
www.dvice.com/archives/2012/07/furby_returns_f.php.
‡
www.hasbro.com/furby/en_us/shop/furby-boom.cfm.
Figure 2.2 Furby with one of the authors. (Photo courtesy of Rinie van Est.)
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.