ArticlePDF Available

Review: Ambient intelligence: Technologies, applications, and opportunities

Article

Review: Ambient intelligence: Technologies, applications, and opportunities

Abstract and Figures

Ambient intelligence is an emerging discipline that brings intelligence to our everyday environments and makes those environments sensitive to us. Ambient intelligence (AmI) research builds upon advances in sensors and sensor networks, pervasive computing, and artificial intelligence. Because these contributing fields have experienced tremendous growth in the last few years, AmI research has strengthened and expanded. Because AmI research is maturing, the resulting technologies promise to revolutionarize daily human life by making people’s surroundings flexible and adaptive.In this paper, we provide a survey of the technologies that comprise ambient intelligence and of the applications that are dramatically affected by it. In particular, we specifically focus on the research that makes AmI technologies “intelligent”. We also highlight challenges and opportunities that AmI researchers will face in the coming years.
Content may be subject to copyright.
Ambient Intelligence:
Technologies, Applications, and Opportunities
Diane J. Cook, Juan C. Augusto, and Vikramaditya R. Jakkula
School of Electrical Engineering and Computer Science,
Washington State University, Pullman, WA, USA
School of Computing and Mathematics,
University of Ulster, UK
Email: cook@eecs.wsu.edu, jc.augusto@ulster.ac.uk, vjakkula@eecs.wsu.edu
October 3, 2007
Abstract
Ambient intelligence is an emerging discipline that brings intelligence to our every-
day environments and makes those environments sensitive to us. Ambient intelligence
(AmI) research builds upon advances in sensors and sensor networks, pervasive com-
puting, and artificial intelligence. Because these contributing fields have experienced
tremendous growth in the last few years, AmI research has strengthened and expanded.
Because AmI research is maturing, the resulting technologies promise to revolutionarize
daily human life by making people’s surroundings flexible and adaptive.
In this paper we provide a survey of the technologies that comprise ambient intel-
ligence and of the applications that are dramatically affected by it. In particular, we
specifically focus on the research that makes AmI technologies “intelligent”. We also
highlight challenges and opportunities that AmI researchers will face in the coming
years.
Keywords: Ambient Intelligence, Artificial Intelligence, Sensors, Decision Making, Context
Awareness, Privacy
1 Introduction
Computer science is a relatively new branch of science and as such it has gone through rapid
and yet important transformations during the first decades of its existence. Those trans-
formations have produced a very interesting mix of available experiences, and expectations
which are making possible the creation and deployment of technology to ultimately improve
the way our environments help us. This technical possibility is being explored in an area
called Ambient Intelligence. Here we survey the field of Ambient Intelligence. Specifically,
we review the technologies that led to and that support research in AmI. We also provide an
1
Figure 1: A shift in people-computing power ratio.
overview of current uses of AmI in practical settings, and present opportunities for continued
AmI research.
1.1 Emergence of AmI
The European Commission first charted a path for AmI research in 2001 [65]. A significant
factor in this birth of the field of AmI is the evolution of technology. Computers were
initially very expensive as well as difficult to understand and use. Each computer was a rare
and precious resource. A single computer would typically be used by many individuals (see
Figure 1). In the next evolutionary step, many users no longer needed to take turns to use
a computer as many were able to access it simultaneously. The PC revolution in the 80s
changed the ratio to one user per computer. As industry progressed and costs dropped, one
user often was able to access more than one computer. The type of computational resources
that we have at our disposal today is dramatically more varied than a few decades ago.
Today, access to multiple computers does not necessarily just mean owning both a PC
and a laptop. Since the miniaturization of microprocessors, computing power is embedded
in familiar objects such as home appliances (e.g., washing machines, refrigerators, and mi-
crowave ovens), they travel with us outside the home (e.g., mobile phones and PDAs), and
they help guide us to and from our home (e.g., cars and GPS navigation). Computers that
perform faster computation with reduced power and tailor the computation to accomplish
very specific tasks are gradually spreading through almost every level of our society. This
widespread availability of resources sparked the realization of Ambient Intelligence.
Possessing the necessary supporting technology is not enough for an area of science to
flourish. User’s experiences with computers over recent decades have created an interesting
context where expectations of these systems are growing and people’s fear of using them
has decreased. Concomitantly with this difference in the way society perceives technology
there is also a change in the way services are handled. An important example of this
is the decentralization of health care and development of health and social care assistive
2
technologies. Because governments and health professionals are departing from the hospital-
centric health care system, the way is paved for AmI systems to support caring for patients
closer to home, within their communities. Developments, competencies and drivers are
converging at the same time in history and all of the necessary components are in place: the
need to distribute technology around us, the will to change the way our society interacts
with technology, the available technological knowledge and all the elements to satisfy the
demand.
The idea of Ambient Intelligence is not new, but what is new is that we can now seriously
think about it as a reality and as a discipline with a unique set of contributions. Most of
us have come across science fiction movies where doors opened when someone approached
or computers were able to identify the interlocutor without their name being explicitly
mentioned. Some of those features were far fetched for the technology available at the
time, but gradually some features that indicate sensible autonomy on behalf of the system
were targeted by industry, and AmI was born.
Technically, many of us today live in homes that were considered “smart” by 1960s
standards, and for a very reasonable cost. Thermostats and movement sensors that control
lighting are commonplace. Now the bar has moved much higher: even the ability to link
movement sensors to a security alarm for detecting intruders will not impress a society which
regularly interacts with such facilities.
Recent computational and electronic advances have increased the level of autonomous
semi-intelligent behavior exhibited by systems like smart homes so much that new terms
like Ambient Intelligence started to emerge [2, 5, 65, 182]. The basic idea behind Ambient
Intelligence (AmI) is that by enriching an environment with technology (e.g., sensors and
devices interconnected through a network), a system can be built such that acts as an
“electronic butler”, which senses features of the users and their environment, then reasons
about the accumulated data, and finally selects actions to take that will benefit the users in
the environment.
1.2 What is AmI?
Ambient Intelligence has been characterized by researchers in different ways. These defini-
tions, summarized in Table 1.2, highlight the features that are expected in AmI technologies:
sensitive, responsive, adaptive, transparent, ubiquitous, and intelligent.
From these definitions, and the features that we are using (summarized in Table 1.2) to
characterize Ambient Intelligence, we can see how the discipline compares and contrasts with
fields such as pervasive computing, ubiquitous computing, and artificial intelligence. The
fact that AmI systems must be sensitive,responsive, and adaptive highlights the dependence
that AmI research has on context-aware computing.
Similarly, the AmI feature of transparency is certainly aligned with the concept of the
disappearing computer. This methodological trend was envisioned by Weiser [168], who
stated:
“The most profound technologies are those that disappear. They weave themselves
into the fabric of everyday life until they are indistinguishable from it.”
3
Definition S R A T U I
A developing technology that will increasingly make our √ √
everyday environment sensitive and responsive to our
presence. [3]
A potential future in which we will be surrounded by √ √ √ √
intelligent objects and in which the environment will
recognize the presence of persons and will respond to it
in an undetectable manner. [65]
“Ambient Intelligence” implies intelligence that is all
around us. [102] √ √
The presence of a digital environment that is sensitive, √ √ √
adaptive, and responsive to the presence of people. [132]
A vision of future daily life..contains the assumption √ √
that intelligent technology should disappear into our
environment to bring humans an easy and entertaining
life. [38]
A new research area for distributed, non-intrusive, and √ √
intelligent software systems [141]
In an AmI environment people are surrounded with networks √ √ √ √
of embedded intelligent devices that can sense their
state, anticipate, and perhaps adapt to their needs. [163]
A digital environment that supports people in their √ √
daily lives in a nonintrusive way. (Raffler) [133]
A digital environment that proactively, but sensibly, assists people in their daily lives. [17] √ √ √ √
Table 1: Features of Ambient Intelligence captured by AmI definitions. Features include
Sensitive (S), Responsive (R), Adaptive (A), Transparent (T), Ubiquitous (U), and Intelligent
(I).
4
Figure 2: Relationship between AmI and contributing technologies.
A recent description of the state of the art in this area of research is provided by Streitz and
Nixon [158].
The notion of a disappearing computer is directly linked to the notion of “Ubiquitous
Computing” [169], or “Pervasive Computing” as IBM later called it [85]. Some technical
publications equate Ubiquitous Computing, Pervasive Computing, or Everyware Computing
[63] with Ambient Intelligence. The nature of Ubiquitous or Pervasive Computing is captured
in part by the Oxford Dictionary definition of ubiquitous:
Ubiquitous: adj. present, appearing, or found everywhere [124]
and of pervasive:
Pervasive: adj. (esp. of an unwelcome influence or physical effect) spreading
widely throughout an area or a group of people [124]
Note that while Ambient Intelligence incorporates aspects of context-aware computing,
disappearing computers, and pervasive / ubiquitous computing into its sphere, there is also
an important aspect of intelligence in this field. As a result, AmI incorporates artificial
intelligence research into its purview, encompassing contributions from machine learning,
agent-based software, and robotics. As Maeda and Minami point out, AmI research can
include work on hearing, vision, language, and knowledge, which are all related to human
intelligence, and there is where AmI differs from ubiquitous computing [102]. By drawing
from advances in artificial intelligence, AmI systems can be even more sensitive, responsive,
adaptive, and ubiquitous. We characterize AmI technologies as those that exhibit character-
istics listed in Table 1. The review that we offer in the next section of the paper summarizes
advances that have been made in related areas that contribute to the goal of AmI systems
that we have set forth.
2 Contributing Technologies
From its definition, we can see that AmI has a decisive relationship with many areas in
computer science. We organize the contributing technologies into five areas, shown in Fig-
ure 2. A key factor in AmI research is the presence of intelligence. We adopt the notion
of an intelligent agent as defined by Russell and Norvig [145]. As such, the AmI algorithm
5
Figure 3: Ambient intelligence interaction with the environment.
perceives the state of the environment and users with sensors, reasons about the data using
a variety of AI techniques, and acts upon the environment using controllers in such a way
that the algorithm achieves its intended goal. The process is illustrated in Figure 3. Hence
we focus on technologies that assist with sensing, reasoning, and acting.
On the other hand, while AmI draws from the field of AI, it should not be considered
synonymous with AI. The IST Advisory Group lists five key technologies that are required
to make AmI a reality [65]. Two of these technologies clearly fall outside the typical scope of
AI research and are addressed separately in this survey. These are human-centric computer
interfaces and secure systems and devices. Next we discuss recent work in some of these
contributing areas that enhance development of AmI.
3 Sensing
Because Ambient Intelligence is designed for real-world, physical environments, effective use
of sensors is vital. Without physical components that allow an intelligent agent to sense and
act upon the environment, we end up with theoretical algorithms that have no practical use.
Sensors are the key that link available computational power with physical applications.
Ambient Intelligence algorithms rely on sensory data from the real world. As Figure 3
shows, the software algorithm perceives the environment and uses this information to reason
about the environment and the action that can be taken to change the state of the environ-
ment. Perception is accomplished using a variety of sensors. Sensors have been designed for
position measurement [170], for detection of chemicals and humidity sensing [42], and to de-
termine readings for light, radiation, temperature, sound, strain, pressure, position, velocity,
and direction, and physiological sensing to support health monitoring [116, 153]. Sensors are
typically quite small and thus can be integrated into almost any AmI application.
Wireless sensor network research has become a popular area of research in recent years
[15, 137]. The sensor networks community has explored applications such as environmental
monitoring, situational awareness, and structural safety monitoring [58, 105]. A challenge
that is prevalent particularly with wireless sensors and wireless sensor networks such as the
popular Motes platform [14] is resource management to support long-term data collection.
6
Most work in sensor networks has required battery power. For many applications, it is
inconvenient to frequently replace batteries. Finding effective alternatives to battery power
for sensor networks, however, is an ongoing research direction.
Making sense of sensor data is a complex task. Sensor data comes with unique features
that challenge conventional data analysis techniques. They generate large volumes of mul-
tidimensional data, defying attempts to manually analyze it. If the sensors are imprecise
the data can be noisy, and if a sensor fails there may be missing values. Sensor data often
needs to be handled on the fly or as streaming data [101], and the data may have a spatial
or temporal component to it.
When analyzing sensor data, AmI systems may employ a centralized or distributed model
[10]. Sensors in the centralized model transmit data to a central server, which fuses and an-
alyzes the data it receives. In the distributed model, each sensor has onboard processing
capabilities and performs local computation before communicating partial results to other
nodes in the sensor network. The choice of model will have a dramatic effect on the com-
putational architecture and type of sensor that is used for the task [19, 80]. In both cases,
sensor data is collected from disparate sources and later combined to produce information
that is more accurate, more complete, or more insightful than the individual pieces. Kalman
filters are a common technique for performing sensor data fusion [26]. Probabilistic ap-
proaches have also been effective for modeling sensors [49, 70] and combining information
from disparate sources [20, 103].
4 Reasoning
Sensing and acting provide links between intelligent algorithms and the real world in which
they operate. In order to make such algorithms responsive, adaptive, and beneficial to users,
a number of types of reasoning must take place. These include user modeling, activity
prediction and recognition, decision making, and spatial-temporal reasoning.
4.1 Modeling
One feature that separates general computing algorithms from those that are responsive to
the user is the ability to model user behavior. If such a model can be built, it can be used
to customize the behavior of the AmI software toward the user. If the model results in
an accurate enough baseline, the baseline can provide a basis for detecting anomalies and
changes in resident patterns. If the model has the ability to refine itself, the environment
can then potentially adapt itself to these changing patterns. In this overview we characterize
AmI user modeling approaches based on three characteristics: a) The data that is used to
build the model, b) The type of model that is built, and c) The nature of the model-building
algorithm (supervised, unsupervised).
The most common data source for model building is low-level sensor information. This
data is easy to collect and process. However, one challenge in using such low-level data is
the voluminous nature of the data collection. In the MavHome smart home project [176],
for example, collected motion and lighting information alone results in an average of 10,310
events each day. In this project, a data mining pre-processor identifies common sequential
7
patterns in this data, then uses the patterns to build a hierarchical model of resident behavior.
Loke [97] also relies upon this sensor data to determine the resident action and device state,
then pulls information from similar situations to provide a context-aware environment. Like
the MavHome project, the iDorm research conducted by Doctor, et al. [48] focuses on
automating a living environment. However, instead of a Markov model, they model resident
behavior by learning fuzzy rules that map sensor state to actuator readings representing
resident actions.
The amount of data created by sensors can create a computational challenge for model-
ing algorithms. However, the challenge is even greater for researchers who incorporate audio
and visual data into the resident model. Luhr [99] uses video data to find intertransaction
(sequential) association rules in resident actions. These rules then form the basis for identi-
fying emerging and abnormal behaviors in an Intelligent Environment. Brdiczka, et al. [27]
rely on speech detection to automatically model interacting groups of individuals. Moncrieff
[112] also employs audio data for generating resident models. However, such data is com-
bined with sensor data and recorded time offsets, then used to sense dangerous situations
by maintaining an environment anxiety level.
Identifying social interactions has been a common theme in AmI research. In addition
to the work of Brdiczka, Laibowitz, et al. [87] have also used wireless sensor networks to
analyze social dynamics in large meetings. They have been able to detect key interaction
characteristics such as interest and affiliation from sensor data in groups of over 100 people.
4.2 Activity Prediction and Recognition
A second contribution that reasoning algorithms offer is the ability to predict and recognize
activities that occur in AmI environments. Much of this work has occurred in smart envi-
ronments research, where the AmI application is focused on a single environment which is
outfitted with sensors and designed to improve the experience of the resident in the envi-
ronment [36]. The Neural Network House [113], the Intelligent Home [93], the House n [109]
and the MavHome [39, 181] projects adaptively control home environments by anticipating
the location, routes and activities of the residents (i.e., a person moving within an AmI
space). Prediction algorithms have been developed for both the single [60] and the multiple
[144, 143] resident cases. Predicting resident locations, and even resident actions, allows
the AmI system to anticipate the resident’s needs and assist with (or possibly automate)
performing the action [73].
The modeling techniques described so far can be characterized as unsupervised learn-
ing approaches. However, if prelabeled resident activity data is available, then supervised
learning approaches can be used to build a model of resident activity and use it to recog-
nize observed activities. Such recognition can aid in providing support for the activities or
reminding individuals to complete activities that they normally perform or that are impor-
tant for the individual. Muehlenbrock, et al. [114] combine unsupervised learning with a
naive Bayes learner to identify an individual’s activity and current availability based on data
such as PC/PDA usage. Tapia, et al. [159] also employ a naive Bayes learner to identify
resident activity from among a set of 35 possible classes, based on collected sensor data.
Other researchers have learned Markov models to perform activity recognition [171] using
just location information [95] or enhancing the model with object interaction data [128] or
8
data collected from wearable devices such as accelerometers and heart rate monitors [115].
4.3 Decision Making
Over the last few years, supporting technologies for Ambient Intelligence have emerged,
matured, and flourished. Building a fully automated AmI application on top of these foun-
dations is still a bit of a rarity. Automated decision making and control techniques are
available for this task. Simpson, et al. [151] discuss how AI planning systems could be em-
ployed to not only remind individuals of their typical next daily activity, but also to complete
a task if needed. Augusto and Nugent [16] describe the use of temporal reasoning with a
rule-based system to identify hazardous situations and return an environment to a safe state
while contacting the resident.
Few fully-implemented applications decision making technologies have been implemented.
One of the first is Mozer’s Adaptive Home [113], which uses a neural network and a reinforce-
ment learner to determine ideal settings for lights and fans in the home. This is implemented
in a home setting and has been evaluated based on an individual living in the Adaptive Home.
Youngblood, et al. [179] also use a reinforcement learner to automate physical environments
with volunteer residents, the MavPad apartment and the MavLab workplace.
The iDorm project of Hagras, et al. [66] is another of these notable projects that has real-
ized a fully-implemented automated living environment. In this case, the setting is a campus
dorm environment. The environment is automated using fuzzy rules learned through obser-
vation of resident behavior. These rules can be added, modified, and deleted as necessary,
which allows the environment to adapt to changing behavior. However, unlike the reinforce-
ment learner approaches, automation is based on imitating resident behavior and therefore
is more difficult to employ for alternative goals such as energy efficiency.
Amigoni, et al. [13] employs a Hierarchical Task Network (HTN) planner to generate
sequences of actions and contingency plans that will achieve the goal of the AmI algorithm.
For example, the AmI system may respond to a sensed health need by calling a medical
specialist and sending health vitals using any available device (cell phone, email, or fax). If
there is no response from the specialist, the AmI system would phone the nearest hospital
and request ambulance assistance.
4.4 Spatial and Temporal Reasoning
Very little can be done within an AmI system without an explicit or implicit reference to
where and when the meaningful events occurred. For a system to make sensible decisions it
has to be aware of where the users are and have been during some period of time. These in-
sights, together with other information, will provide important clues on the type of activities
the user is engaged in and the most adequate response.
Spatial and temporal reasoning are two well established areas of AI [56]. They have been
the subject of intense research for a couple of decades and there are well known formalisms
and algorithms to deal with spatial, temporal, and spatio-temporal reasoning. Gottfried et
al. [61] has shown how the traditional frameworks for spatial reasoning and for temporal
reasoning can be used to have a better understanding of the activities in an AmI application.
In an environment such as an airport or a home, for example, such reasoning can be used to
9
AtKitchen Not AtKitchen
151413 164321 5
...
AtReception
CookerOn
Figure 4: Detecting hazards in the kitchen.
analyze trajectories of people within a room and classify them as “having a clear goal” or
“being erratic” [96].
Both dimensions, space and time, are useful to understand key elements of a situation
under development. For example, lets assume we are monitoring activities in order to pre-
vent hazardous situations at home. Whenever someone turns on the cooker and leaves it
unattended for more than 10 units of time, then the system has to take action (turning off
the cooker automatically or warning the user, U). Consider a scenario in which the AmI
environment sensed the cooker has been turned on, after which a sequence of sensor signals
(e.g., movement sensors combined with RFID sensors) was captured detecting the location
of Umoving from the kitchen to a reception area and then into the bedroom. Finally, the
bed occupancy sensor (a pressure pad) detects the person is in bed. By the point at which
the person is in bed (time 13 since the cooker was turned on) the condition that more than
10 units have elapsed since the person turned the cooker on without returning to the kitchen
is satisfied. All the conditions will be fulfilled for the warning rule to be triggered. This
situation is pictured in Figure 4.
Lets consider another situation in which the doorbell has been rung and the resident
does not respond within 5 minutes. However, the AmI system detects that the person is at
home and knows the resident is not hearing impaired. This can be identified as a potential
emergency and may trigger a procedure where caregivers are notified and will try to contact
the individual visually or by telephone.
Situations of arbitrary complexity can be detected by using language which allows the
specifications of situations involving repetitions, sequences, frequencies and durations of
activities related to the activities of entering to rooms or moving from one room to the next
one [57]. In [16] such a language is used to integrate both concepts in the same formalism
and to obtain spatio-temporal reasoning combined with active databases in the identification
of interesting situations like those described above.
An alternative formalism for reasoning about time is based on Allen’s temporal logic [12].
Allen suggested that it is more common to describe scenarios by time intervals than by time
points, and defined thirteen relations that comprise a temporal logic: before, after, meets,
meet-by, overlaps, overlapped-by, starts, started-by, finishes, finished-by, during, contains,
and equals. Jakkula, et al. [79] found that these temporal relations play a beneficial role
in prediction and anomaly detection for ambient environments. Consider, as an example,
a medicine compliance tool that makes sure an elderly person consistently takes pills right
10
Figure 5: Boundary conditions for nine of Allen’s temporal intervals. These nine have been
used for event prediction and anomaly detection [79].
after eating food. The two activities are related by the “after” relationship. When the
relationship is violated, the system can respond with a reminder for the individual. The
nine intervals that were used for prediction and anomaly detection in Jakkula’s TempAl
algorithm are shown in Figure 5.
5 Acting
AmI systems tie reasoning to the real world through sensing and acting. Intelligent and
assistive devices provide a mechanism by which AmI systems can executive actions and affect
the system users. Another mechanism is through robots. Relationships between human and
machines have been explored extensively in science fiction stories. However as Turkle points
out [160], watching children and the elderly now commonly interact tenderly with robot pets
brings “science fiction into everyday life and technophilosophy down to earth”. Research in
robotics has progressed to the point where users need no longer wrestle with how to give
them to move to a specified location, but instead can formulate requests such as “bring me
the medicine on the counter”. Indeed, such robot assistants are already found in nursing
homes [135] and provide an outlet for nurturing contact for the elderly.
Robots are able to provide an even wider range of assistive tasks to support AmI. They
can monitor the vital signs of their masters and provide conversational stimulation. Robots
are now capable of exhibiting much more human-like emotions and expressions than in the
past [29] and can even influence human decision. One such case is the museum traffic control
project [50], where a robot generated cues that caused visitors to travel to portions of the
museum that were normally avoided. Robots provide AmI systems with self-mobility and
human-likeness, which facilitates human interaction and allows the influence of AmI to more
greatly pervade human culture.
11
6 Human-Computer Interaction
A characteristic that the IST Advisory Group highlighted as necessary to further societal
acceptance of AmI [65] is that AmI should be made easy to live with. This is further detailed
as a need to define human-centric computer interfaces that are context aware and natural.
Here we highlight some recent advances in these areas.
6.1 Context Awareness
Models of 21st century ubiquitous computing scenarios [168] depend not just on the devel-
opment of capability-rich mobile devices (such as web-phones or wearable computers), but
also on the development of automated machine-to-machine computing technologies, whereby
devices interact with their peers and the networking infrastructure, often without explicit
operator control. To emphasize the fact that devices must be imbued with an inherent
consciousness about their current location and surrounding environment, this computing
paradigm is also called sentient [76] or context-aware computing.
“Context (e.g., location and activity) awareness” is a key to building Ambient Intelligence
and associated applications. If devices can exploit emerging technologies to infer the current
activity state of the user (e.g., whether the user is walking or driving, whether he/she is at
office, at home or in a public environment) and the characteristics of their environment (e.g.,
the nearest Spanish-speaking ATM), they can then intelligently manage both the information
content and the means of information distribution. For example, the embedded pressure
sensors in the Aware Home [123] capture residents’ footfalls, and the home uses these data
for position tracking and pedestrian recognition.
Research in context-aware computing includes mechanisms of determine a user’s context
even with imperfect information [45] and designing context services as found in IBM’s Con-
text Sphere [92]. Providing this type of context-aware infrastructure makes it possible to
design office spaces that smoothly move information between displays, walls, and tables [142]
and learn to customize lighting and temperature based on learned resident preferences [175].
Cheverst, et al. [32] have built upon these capabilities to design a location-aware electronic
tourist guide.
6.2 Natural Interfaces
An important aspect of AmI has to do with interaction. On one side there is a motivation
to reduce the human-computer interaction (HCI) [46]. The system is supposed to use its
intelligence to infer situations and user needs from the recorded activities, much as a butler
observes activities unfold with the expectation of helping when (and only if) needed. This
is the idea of an “intelligent social user interface” [162]. On the other hand, a diversity of
users may need or voluntarily seek direct interaction with the system to indicate preferences,
needs, etc. HCI has been an important area of study since the inception of computers.
Today, with so many gadgets incorporating computing power of some sort, HCI continues
to thrive as an important area that prevents AmI technologies from becoming “ubiquitous
clutter” [147].
12
Although designers of Ambient Intelligence systems are encouraged by the progress that
has been made in the field over the last few years, much of this progress will go unused
if the technologies are difficult or unnatural for residents. Abowd and Mynatt [7] point
out that explicit input must now be replaced with more human-life communication capa-
bilities and with implicit actions. The maturing of technologies including motion tracking,
gesture recognition [126], facial expression recognition [125] and emotion recognition [117],
speech processing [173], and even whistle processing [52] facilitate natural interactions with
intelligent environments. In some cases, diverse interface mechanisms are combined to form
multi-modal interfaces [9, 30].
Work on natural interfaces for Ambient Intelligence has taken AmI applications out of
single rooms and buildings to even richer settings. UCLA’s HyperMedia Studio project [107]
adapts light and sound on a performance stage automatically in response to performers’ posi-
tions and movements. The driver’s intent project at MIT [126] recognizes driver’s upcoming
actions such as passing, turning, stopping, car following, and lane changing by monitoring
hand and leg motions. The use of facial expression recognition enhances the automobile by
recognizing when the driver is sleepy, or change the classroom interaction when detecting
that the students are bored or confused. New Songdo City, a “ubiquitous city” being built in
South Korea, is implementing many AmI ideas on a city-wide scale [121]. Such a large-scale
sharing of data facilitates easy access to city resources for residents.
Images also help assess a situation where safety can be compromised. The Wireless Sensor
Networks Lab at Stanford University uses a network of video cameras to infer a sequence of
body postures (Figure 6) and hence detect possible hazards like a fall [84].
Images can be also used as in visual arts. Now more than ever, art can be “experienced”.
One example of the use of AmI to transform the way people relate and understand their
environment is being implemented through the UNSEEN project [23]. A nature interpreta-
tion center set in eastern Qu´ebec where real-time images of native plants are examined and
used by the system to present the plants and their current state of development through
challenging and original perspectives.
7 Privacy and Security Challenges
Ambient Intelligence offers great benefits to users by customizing their environments and
unobtrusively meeting their needs. As Brey points out [28], AmI potentially gives more
control to humans by making their environments more responsive to intended actions, by
supplying humans with customized information, and by reducing the cognitive or physical
effort that is required to perform a task. At the same time, AmI can take away control when
the environment performs the wrong action, when it forces humans to perform extra or cor-
rective actions, when it shares information with third parties, and when it gives monitoring
and data collection access to third parties.
Wright [172] argues that delivering personalized services opens up the possibility for the
corresponding personal information to be stored and shared. As Bohn, et al. [24] point
out, the sayings that “the walls have ears” and “if these walls could talk” have become a
reality which is disturbing to many. In fact, in a 2003 survey [139], respondees indicated
that privacy protection was more important to them than any potential benefits provided
13
Figure 6: (a) Sample images from two cameras showing different postures. (b) Elliptical
model representations with average motion vectors for the moving body parts. Images pro-
vided by the Stanford Wireless Sensor Networks Lab.
14
by technologies found in Ambient Intelligence applications. In addition, what is considered
obtrusive and privacy invading differs by age group and culture [74]. For example, one reason
that U-City in Korea is developing so quickly is that there is a quicker acceptance of loss of
privacy by residents there [121].
The use of image processing through video cameras as a potential kind of sensor is a
controversial area. Naturally the amount of information that can be collected in that way is
very valuable in terms of assessing a situation. On the other hand, it raises clear issues of
privacy and the “big brother” syndrome. Still there are applications where users think the
benefits out weight the drawbacks and are decided to accept it as part of the system that
is build to benefit them. One such example is the image processing system that recognizes
hand based gestures that can be used to give orders to a system and control several different
appliances in an easy way without individual remote control units [47].
Not all image processing techniques compromise privacy. For example, in [68] a system is
reported which monitors the top of a cooking unit scanning with a camera capable to process
images from a thermal perspective. If the cooker has been left unattended for an important
length of time and the image processing unit can classify the warmth emanating from the
cooker into a dangerous level it will trigger an alarm. It is important the way the image
is used and the level of acceptance the user has for the successful use of this technology.
On the other hand, non-camera sensors do not necessarily perform a better job of ensuring
privacy. As Bohn, et al. [24] argue, individual models, even seemingly innocuous ones such
as walking patterns and eating habits, can be combined to provide very detailed information
on a person’s identify and lifestyle.
In addition to intentional privacy violations, Ambient Intelligence technologies can raise
other security issues [43]. At the sensor level, sensor reliability, handling errors, and installa-
tion errors can create security risks. To ensure security in sensor networks, the designer must
consider these factors together with sensor communication channel reliability and security,
and sensor data security. While encrypting collected data can address some of the data
privacy issues, the challenge is to implement the required security using minimal resources.
There is a great deal of research being investigated to mitigate the privacy and security
risks of Ambient Intelligence. Some of these projects focus on keeping sensed data such as
location information private [33], while other projects are designing devices that can act
as secure keys for providing and receiving personal information [184]. In lieu of transport-
ing specialized devices, biometric information can be used to access sensors and collected
information [165].
As Joinson, et al. [82] reveal from their survey of potential AmI users, privacy is a
preference that should be customizable by users. There are situational aspects of AmI
environments that trigger different privacy concerns in different people. As a result, privacy
should be a decision that is influenced by context. This type of approach is advocated by
Preuveneers, et al. [138], who are designing such context-driven privacy measures. Their
solution is to obtain the minimal amount of personal information that is needed to achieve the
user’s goal. Detailed personal information can be reduced by inferring needed information
from previously-processed data, and the impact of obtaining personal information on the
user’s goal can be analyzed to determine whether the information should be obtained.
15
8 AmI Applications
There are many settings in which Ambient Intelligence can greatly impact our lives. Some
of these applications have already been pursued by AmI researchers. In this section, we
highlight current AmI applications. By summarizing existing implementations, we also draw
attention to the technologies that are necessary to create the implementations and the chal-
lenges that AmI researchers still face. It is important to note that not all of the applications
described in this section embody the six features of AmI systems that we listed in Table 1.2.
However, they all reflect a subset of the AmI features. Perhaps even more importantly, the
ideas presented in the applications themselves may spark the creation of future AmI solutions
that do reflect all of our defined AmI characteristics.
8.1 Smart Homes
An example of an environment enriched with Ambient Intelligence is a “smart home”. Several
artifacts and items in a house can be enriched with sensors to gather information about
their use and in some cases even to act independently without human intervention. Some
examples of such devices are electrodomestics (e.g., cooker and fridge), household items (e.g.,
taps, bed and sofa) and temperature handling devices (e.g., air conditioning and radiators).
Expected benefits of this technology can be: (a) increasing safety (e.g., by monitoring lifestyle
patterns or the latest activities and providing assistance when a possibly harmful situation
is developing), (b) comfort (e.g., by adjusting temperature automatically), and (c) economy
(e.g., by controlling the use of lights). This is a popular use of many technologies such as
active badges [167] and indoor positioning systems [71].
In addition to investigating intelligent devices in a home, an example of Ambient Intel-
ligence is allowing the home itself to possess intelligence and make decisions regarding its
state and interactions with its residents. There are several physical smart homes that have
been designed with this theme in mind. The MavHome project treats an environment as an
intelligent agent, which perceives the environment using sensors and acts on the environment
using powerline controllers [177].
At the core of its approach, MavHome observes resident activities as noted by the sensors.
These activities are mined to identify repetitive patterns and compression-based predictors
are employed to identify likely future activities. The results from these two algorithms are
employed in building a hierarchical Markov model of the resident and the environment,
based on which a policy can be learned for automating environmental control. Initially the
approach was evaluated for its ability to predict and automate daily interactions with the
environment that the resident would typically perform manually (e.g., turn on the overhead
light when entering the apartment). From one month of data collected on a volunteer
resident, MavHome was able to reduce the needed daily interactions by 76%, on average
[180].
The Gator Tech Smart House is built from the ground up as an assistive environment
to support independent living for older people and individuals with disabilities. The home
is equipped with a large number of sensors and actuators, and generates a large volume
of data streams [72]. Data streams are filtered through an OSGi service bundle, providing
opportunity for data folding, modeling, and encryption [11].
16
Figure 7: Example smart homes: the MavHome (upper left) iDorm (upper right), Gator
Tech Smart House (middle), Aware Home (lower left), and Domus Lab (lower right)
17
The Gator Tech project currently uses a self-sensing service to enable remote monitoring
and intervention caregivers of elderly persons living in the house. The application is a
classical example that demonstrates the tension found between two noble goals: preserving
privacy and providing useful smart environment benefits.
The University of Essex’s intelligent dormitory (iDorm) [48] is a real AmI test-bed com-
prised of a large number of embedded sensors, actuators, processors and networks in the
form of a two bed roomed apartment. It is a full-size domestic apartment containing the
usual rooms for activities such as sleep, work, eating, washing and entertaining.
A common interface to the iDorm and its devices is implemented through Universal Plug
and Play (UPnP), and any networked computer running a standard Java process can access
and control the iDorm directly [66]. Fuzzy rules are learned from observing resident activities
[67] and are used to control select devices in the dorm room.
The Aware Home [7] project has been developed by the Georgia Institute of Technol-
ogy. This home consists of two identical but independent living spaces, each one with:
two bedrooms, two bathrooms, one office, kitchen, dining room, living room and laundry
room. There is also a shared basement with a home entertainment area and control room for
centralized computing services. The house has been built using standard construction tech-
niques. Some of the technology deployed in the house are human position tracking through
ultrasonic sensors, RF technology and video, recognition through floor sensors and vision
techniques.
One of the applications of the tracking and sensing technologies in the Aware Home
is a system for finding Frequently Lost Objects such as keys, wallets, glasses, and remote
controls. The system uses RF tags attached to each object the user would like to track and
a long-range indoor positioning system to track these objects. The user will interact with
the system via LCD touch panels. The system will guide the user to the lost object using
spatialized audio cues (e.g., ”Your keys are in the bedroom.”). Other tracking technologies
in the house can assist with the task of locating objects.
The DOMUS lab is based at the Computer Science Department of the University of
Sherbrooke (Quebec, Canada) and it has been in operation since 2003. It is run by a
multidisciplinary team and one of the main aims of the lab is to achieve an implementation
of smart homes based on pervasive assistants which can provide mobile orthosis [134].
The three main lines of investigation within DOMUS are [25]: a) Pervasive Cognitive
Assistant: provide assistance adapted to specific cognitive deficits (memory, initiation, plan-
ning, attention, b) Cognitive Modeling: describe the specific behaviors of the cognitive
impaired people using descriptive representations, c) Mobile Orthose: Help people to man-
age their ADLs outside and allow caregivers to monitor them and collect ecological data on
symptoms and medication side-effects.
Recognizing the emerging popularity of smart homes and their benefits, several industry-
led projects are also developing smart homes. Siemens [150] has invested in smart homes that
provide services to enhance entertainment, security and economy. Energy saving, lighting
control, networked home entertainment, and safety devices for the monitoring of children
are some of the features that Siemens advertises. Touch screens can be used to operate the
central control unit and units can be remotely activated and controlled, for example by using
the mobile phone. Areas of application for Siemen’s research vary from adaptive offices and
Smart Homes to intelligent cars [21].
18
Philips [131] has already developed smart homes for the market that highlight innovative
technology on interactive displays. For several years, the company has been overseeing the
HomeLab at Eindhoven (NL) [41]. Research conducted at the HomeLab has been focused
on interaction and how the houses of today can increase their support to daily living from
three perspectives: a) Need to belong and share experiences, b) Need for thrills, excitement
and relaxation, and c) Need to balance and organize our lives [130]. One important aspect of
Philips research is the level of social awareness that has to be embedded in the AmI system
to be adequate and acceptable to users, in particular to elderly people [44]. The company
has been very active in the market [4].
Microsoft also has a laboratory devoted to the research on the interaction of humans
with systems and the use of artificial intelligence to support daily life activities. Some of the
topics that have been investigated are related to the availability of users and the suitability
of interrupting them [77].
These by no means are the only Smart Home projects being developed throughout the
world and there are significant developments in many regions of the world. There is a long
list of projects being currently developed in many other countries, especially Japan and
Korea. We address the readers to other sources of literature (e.g., [59, 111, 119, 183]) for
more details.
8.2 Health Monitoring and Assistance
There are many potential uses for an Intelligent Environment. Indeed, we anticipate that
features of Intelligent Environments would pervade our entire lives. They will automate
aspects of our life, increase the productivity at work, customize our shopping experiences,
and accomplishing all of these tasks will also improve the use of resources such as water and
electricity. In this section we focus on one class of applications for Ambient Intelligence:
health monitoring and assistance.
One reason for singling out this topic is the amount of research activity found here, as
well as the emergence of companies with initiatives to bring intelligent elder care technologies
into the home [78, 120, 90]. Another reason is the tremendous need for research on Ambient
Intelligence to support the quality of life for individuals with disabilities and to promote
aging in place. The need for technology in this area is obvious from looking at our current
and project future demographics. By 2040, 23% of the population will be 65+ [89] and
over 11 million people will suffer from dementia related to Alzheimer’s disease [75], with the
long-term projected total losses to the US economy expected to be nearly 2 trillion dollars
[51].
Given the costs of US nursing home care (approximately $40,000 a year with an addi-
tional $197 billion of free care being offered by family members) [62] and the importance that
Americans place on wanting to remain in their current residence as long as possible [64], use
of technology to enable individuals with cognitive or physical limitations to remain in their
homes longer may be more cost effective and promote a better quality of life. Placement
in nursing homes may sometimes be premature because of family concerns related to safety
issues [152] and AARP reports [1] strongly encourage increased funding for home modifica-
tions that keep older adults independent in their own homes. The need for this technology
is not limited to the United States: The Commission of the European Communities [34]
19
indicates that early patient discharge from hospitals due to introduction of mobile health
monitoring would save 1.5 billion Euros each year in Germany alone.
With the maturing of supporting technologies, at-home automated assistance can allow
people with mental and physical challenges to lead independent lives in their own homes
[136, 146, 155] and reduce the physical and emotional toll that is taken on caregivers [166].
Some of these technologies focus on assurance, or making sure our friends and loved ones are
safe and healthy at home. AmI techniques for recognizing activities [18, 118, 122], monitoring
diet and exercise [53, 69], and detecting changes or anomalies [37] support this goal.
The next category of health technologies targets the goal of providing support to in-
dividuals with cognitive or physical impairments. AmI techniques can be used to provide
reminders of normal tasks [98] or the sequence of steps that comprise these tasks [108]. Use
of devices such as the activity compass [83] can actually remind individuals of the route that
will get them back to a safe location if they have wandered off. For those with physical
limitations, automation of their home and work environment [178] can allow them to live
independent lives at home.
AmI technologies can also be used to assess the cognitive limitations of individuals.
Carter and Rosen [31] demonstrate such an assessment based on the ability of individuals to
efficiently complete kitchen tasks. Jimison, et al. [81] also provide such an assessment. In
their case, individuals are monitored while playing computer games, and assessment is based
on factors such as game difficulty, player performance, and time to complete the game.
Finally, AmI can be used to enhance the quality of life for individuals who would otherwise
lead solitary lives at home. Intel has created the “Proactive Health Group” which performs
research and development of technologies that can increase the quality of life of older adults
[100]. One important aspect of older adults related to wellbeing is their social network. Intel
has developed systems which using wireless sensors examines this particular aspect of the
daily life of a person. Intel’s technologies provide information to caregivers summarizing the
social interactions the individual has had at home and offers advice on how to improve that
aspect of a person’s life.
8.3 Hospitals
While bringing health care to homes is an exciting development, hospitals are still needed
for a variety of reasons. The concentration of costly equipment and specialized professionals
is valuable in many situations. Applications of AmI in hospitals can vary from enhancing
safety for patients and professionals to following the evolution of patients after surgical
intervention. Many of the AmI technologies found in smart homes can be adapted for use
in specific rooms or areas of a hospital.
At a different level, AmI can be used to improve the experience of hospital visitors. For
example, the Lutheran General Hospital in Chicago has built the Yacktman Children’s CT
Pavilion where patients are entertained and helped by Ambient Intelligence during their
examination sessions [106]. Patients can select a topic of preference for their visit and as
they enter to the hospital their identity is read from their RFID-encoded cards. The system
is then aware of their presence at the unit, and also of their preferences, being able to tailor
the lighting and wall/ceiling projections when they are in a particular room. The images
projected can be used to calm the anxiety of the patient but also to guide them. For example,
20
if a child is required to hold his breath during an examination a figure in the projection will
do the same. The child’s fear may be reduced by letting them understand the procedure
they are about to undertake. Children waiting for a scan can use a small scale toy scanner
unit and scan toy animals, which are recognized by RFID sensors, and the toys will trigger
the appropriate images on a display.
Ambient Intelligence can also be used to link hospital care with smart home technology.
As another example, the Ulster Community Hospitals Trust of Northern Ireland [161] has set
up the PathFinder project with the goal of caring for elderly and vulnerable people in their
homes. By eventually equipping 3,000 homes in the community with sensors and monitoring
their well-being, PathFinder can increase the level of autonomy, independence and safety for
these individuals, particularly if they have a medical condition which may be detrimental to
their lifestyle.
Hospitals can increase the efficiency of their services by monitoring patients’ health and
progress through analysis of activities in their rooms. They can also increase safety by, for
example, only allowing authorized personnel and patients to have access to specific areas and
devices. The latest issue of Consumer Reports [35] laments the status of assisted care facilities
in the US and the need in most for additional staffing. Ambient Intelligence capabilities can
be used in this setting to reduce the burden of staff nurses in assisted care facilities, and to
make them aware more quickly of residents’ needs they arise. In addition, tracking is used
to find doctors or nurses in a hospital setting when they are urgently needed. This is the
most common use of many technologies such as active badges [110].
8.4 Transportation
Transport means are also valuable settings for AmI technologies. We spend a significant part
of our life traveling in different ways. Train stations, buses, and cars can be equipped with
technology that can provide fundamental knowledge about how the system is performing at
each moment. Based on this knowledge, preventive actions can be applied and the experience
of people using that transport can be increased by using the system more effectively. Public
transport can benefit from AmI technology including GPS-based spatial location, vehicle
identification and image processing to make transport more fluent and hence more efficient
and safe. As an example we can consider the I-VAITs project [140] aiming to assist drivers
by gathering important information through the way they use different elements of the car
(pressure on breaks) or their movements and image processing of the driver’s face expressions
(as mood indicators). This can allow a system to assist the driver more effectively when is
help most needed, such as while executing tricky maneuvers.
Pentland, in partnership with Nissan Cambridge Basic Research, [127] has built a system
that allows the car to “observe” the driver, continuously estimating the driver’s internal state
and responding appropriately. An HMM model of the driver’s hand and leg motions and
associated actions (e.g., passing, turning, stopping, car following, lane change, or speeding
up) was built. This was used to classify real driver’s actions in relation to the artificial model.
The system was able to accurately identify what action the driver was beginning to execute.
This detection can be done as soon as the action started. with high accuracy (97% within
0.5 seconds of the beginning an action, rising to over 99% accuracy within two seconds).
This quick scenario identification allow a real-time optimization of the car’s performance to
21
suit a particular situation, and to warn the driver about possible dangers.
Microsoft also employs AmI technologies for driver assistance by providing route planners.
They also generate inference about possible preferred routes and provide customized route
suggestions for drivers [86, 94].
8.5 Emergency Services
Safety-related services like fire brigades can improve the reaction to a hazard by locating
the site of an accident more efficiently and also by preparing a route to reach the place
in connection with street services. This can be realized using image processing and traffic
monitoring as found in the e-Road project [40]. This service can also quickly locate a place
where a hazard is occurring or is likely to occur and prepare a better access to it for security
personnel.
Similarly, the PRISMATICA project [164] uses cameras to monitor public transportation
locations. By detecting situations such as overcrowding, the presence of people or objects
that are not moving, motion in a forbidden direction, and intrusion, the environment and
officials can respond quickly to ensure the safety of individuals using public transportation.
8.6 Education
Not only do students learn about technologies such as Ambient Intelligence in the classroom,
but AmI can also help improve the learning experience for these students. Education-related
institutions can use technology to track students’ progression on their tasks and frequency
of their attendance at key events. In the Georgia Tech Classroom 2000 project [6], Abowd
provides human-computer interfaces through devices such as an interactive whiteboard that
stores content in a database. The smart classroom of Shi, et al. [149], also uses an interactive
whiteboard, and allows lecturers to write notes directly on the board with a digital pen. This
classroom experience is further enhanced by video and microphones that recognize a set of
gestures, motions, and speech that can be used to retrieve information or focus attention on
appropriate displays and material.
The intelligent classroom at Northwestern University [55] employs many of these same
devices, and also uses the captured information to infer speaker intent. From the inferred
intent the room can control light settings, play videos, and display slides. In none of these
cases is explicit programming of the Ambient Intelligence system necessary – natural actions
of the inhabitants elicit appropriate responses from the environment.
In their Reconfigurable Context Sensitive Middleware (RCSM) Project at Arizona State
University [174], Yau provides enhanced collaborative learning in smart classroom environ-
ments using PDAs to monitor the situation in terms of the locations of PDAs, noise, light,
and mobility activity. They use the current situation to trigger communication activity
among the students and the instructor for group discussion and automatic distribution of
presentation materials. At San Francisco State University [88] the smart classroom project
focuses not only on supporting user interactions but also visualizing the behaviors.
Targeting early childhood education, a Smart Table was designed as part of the Smart
Kindergarten project at UCLA [156]. By automatically monitoring kids’ interaction with
blocks on a table surface, the Smart Table enables teachers to observe learning progress for
22
children in the class. Children respond particularly well to such natural interfaces, as in the
case of the KidsRoom at MIT [22]. The room immerses children in a fantasy adventure in
which the kids must work together to explore the story. KidsRoom presents children with an
interactive fantasy adventure. Only through teamwork actions such as rowing a virtual boat
and yelling a magic word will the story advance, and these activities are captured through
cameras and microphones placed around the room.
8.7 Workplaces
Facilitating interaction is particularly important in a workplace environment, where workers
want to focus on the project at hand without being tripped up by technology. The AIRE
project [8], for example, has designed intelligent workspaces, conference rooms, and kiosks
that use a variety of mechanisms such as gaze-aware interfaces and multi-modal sketching
that the full meaning of a discussion between co-workers through the integration of captured
speech and captured writing on a whiteboard.
The Monica project [91] identifies gestures and activities in order to retrieve and project
needed information in a workplace environment. Similarly, the Interactive Room (iRoom)
project at Stanford [54] enables easy retrieval and display of useful information. Users can
display URLs on a selected surface by simply dragging the URL onto the appropriate PDA
icon.
NIST Smart Space and Meeting Recognition projects are developing tools for data for-
mats, transport, distributed processing,and metadata which aid context-aware smart meeting
rooms that sense ongoing human activities and respond to them [154]. A cooperative ef-
fort from a network of Portuguese Universities [104] is developing a system which is able to
support Distributed Decision Making Groups through the use of an agent-based architec-
ture. The system is able to exhibit intelligent, and emotional-aware behavior, and supports
argumentation, through interaction with individual persons or groups.
Production-centered places like factories are capable of self-organizing according to the
production / demand ratio of the goods produced. This demands careful correlation between
the collection of data through sensors within the different sections of the production line
and the pool of demands via a diagnostic system which can advice the people in charge of
the system at a decision-making level. Production environments can be also enriched with
AmI technology in order to increase important aspects of the process, such as safety of the
employees. The MOSES system [157] uses AmI to infer where the personnel is located and
what task are performing. The system relies upon RFID technology to recognize positioning
of the important elements of the environment. As workers are equipped with RFID readers,
the system can track the development of activities and therefore can advise the employee
what tasks remain to be done. The iShopFloor [148] provides an architecture for intelligent
manufacturing process planning, scheduling, sensing, and control. The system is based on
three main agents: resource agents (manufacturing devices), product/part agents (parts),
and service agents (coordination of resource and parts agents).
23
9 Final Words
Humans have learned through the millennia how to benefit from their environments. Whether
it was by obtaining food or shelter we learned how different habitats can give us fundamental
elements for our survival or comfort.
In search of security and predictability our modern society began to imbue their surround-
ings with technology in order to more easily obtain essential elements for the functioning of
society and in order to make key elements of survival and comfort available to the masses.
Until recently, this technology has been passive.
We have reached a point where technology allows humanity to make these technologies
more active. The goal of ambient intelligence is not only to provide such active and intelligent
technologies, but to weave them seamlessly into the fabric of everyday lives and settings and
to tailor them to each individual’s specific needs.
9.1 Ongoing Challenges
Systems which have to interact with humans face important challenges. They differ in many
ways from those that validate their usefulness only by accepting a well-delimited input,
computing a solution and displaying the result. AmI systems have to interact sensibly with
the user in a variety of sophisticated ways.
Crucially, AmI systems need to be aware of the users preferences, intentions, and needs.
AmI systems should know when it is convenient to interrupt a user, and when to make a
suggestion but also when is more convenient to refrain from making a suggestion. Sometimes
acting may be essential to save a life or to prevent an accident. Too much intervention from
the system can be inadequate and even can make the system useless if the user get tired of it
and decides not to pay attention anymore. All that social tact that humans learn throughout
life is not simple to achieve.
There are many practical challenges that need to be met in each of the contributing
technological areas we have surveyed. For example, many AmI applications relying upon
wireless sensors are at the mercy of the battery life for the sensors. Researchers are starting
to investigate batteryless approaches to sensing [129], but much work remains to be done
to make this approach robust and easy to use. In the area of user modeling and activity
analysis, an ongoing challenge is to model multiple residents in an environment. While this
has been investigated for location tracking in a limited context [144], solving the general
problem of activity modeling, recognition, and prediction for multiple-resident settings is an
open and very difficult problem.
In this paper we discussed issues related to security and privacy for AmI systems. Some
steps have been taken to better understand privacy issues and to address these in AmI
systems. The dependability of AmI systems has not been researched to the same extent.
An ongoing challenge for AmI researchers is to design self-testing and self-repairing AmI
software that can offer quantitative quality-of-service guarantees.
In addition, the IST Advisory Group has stated a goal that AmI facilitates human contact
[65]. In contrast, current AmI research has actually raised fears of isolationism [24]. A new
direction that can be forged for AmI researchers is to investigate mechanisms for supporting
24
and enriching human socialization and interaction, and orient AmI toward community and
cultural enhancement.
Much study and experimentation is still needed to know what sensors, in which quantity
and in which particular distribution are needed to guarantee an acceptable level of service.
As technology advances and provides new sensors the line will be continually moving.
9.2 Conclusions
Ambient Intelligence is establishing fast as an area where a confluence of topics can converge
to help society through technology. We have summarized the flexibility of the idea, the
current state of the art and current trends at research labs and companies.
There are still many challenges ahead and improvements are needed at all levels: infras-
tructure, algorithms and human-computer interaction for AmI systems to be widely accepted
and more important of all, be useful to society. We are conscious that the realization of
AmI’s aims are not easily reachable but the field is gaining momentum. Many important
elements are advancing and we are optimistic that this will bring the synergy that is needed
to materialize the goal of Ambient Intelligence.
Acknowledgment
These notes greatly benefited with the kind input of many colleagues around the world: Emile
Aarts, Marc B¨ohlen, Vic Callaghan, Boris de Ruyter, Bj¨orn Gottfried, Hans Guesgen, Sylvain
Gyroux, Sumi Helal, Eric Horvitz, Sebastian H¨uebner, Helene Pigot, Maureen Schmitter-
Edgecombe, Hamid Aghajan, Sajal Das, and the anonymous reviewers.
References
[1] AARP. These four walls... Americans 45+ talk about home and community, 2003.
[2] E. Aarts and L. Appelo. Ambient intelligence: thuisomgevingen van de toekomst. IT
Monitor, 9:7–11, 1999.
[3] E. Aarts and J. Encarnacao. True Visions: The Emergence of Ambient Intelligence.
Springer, 2006.
[4] E. Aarts and J. Encarnaao, editors. True Visions: Tales on the Realization of Ambient
Intelligence. Springer, Berlin, 2006.
[5] E. Aarts, R. Harwig, and M. Schuurmans. Ambient intelligence. In The invisible future:
the seamless integration of technology into everyday life, pages 235–250. McGraw-Hill,
2002.
[6] G. D. Abowd. Classroom 2000: An experiment with the instrumentation of a living
educational environment. IBM Systems Journal, 38(4):508–530, 1999.
25
[7] G. D. Abowd and E. D. Mynatt. Designing for the human experience in smart envi-
ronments. In D. J. Cook and S. K. Das, editors, Smart Environments: Technology,
Protocols, and Applications, pages 153–174. Wiley, 2004.
[8] A. Adler and R. Davis. Speech and sketching for multimodal design. In Proceedings of
the 9th International Conference on Intelligent User Interfaces, pages 214–216, 2004.
[9] H. Ailisto, A. Kotila, and E. Strommer. Ubicom applications and technologies, 2007.
www.vtt.fi/inf/pdf/tiedotteet/2003/T2201.pdf.
[10] I. F. Akyldiz, W. Su, Y. Sankarasubramanian, and E. Cayirci. A survey on sensor
networks. IEEE Communications Magazine, 40:102–114, 2002.
[11] M. Ali, W. Aref, R. Bose, A. Elmagarmid, A. Helal, I. Kamel, and M. Mokbel. NILE-
PDT: a phenomenon detection and tracking framework for data stream management
systems. In Proceedings of VLDB’05, pages 1295–1298, 2005.
[12] J. F. Allen and G. Ferguson. Actions and events in interval temporal logic. Journal of
Logic and Computation, 4:531–579, 1994.
[13] F. Amigoni, N. Gatti, C. Pinciroli, and M. Roveri. What planner for ambient intelli-
gence applications? IEEE Transactions on Systems, Man, and Cybernetics, Part A:
Systems and Humans, 35(1):7–21, 2005.
[14] G. Anastasi, A. Falchi, A. Passarella, M. Conti, and E. Gregori. Performance mea-
surements of motes sensor networks. In Proceedings of the International Workshop
on Modeling Analysis and Simulation of Wireless and Mobile Systems, pages 174–181,
2004.
[15] G. Asada, I. Bhatti, T. H. Lin, S. Natkunanthanan, F. Newberg, R. Rofougaran,
A. Sipos, S. Valoff, G. J. Pottie, and W. J. Kaiser. Wireless integrated network sensors.
In Proceedings of SPIE, pages 11–18, 1999.
[16] J. Augusto and C. Nugent. The use of temporal reasoning and management of complex
events in smart homes. In R. L. de M´antaras and L. Saitta, editors, Proceedings of
European Conference on Artificial Intelligence (ECAI 2004), pages 778–782. IOS Press
(Amsterdam, The Netherlands), 2004. August, 22-27.
[17] J. C. Augusto and P. McCullah. Ambient intelligence: Concepts and applications.
International Journal on Computer Science and Information Systems, 4(1):1–28, 2007.
[18] T. S. Barger, D. E. Brown, and M. Alwan. Health status monitoring through analysis
of behavioral patterns. IEEE Transactions on Systems, Man, and Cybernetics, Part
A, 35(1):22–27, 2005.
[19] L. Benini and M. Poncino. Ambient intelligence: A computational platform perspec-
tive. In Ambient Intelligence: Impact on embedded system design, pages 31–50. Kluwer,
2003.
26
[20] J. O. Berger. Statistical Decisions. Springer-Verlag, 1985.
[21] M. Berger. Towards ambient intelligence - an industrial view. In J. Augusto and
D. Shapiro, editors, Proceedings of the 1st Workshop on Artificial Intelligence Tech-
niques for Ambient Intelligence (AITAmI’2006), page 8, Riva del Garda, Italy, 2006.
[22] A. F. Bobick, S. S. Intille, J. W. Davis, F. Baird, C. S. P. ez, L. W. Campbell, Y. A.
Ivanov, A. Schuette, and A. Wilson. The kidsroom: A perceptually-based interactive
and immersive story environment. Presence, 8(4):369–393, 1999.
[23] M. Bohlen and N. Tan. Garden variety pervasive computing. IEEE pervasive computing
mobile and ubiquitous systems, 3(1):29–34, 2004.
[24] J. Bohn, V. Coroama, M. Langheinrich, F. Mattern, and M. Rohs. Social, economic and
ethical implications of ambient intelligence and ubiquitous computing. In W. Weber,
J. Rabaey, and E. Aarts, editors, Ambient Intelligence, pages 5–29. Springer, 2005.
[25] B. Bouchard, A. Bouzouane, and S. Giroux. A keyhole plan recognition model for
alzheimer’s patients: First results. Journal of Applied Artificial Intelligence, 22(7):1–
34, 2007.
[26] K. Brammer and G. Siffling. Kalman Bucy Filters. Artech House, 1989.
[27] O. Brdiczka, J. Maisonnasse, and P. Reignier. Automatic detection of interaction
groups. In Proceedings of the International Conference on Multimodal Interfaces, pages
28–34, 2005.
[28] P. Brey. Freedom and privacy in ambient intelligence. Ethics and Information Tech-
nology, 7(3):157–166, 2005.
[29] C. Brezeal and B. Scassellati. A context-dependent attention system for a social robot.
In Proceedings of the International Joint Conference on Artificial Intelligence, pages
1146–1151, 1999.
[30] R. Cantoni. Bodyarchitecture: The evolution of interface towards ambient intelligence.
In G. Riva, F. Vatalaro, F. Davide, and M. Alcaniz, editors, Ambient Intelligence. IOS
Press, 2005.
[31] J. Carter and M. Rosen. Unobtrusive sensing of activities of daily living: A preliminary
report. In Proceedings of the 1st Joint BMES/EMBS Conference, page 678, 1999.
[32] K. Cheverst, N. Davies, K. Mitchell, A. Friday, and C. Efstratiou. Developing a
context-aware electronic tourist guide: Some issues and experiences. In Proceedings of
the Conference on Human Factors in Computing Systems, pages 17–24, 2000.
[33] Y. Cho, S. Cho, D. Choi, S. Jin, K. Chung, and C. Park. A location privacy protection
mechanism for smart space. In Information Security Applications, pages 162–173.
Springer, 2004.
27
[34] Commission of the European Communities. Ageing well in the information society:
An i2010 initiative, 2007.
[35] Consumer Reports. Profits vs. patients: CR investigates nursing homes, August 2007.
[36] D. J. Cook and S. K. Das. How smart are our environments? an updated look at the
state of the art. Journal of Pervasive and Mobile Computing (to appear), 2007.
[37] D. J. Cook, G. M. Youngblood, and G. Jain. Algorithms for smart spaces. In Technology
for Aging, Disability and Independence: Computer and Engineering for Design and
Applications. Wiley, 2006.
[38] C. K. M. Crutzen. Invisibility and the meaning of ambient intelligence. International
Review of Information Ethics, 6:1–11, 2006.
[39] S. K. Das, D. J. Cook, A. Bhattacharya, E. O. Heierman, and T.-Y. Lin. The role
of prediction algorithms in the mavhome smart home architecture. IEEE Wireless
Communications, 9(6):77–84, 2002.
[40] S. Dashtinezhad, T. Nadeem, B. Dorohonceanu, C. Borcea, P. Kang, and L. Iftode.
Trafficview: a driver assistant device for traffic monitoring based on car-to-car com-
munication. In Proceedings of the Vehicular Technology Conference, pages 2946–2950,
2004.
[41] B. de Ruyter and E. Aarts. Ambient intelligence: Visualizing the future. In Proceedings
of the Conference on Smart Objects and Ambient Intelligence, pages 203–208, 2005.
[42] G. Delapierre, H. Grange, B. Chambaz, and L. Destannes. Polymer-based capacitive
humidity sensor. Sensors and Actuators, 4(1):97–104, 1983.
[43] J. Delsing and P. Lindgren. Sensor communication technology towards ambient intel-
ligence. Measurement Science and Technology, 16:37–46, 2005.
[44] B. deRuyter. Social interactions in ambient intelligent environments. In J. Augusto
and D. Shapiro, editors, Proceedings of the 1st Workshop on Artificial Intelligence
Techniques for Ambient Intelligence (AITAmI’2006), pages 9–10, Riva del Garda, Italy,
2006.
[45] A. K. Dey, J. Mankoff, G. D. Abowd, and S. Carter. Distributed mediation of ambigu-
ous context in aware environments. In Proceedings of the Annual Symposium on User
Interface Software and Technology, pages 121–130, 2002.
[46] A. Dix, J. Finlay, G. D. Abowd, and R. Beale. Human–Computer Interaction 3d
edition. Prentice Hall, 2003.
[47] J.-H. Do, S. H. Jung, H. Jang, S.-E. Yang, J.-W. Jung, and Z. Bien. Gesture-based
interface for home appliance control in smart home. In C. Nugent and J. C. Augusto,
editors, Smart Homes and Beyond - Proceedings of the 4th International Conference On
Smart homes and health Telematics (ICOST2006), volume 19 of Assistive Technology
Research Series, pages 23 – 30, Belfast, UK, June 2006. IOS Press.
28
[48] F. Doctor, H. Hagras, and V. Callaghan. A fuzzy embedded agent-based approach for
realizing ambient intelligence in intelligent inhabited environments. IEEE Transactions
on Systems, Man, and Cybernetics, Part A, 35(1):55–65, 2005.
[49] H. F. Durrant-Whyte. Uncertain geometry in robotics. IEEE Transactions on Robotics
and Automation, 4(1):23–31, 1988.
[50] K. Eng, R. J. Douglas, and P. F. M. J. Verschure. An interactive space that learns
to influence human behavior. IEEE Transactions on Systems, Man, and Cybernetics,
Part A: Systems and Humans, 35(1):66–77, 2005.
[51] R. L. Ernst and J. W. Hay. The US economic and social costs of alzheimer’s disease
revisited. American Journal of Public Health, 84:1261–1264, 1994.
[52] U. Esnaola and T. Smithers. Whistling to machines. In Y. Cai and J. Abascal, editors,
Ambient Intelligence in Everyday Life, pages 198–226. Springer, 2006.
[53] J. Farringdon and S. Nashold. Continuous body monitoring. In Y. Cai, editor, Ambient
Intelligence for Scientific Discovery. Lecture Notes in Computer Science 3345, pages
202–223. Springer Verlag, 2005.
[54] A. Fox, B. Johanson, P. Hanrahan, and T. Winograd. Integrating information appli-
ances into an interactive space. IEEE Computer Graphics and Applications, 20(3):54–
65, 2000.
[55] D. Franklin. Cooperating with people: The intelligent classroom. In Proceedings of
the National Conference on Artificial Intelligence, pages 555–560, 1998.
[56] A. Galton. Qualitative Spatial Change. Oxford University Press, 2000.
[57] A. Galton. Eventualities. In Fisher, Gabbay, and Vila, editors, Handbook of Temporal
Reasoning in Artificial Intelligence, pages 25–28. Elsevier, 2005.
[58] M. Gaynor, S. L. Moulton, M. Welsh, E. LaCombe, A. Rowan, and J. Wynne. Inte-
grating wireless sensor networks with the grid. IEEE Internet Computing, 8(4):32–39,
2004.
[59] S. Giroux and H. Pigot, editors. From Smart Homes to Smart Care (Proceedings of the
3rd International Conference on Smart Homes and Health Telematic - ICOST2005),
volume 15 of Assistive Technology Research, Sherbrooke, Canada, July 2005. IOS Press.
[60] K. Gopalratnam and D. J. Cook. Online sequential prediction via incremental parsing:
The active lezi algorithm. IEEE Intelligent Systems, 22(2), 2007.
[61] B. Gottfried, H. Guesgen, and S. H¨uebner. Spatiotemporal reasoning for smart homes.
In J. C. Augusto and C. D. Nugent, editors, Designing Smart Homes - The Role of
Artificial Intelligence, volume 4008 of LNCS, pages 16–34. Springer, Heidelberg, 2006.
29
[62] P. J. Grayons. Technology and home adaptation. In S. lanspery and J. Hyde, ed-
itors, Staying put: Adapting the places instead of the people, pages 55–74. Baywood
Publishing Company, 1997.
[63] A. Greenfield. Everyware: The Dawning Age of Ubiquitous Computing. Peachpit Press,
2006.
[64] J. Gross. A grass-roots effort to grow old at home, August 14, 2007.
[65] I. A. Group. Scenarios for ambient intelligence in 2010, 2001.
[66] H. Hagras, V. Callaghan, M. Colley, G. Clarke, A. Pounds-Cornish, and H. Duman.
Creating an ambient-intelligence environment using embedded agents. IEEE Intelligent
Systems, 19(6):12–20, 2004.
[67] H. Hagras, F. Doctor, A. Lopez, and V. Callaghan. An incremental adaptive life long
learning approach for type-2 fuzzy embedded agents in ambient intelligent environ-
ments. IEEE Transactions on Fuzzy Systems, 2007.
[68] J. Halls. Ultra low resolution thermal imaging for kitchen hazard detection. In C. Nu-
gent and J. C. Augusto, editors, Smart Homes and Beyond - Proceedings of the 4th
International Conference On Smart homes and health Telematics (ICOST2006), vol-
ume 19 of Assistive Technology Research Series, pages 231–238, Belfast, UK, June
2006. IOS Press.
[69] K. hao Chang, S. yen Liu, H. hua Chu, J. Y. jen Hsu, C. Chen, T. yun Lin, C. yu Chen,
and P. Huang. The diet-aware dining table: Observing dietary behaviors over a table-
top surface. In Proceedings of the International Conference on Pervasive Computing.
Lecture Notes in Computer Science 3968, pages 366–382. Springer Verlag, 2006.
[70] H. R. Hashemipour, S. Roy, and A. J. Laub. Decentralized structures for parallel
Kalman filters. IEEE Transactions on Automatic Control, 33:88–93, 1988.
[71] M. Hazas and A. Hopper. Broadband ultrasonic location systems for improved indoor
positioning. IEEE Transactions on Mobile Computing, 5(5):536–547, 2006.
[72] A. Helal, W. Mann, H. El-Zabadani, J. King, Y. Kaddoura, and E. Jansen. The gator
tech smart house: A programmable pervasive space. IEEE Computer, 38(3):50–60,
2005.
[73] S. Helal, B. Winkler, C. Lee, Y. Kaddourah, L. Ran, C. Giraldo, and W. Mann. En-
abling location-aware pervasive computing applications for the elderly. In Proceedings
of the First IEEE Pervasive Computing Conference, 2003.
[74] B. K. Hensel, G. Demiris, and K. L. Courtney. Defining obtrusiveness in home tele-
health technologies: A conceptual framework. Journal of the American Medical Infor-
matics Association, 13(4):428–431, 2006.
30
[75] L. E. Herbert, P. A. Scherr, J. L. Bienias, D. A. Bennett, and D. A. Evans. Alzheimer’s
disease in the US population: Prevalance estimates using the 2000 census. Archives of
Neurology, 60:1119–1122, 2000.
[76] A. Hopper. Sentient computing, 1999.
[77] E. Horvitz, P. Koch, and J. Apacible. Busybody: Creating and fielding personalized
models of the cost of interruption. In Proceedings of CSCW, Conference on Computer
Supported Cooperative Work, pages 507 – 510. ACM Press, 2004.
[78] Intel Proactive Health, 2006. www.intel.com/research/prohealth/cs-
aging in place.htm.
[79] V. Jakkula, A. Crandall, and D. J. Cook. Knowledge discovery in entity based smart
environment resident data using temporal relations based data mining. In Proceedings
of the ICDM Workshop on Spatial and Spatio-Temporal Data Mining, 2007.
[80] D. N. Jayasimha, S. S. Iyengar, and R. L. Kashyap. Information integration and
synchronization in distributed sensor networks. IEEE Transactions on Systems, Man,
and Cybernetics, 21(5):1032–1043, 1991.
[81] H. B. Jimison, M. Pavel, and J. Pavel. Adaptive interfaces for home health. In
Proceedings of the International Workshop on Ubiquitous Computing for Pervasive
Healthcare, 2003.
[82] A. N. Joinson, C. Paine, U.-D. Reips, and T. Buchanan. Privacy and trust: The role
of situational and dispositional variables in online disclosure. In Workshop on Privacy,
Trust and Identity Issues for Ambient Intelligence, 2006.
[83] H. Kautz, L. Arnstein, G. Borriello, O. Etzioni, and D. Fox. An overview of the assisted
cognition project. In Proceedings of the AAAI Worskhop on Automation as Caregiver:
The Role of Intelligent Technology in Elder Care, pages 60–65, 2002.
[84] A. Keshavarz, A. M. Tabar, and H. Aghajan. Distributed vision-based reasoning for
smart home care. In Proc. of ACM SenSys Workshop on DSC, October 2006.
[85] P. Krill. IBM research envisions pervasive computing. InfoWorld, 2000.
[86] J. Krumm and E. Horvitz. Predestination: Inferring destinations from partial trajec-
tories. In UbiComp 2006: Eighth International Conference on Ubiquitous Computing,
pages 243–260, 2006.
[87] M. Laibowitz, J. Gips, R. Aylward, A. Pentland, and J. A. Paradiso. A sensor network
for social dynamics. In Proceedings of the International Conference on Information
Processing in Sensor Networks, pages 483–491, 2006.
[88] E. Lank, A. Ichnowski, and S. Khatri. Zero knowledge access to a smart classroom
environment. In Proceedings of the Workshop on Ubiquitous Display Environments,
2004.
31
[89] S. Lanspery, J. J. Callahan, Jr., J. R. Miller, and J. Hyde. Introduction: Staying put.
In S. Lanspery and J. Hyde, editors, Staying Put: Adapting the Places Instead of the
People, pages 1–22. Baywood Publishing Company, 1997.
[90] C. Larson. In elder care, signing on becomes a way to drop by, February 4, 2007.
[91] C. Le Gal. Smart offices. In D. J. Cook and S. K. Das, editors, Smart Environments:
Technology, Protocols, and Applications. Wiley, 2004.
[92] H. Lei. Context awareness: a practitioners perspective. In Proceedings of the Interna-
tional Workshop on Ubiquitous Data Management, pages 43–52, 2005.
[93] V. Lesser, M. Atighetchi, B. Benyo, B. Horling, A. Raja, R. Vincent, T. Wagner,
X. Ping, and S. X. Zhang. The intelligent home testbed. In Proceedings of the Auton-
omy Control Software Workshop, 1999.
[94] J. Letchner, J. Krumm, and E. Horvitz. Trip router with individualized preferences:
Incorporating personalization into route planning. In Eighteenth Conference on Inno-
vative Applications of Artificial Intelligence, 2006.
[95] I. L. Liao, D. Fox, and H. Kautz. Location-based activity recognition using relational
Markov networks. In Proceedings of the International Joint Conference on Artificial
Intelligence, pages 773–778, 2005.
[96] J. Liu, J. C. Augusto, H. Wang, and J. B. Yang. Considerations on uncertain spatio-
temporal reasoning in smart home systems. In Proceedings of the International Con-
ference on Applied Artificial Intelligence, 2006.
[97] S. W. Loke. Representing and reasoning with situations for context-aware pervasive
computing: a logic programming perspective. The Knowledge Engineering Review,
19(3):213–233, 2005.
[98] E. F. LoPresti, A. Mihailidis, and N. Kirsch. Assistive technology for cognitive reha-
bilitation: State of the art. Neuropsychological Rehabilitation, 14(1/2):539, 2004.
[99] S. Luhr. Recognition of emergent human behaviour in a smart home: A data mining
approach. Journal of Pervasive and Mobile Computing, special issue on Design and
Use of Smart Environments (to appear), 2007.
[100] J. Lundell. Ubiquitous computing to support older adults and informal caregivers. In
S. Giroux and H. Pigot, editors, From Smart Homes to Smart Care (Proceedings of the
3rd International Conference on Smart Homes and Health Telematic - ICOST2005),
volume 15 of Assistive Technology Research, pages 11–22, Sherbrooke, Canada, July
2005. IOS Press.
[101] S. Madden and M. J. Franklin. Fjording the stream: An architecture for queries
over streaming sensor data. In Proceedings of the International Conference on Data
Engineering, pages 555–566, 2002.
32
[102] E. Maeda and Y. Minami. Steps toward ambient intelligence. NIT Technical Review,
4(1), 2006.
[103] J. Manyika and H. Durrant-Whyte. Data Fusion and Sensor Management: A Decen-
tralized Information-Theoretic Approach. Ellis Horwood, 1994.
[104] G. Marreiros, R. Santos, C. Ramos, J. Neves, P. Novais, J. Machado, and J. Bulas-
Cruz. Ambient intelligence in emotion based ubiquitous decision making. In J. C.
Augusto and D. Shapiro, editors, Proceedings of the Second Workshop on Artificial
Intelligence Techniques for Ambient Intelligence, pages 86–91, 2007.
[105] K. Martinez, J. K. Hart, and R. Ong. Environmental sensor networks. Computer,
37(8):50–56, 2004.
[106] S. Marzano. People as a source of breakthrough innovation. Design Management
Review, 16(2), 2005.
[107] E. Mendelowitz and J. Burke. Kolo and nebesko: A distributed media control frame-
work for the arts. In Proceedings of the International Conference on Distributed Frame
works for Multimedia Applications, pages 113–120, 2005.
[108] A. Mihailidis, J. C. Barbenel, and G. Fernie. The efficacy of an intelligent cogni-
tive orthosis to facilitate handwashing by persons with moderate-to-severe dementia.
Neuropsychological Rehabilitation, 14(1/2):135–171, 2004.
[109] MIT. House n Living Laboratory Introduction, 2006. architecture.mit.edu/
house n/publications.
[110] S. Mitchell, M. D. Spiteri, J. Bates, and G. Coulouris. Context-aware multimedia
computing in the intelligent hospital. In Proceedings of the ACM SIGOPS European
Workshop, 2000.
[111] M. Mokhtari, editor. Independent Living for Persons with disabilities and elderly people
(Proceedings of the 1st International Conference on Smart Homes and Health Telematic
- ICOST2003), volume 12 of Assistive Technology Research, Paris, France, 2003. IOS
Press.
[112] S. Moncrieff. Multi-modal emotive computing in a smart house environment. Jour-
nal of Pervasive and Mobile Computing, special issue on Design and Use of Smart
Environments (to appear), 2007.
[113] M. C. Mozer. Lessons from an adaptive home. In D. J. Cook and S. K. Das, editors,
Smart Environments: Technology, Protocols, and Applications, pages 273–298. Wiley,
2004.
[114] M. Muehlenbrock, O. Brdiczka, D. Snowdon, and J. Meunier. Learning to detect user
activity and availability from a variety of sensor data. In Proceedings of the IEEE
International Conference on Pervasive Computing and Communications, 2004.
33
[115] E. Munguia Tapia, S. Intille, and K. Larson. Real-time recognition of physical activ-
ities and their intensities using wireless accelerometers and a heart rate monitor. In
Proceedings of the International Conference on Wearable Computers, 2007.
[116] B. Najafi, K. Aminian, A. Paraschiv-Ionescu, F. Loew, C. Bula, and P. Robert. Am-
bulatory system for human motion analysis using a kinematic sensor: Monitoring of
daily physical activity in the elderly. IEEE Transactions on Biomedical Engineering,
50(6):711–723, 2003.
[117] R. Nakatsu. Integration of multimedia and art for new human-computer commu-
nications. In Proceedings of the Pacific Rim International Conference on Artificial
Intelligence, pages 19–28, 2002.
[118] M. Nambu, K. Nakajima, M. Noshira, and T. Tamura. An algorithm for the automatic
detection of health conditions. IEEE Enginering Medicine Biology Magazine, 24(4):38–
42, 2005.
[119] C. D. Nugent and J. C. Augusto, editors. Proceedings of the 4th International Con-
ference on Smart Homes and Health Telematic (ICOST2006), volume 19 of Assistive
Technology Research, Belfast, UK, June 2006. IOS Press.
[120] Oatfield Estates, 2006. www.elite-care.com/oatfield.html.
[121] P. L. O’Connell. Korea’s high-tech utopia, where everything is observed, October 5,
2005.
[122] M. Ogawa, R. Suzuki, S. Otake, T. Izutsu, T. Iwaya, and T. Togawa. Long term
remote behavioral monitoring of elderly by using sensors installed in ordering houses.
In Proceedings IEEE-EMBS special topic conference on microtechnologies in medicine
and biology, pages 322–335, 2002.
[123] R. J. Orr and G. D. Abowd. The smart floor: A mechanism for natural user identi-
fication and tracking. In Proceedings of the ACM Conference on Human Factors in
Computing Systems, The Hague, Netherlands, 2000.
[124] Oxford. Oxford Dictionary and Thesaurus. Oxford University Press, 2007.
[125] M. Pantic. Face for ambient intelligence. In Y. Cai and J. Abascal, editors, Ambient
Intelligence in Everyday Life, pages 32–66. Springer, 2006.
[126] A. Pentland. Perceptual environments. In D. J. Cook and S. K. Das, editors, Smart
Environments: Technology, Protocols, and Applications. Wiley, 2004.
[127] A. Pentland. Perceptual environments. In D. Cook and S. Das, editors, Smart Envi-
ronments: Technologies, Protocols and Applications. Wiley, 2005.
[128] M. Philipose, K. P. Fishkin, M. Perkowitz, D. J. Patterson, D. Hahnel, D. Fox, and
H. Kautz. Inferring ADLs from interactions with objects. IEEE Pervasive Computing,
3(4):50–57, 2005.
34
[129] M. Philipose, J. R. Smith, J. Bing, A. Mamishev, S. Roy, and K. Sundara-Rajan.
Battery-free wireless identification and sensing. IEEE Pervasive Computing, 4(1):37–
45, 2005.
[130] Philips. 365 days ambient intelligence research in homelab, 2003.
www.research.philips.com/.
[131] Philips, 2006. www.research.philips.com/technologies/
syst softw/ami/background.html.
[132] Phillips Research. Ambient intelligence: Changing lives for the better, 2007.
www.research.phillips.com/.
[133] Phillips Research. Other perspectives on ambient intelligence, 2007.
[134] H. Pigot, A. Mayers, S. Giroux, B. Lefebvre, and N. N. V. Rialle. Smart house for
frail and cognitive impaired elders. In First International Workshop on Ubiquitous
Computing for Cognitive Aids (UbiCog’02), Goteborg, Sweden, September 2002.
[135] J. Pineau, M. Montemerlo, M. Pollack, N. Roy, and S. Thrun. Towards robotic assis-
tants in nursing homes: Challenges and results. Robotics and Autonomous Systems,
42(3-4), 2003.
[136] M. E. Pollack. Intelligent technology for an aging population: The use of AI to assist
elders with cognitive impairment. AI Magazine, 26(2):9–24, 2005.
[137] G. Pottie and W. Kaiser. Wireless sensor networks. Communications of the ACM,
43(5):51–58, 2000.
[138] D. Preuveneers, L. Demuynck, B. Elen, K. Verslype, B. D. Decker, Y. Berbers, and
P. Verbaeten. Context-driven prevention of unintended identity disclosure. In Work-
shop on Privacy, Trust and Identity Issues for Ambient Intelligence, 2006.
[139] Privacy Rights Clearinghouse. RFID position statement of consumer privacy and civil
liberties organizations, 2003.
[140] A. Rakotonirainy and R. Tay. In-vehicle ambient intelligent transport systems (i-vaits):
Towards an integrated research. In Procedings of 7th international IEEE conference
on intelligent transportation systems (ITSC 2004), pages 648–651, Washington DC,
USA, 2004.
[141] J. Rech and K.-D. Althoff. Artificial intelligence and software engineering: Status and
future trends. Themenschwerpunkt KI & SE, KI, 3:5–11, 2004.
[142] J. Rekimoto and M. Saitoh. Augmented surfaces: A spatially continuous work space
for hybrid computing environments. In Proceedings of the ACM SIGCHI Conference
on Human Factors in Computing Systems, pages 378–385, 1999.
35
[143] N. Roy, A. Roy, , and S. K. Das. Context-aware resource management in multi-
inhabitant smart homes: A framework based on nash h-learning. Journal of Pervasive
and Mobile Computing, 2(4):372–404, 2006.
[144] N. Roy, A. Roy, , and S. K. Das. Context-aware resource management in multi-
inhabitant smart homes: A nash h-learning based approach. Proceedings of the IEEE
Conference on Pervasive Computing and Communications, pages 148–158, 2006.
[145] S. J. Russell and P. Norvig. Artificial Intelligence: A Modern Approach (Second Edi-
tion). Prentice Hall, 2003.
[146] M. Saito. Expanding welfare concept and assistive technology. In Proceedings of the
IEEK Annual Fall Conference, pages 156–161, 2000.
[147] N. Shadbolt. Ambient intelligence. IEEE Intelligent Systems, 18(4):2–3, 2003.
[148] W. Shen, S. Lang, and L. Wang. ishopfloor: An internet-enabled agent-based intelligent
shop floor. IEEE Trans. on Systems, Man, and Cybernetics, Part C, 35(3):371–381,
2005.
[149] Y. Shi, W. Xie, G. Xu, R. Shi, E. i Chen, Y. Mao, and F. Liu. The smart classroom:
Merging technologies for seamless tele-education. IEEE Pervasive Computing, April-
June:47–55, 2003.
[150] Siemens, 2006. networks.siemens.de/smarthome/en/index.htm.
[151] R. Simpson, D. Schreckenghost, E. F. LoPresti, and N. Kirsch. Plans and planning
in smart homes. In J. Augusto and C. Nugent, editors, Designing Smart Homes: The
role of Artificial Intelligence. Springer Verlag, 2006.
[152] G. Smith, S. Della Sala, R. H. Logie, and E. A. Maylor. Prospective and retrospective
memory in normal aging and ementia: A questionnaire study. Memory, 8:311–321,
2000.
[153] V. Stanford. Biosignals offer potential for direct interfaces and health monitoring.
IEEE Pervasive Computing, 3:99–103, 2004.
[154] V. Stanford. Infrastructure for distributed and embedded systems and interfaces. In
Proceedings of the Embedded Systems, Ambient Intelligence and Smart Surroundings
Conference, 2005.
[155] D. H. Stefanov, Z. Bien, and W.-C. Bang. The smart house for older persons and per-
sons with physical disabilities: Structure, technology arrangements, and perspectives.
IEEE Transactions on Neural Systems and Rehabilitation Engineering, 12(2), 2004.
[156] P. Steurer and M. B. Srivastava. System design of smart table. In Proceedings of the
IEEE International Conference on Pervasive Computing and Communications, page
473, 2003.
36
[157] M. Stoettinger. Context-awareness in industrial environments, 2004. www.mobile-
safety.com.
[158] N. Streitz and P. Nixon. Special issue on ’the disappearing computer’. In Communi-
cations of the ACM, V 48, N 3, pages 32–35. ACM Press, March 2005.
[159] E. M. Tapia, S. S. Intille, and K. Larson. Activity recognition in the home using simple
and ubiquitous sensors. In Proceedings of Pervasive, pages 158–175, 2004.
[160] S. Turkle. Whither psychoanalysis in a computer culture?, 2002. futureposi-
tive.synearth.net/2002/11/13.
[161] Ulster Community and Hospitals Trust. Interim poicy on research governance process,
2005.
[162] E. J. van Loenen. The ambience project, 2007. www.hitech-
projects.com/euprojects/ambience/.
[163] A. Vasilakos and W. Pedrycz. Ambient Intelligence, Wireless Networking, and Ubiq-
uitous Computing. Artech House Publishers, 2006.
[164] S. A. Velastin, B. A. Boghossian, B. P. L. Lo, J. Sun, and M. A. Vicencio-Silva. PRIS-
MATICA: toward ambient intelligence in public transport environments. IEEE Trans-
actions on Systems, Man, and Cybernetics, Part A: Systems and Humans, 35(1):164–
182, 2005.
[165] E. Vildjiounaite, S.-M. akel¨a, M. Lindholm, R. Riihim¨aki, V. Kyll¨onen,
J. M¨antyj¨arvi, and H. Ailisto. Unobtrusive multimodal biometrics for ensuring privacy
and information security with personal devices. In Proceedings of the International
Conference on Pervasive Computing, 2006.
[166] P. Vitaliano, D. Echeverria, J. Yi, P. Phillips, H. Young, and I. Siegler. Psychophysio-
logical mediators of caregiver stress and differential cognitive decline. Psychology and
Aging, 20(3):402–411, 2005.
[167] A. M. R. Ward. Sensor-driven computing. PhD thesis, Cambridge University, 1998.
[168] M. Weiser. The computer for the twenty-first century. Scientific American, 165:94–104,
1991.
[169] M. Weiser. Hot topics: Ubiquitous computing. IEEE Computer, 26(10):71–72, 1993.
[170] R. F. Wolffenbuttel, K. M. Mahmoud, and P. L. Regtien. Compliant capacitive wrist
sensor for use in industrial robots. IEEE Transactions on Instrumentation and Mea-
surements, 39:991–997, 1990.
[171] C. Wren and E. Munguia Tapia. Toward scalable activity recognition for sensor net-
works. In Proceedings of the Workshop on Location and Context-Awareness, 2006.
37
[172] D. Wright. The dark side of ambient intelligence. The Journal of Policy, Regulation
and Strategy for Telecommunications, 7(6):33–51, 2005.
[173] R. Xu, G. Mei, Z. Ren, C. Kwan, J. Aube, C. Rochet, and V. Stanford. Speaker
identification and speech recognition using phased arrays. In Y. Cai and J. Abascal,
editors, Ambient Intelligence in Everyday Life, pages 227–238. Springer, 2006.
[174] S. S. Yau, S. Gupta, F. Karim, S. Ahamed, Y. Wang, and B. Wang. A smart class-
room for enhancing collaborative learning using pervasive computing technology. In
Proceedings of the WFEO World Congress on Engineering Education, 2003.
[175] S. Yoshihama, P. Chou, and D. Wong. Managing behavior of intelligent environments.
In Proceedings of the IEEE International Conference on Pervasive Computing and
Communications, pages 330–337, 2003.
[176] G. M. Youngblood. Automating inhabitant interactions in home and workplace en-
vironments through data-driven generation of hierarchical partially-observable Markov
decision processes. PhD thesis, The University of Texas at Arlington, 2005.
[177] G. M. Youngblood and D. J. Cook. Data mining for hierarchical model creation. IEEE
Transactions on Systems, Man, and Cybernetics, Part C, 2007.
[178] G. M. Youngblood, L. B. Holder, and D. J. Cook. A learning architecture for au-
tomating the intelligent environment. In Proceedings of the Conference on Innovative
Applications of Artificial Intelligence, pages 1576–1583, 2005.
[179] G. M. Youngblood, L. B. Holder, and D. J. Cook. Managing adaptive versatile envi-
ronments. Journal of Pervasive and Mobile Computing, 1(4):373–403, 2005.
[180] G. M. Youngblood, L. B. Holder, and D. J. Cook. Managing adaptive versatile en-
vironments. In Proceedings of the International Conference on Pervasive Computing,
pages 351–360, 2005.
[181] M. Youngblood, D. J. Cook, and L. B. Holder. Managing adaptive versatile environ-
ments. In Proceedings of the IEEE International Conference on Pervasive Computing
and Communications, pages 351–360, 2005.
[182] E. Zelkha. The future of information appliances and consumer devices. Palo Alto
Ventures, Palo Alto, California, 1998.
[183] D. Zhang and M. Mokhtari, editors. Toward a Human Friendly Assistive Environment
(Proceedings of the 2nd International Conference on Smart Homes and Health Telem-
atic - ICOST2004), volume 14 of Assistive Technology Research, Singapore, September
2004. IOS Press.
[184] F. Zhu, M. W. Mutka, and L. M. Ni. The master key: A private authentication
approach for pervasive computing environments. In Proceedings of the International
Conference on Pervasive Computing, 2006.
38
... The domestic environment is then increasingly populated by products that can adapt autonomously to the context and the needs of users, entertain a dialogue, recognize them, and track/anticipate their behavior. We are in front of game-changing products, starting to materialize the idea of Ambient Intelligence (AmI) as an "electronic butler" [2], but still responding to the taste for novelty rather than having a significant utility for the user and its living context. Despites their currently unfulfilled promises, this kind of products may actually augment people's control over more responsive and customized environments, enabling the societal acceptance of AmI by making it easy to live with [3]. ...
... Going towards products that are increasingly responsive, natural behaving, and context aware, it is inevitable that they call for a new definition: their methods of evaluation need to be scaled from a single item to its spatial surroundings. As a matter of fact, we can already observe in several responses such a paradigmatic shift from the traditional conception of those dimensions towards some of the essential concepts of AmI [2] (e.g.: empathy, transparency, responsiveness, adaptability, unobtrusiveness).This may suggest a reframing to make them blend with the ones proposed by researchestrustworthiness, conversational, intelligence, meaningfulness -in order to intercept the emerging tendencies of the technological systems embedding AmI principles. ...
Chapter
Artificial Intelligence (AI) is entering our daily life and personal environments, as AI-infused products such as smart speakers are entering millions of houses worldwide. Despite the success and their potential to unleash Ambient Intelligence, they are still considered useless or close to a gadget dimension. Accordingly, we consider it mandatory for the design discipline to acknowledge those products, trying to understand and frame the User Experience (UX) they entail. The present study explores the UX dimensions that could describe and assess the experience enabled by such devices. It is the first step of a research program to create a novel UX evaluation method for AI-infused products. In the study, we employed a multi-method approach to (i) understand the main UX dimensions commonly assessed, (ii) identify specific UX dimensions for AI-infused products, and (iii) verify the assumptions with a sample of advanced users. The results are described, discussed, and framed within the current literature.KeywordsAI-infused productsEvaluating dimensions for AmIUX assessment
... Amidst the wide array of challenges posed by this vision of forthcoming reality [10], until now, the central question of research has been the impact of ambient intelligence (AmI) and of profiling techniques on individual autonomy and refined discrimination. Unauthorized and abusive access to the data collected, loss of control [11], dependency, social exclusion, unwanted and unwarranted surveillance, and more in general privacy [12], trust [13] and security concerns [14] [15] have been identified as possible disbenefits of these technologies [16] [17]. ...
Chapter
Full-text available
Companies have been analyzing data from their own customer interactions on a smaller scale for many years. But only recently, they understood the potential treasure trove of non-traditional and less structured data (such as machine-generated data and social media data) that can be mined both for internal marketing purposes and for licensing to third parties. From the business perspective, the protection of this data is needed to secure the significant economic investment that the “new data economy” can require. Otherwise, data holders may lack the incentives to share the data they own and control, because of the risk that non-authorized users may “free ride” on their investment. Granting property rights is often suggested as a solution to overcome the incentive problem. In the case of data, while relying on contract freedom may seem the favourite solution, between those extremes a spectrum of possible “halfway” approaches has been proposed. They range from “quasi-property” or “semi- commons”, with a liability-like regime, to access rights, requiring to license extractions and reuse of data on FRAND terms.
... The extant marketing literature defines the smart environment as the platform where several heterogeneous smart devices are consistently working to allow inhabitants to live with greater comfort (Cook & Das, 2004). However, at the core of the smart environment is not the devices, but the users, referred to as smart consumers (Cook, Augusto, & Jakkula, 2009;Mavrommati & Darzentas, 2006). Chen et al. (2018) define smart consumers as those consumers who voluntarily engage and are competent to participate in experience sharing. ...
Chapter
Intelligent everyday environments are expected to empower their inhabitants, assisting them in carrying out their everyday tasks, but also ensuring their well-being and prosperity. In this regard, the accessibility of an intelligent environment is of utmost importance to ensure that it fulfills user needs, but also that it is usable and useful for everyone, without imposing barriers or excluding individuals with disabilities or older adults. This chapter carries out a review of the state of the art in the field of interaction techniques in intelligent environments, analyzing their accessibility challenges and benefits for different user categories. Furthermore, toward the direction of universally accessible intelligent environments, the issue of multimodal interaction is discussed, summarizing the modalities that can be employed for each user group.KeywordsAccessibilityIntelligent environmentsAmbient intelligenceDesign for allInteraction techniques
Conference Paper
Full-text available
The elderly population in most of the countries has risen from year to year. Number of elderlies that staying alone at home also increase accordingly since the number of children in one family is decreasing. The cost of hiring caregivers is normally expensive and might cause financial burden to a family in the longterm. Fortunately, with advance technology, alternative approaches such as Global Monitoring System, smart clothing as well as embedded health care system in smart phone and smart watch could be the alternative approaches to monitor the health of some elderly that is able to live alone. The most feasible and cost-saving approach for in-house monitoring should be setting up sensors at different locations in a smart home for independent elderly that stay alone. Daily activities recorded through the sensors can be collected and analyzed to detect if there is any anomaly found. In this paper, analysis of staying alone elderly’s daily activities and behaviors are performed. This analysis is important to further developing suitable model that can be used to detect the anomalies in their routine home activity patterns.
Chapter
We conduct a meta-analysis of scientific contributions in assistive technology, systems, and applications designed for people with motor disabilities and/or limited mobility at the intersection of Ambient Intelligence and Mixed Reality environments. Our findings show that most of the scientific contributions have focused on navigation, rehabilitation, and video games; systems outnumber user studies; and the most frequently implemented attributes of Ambient Intelligence environments have been sensitivity, responsiveness, and adaptation. Our survey also shows that work at the intersection of Ambient Intelligence and Mixed Reality for people with motor disabilities has been scarce despite the opportunities enabled by these two research areas in conjunction to increase access to digital information and services.
Chapter
We discuss the concept of digital proprioception for smart devices and smart environments, which we formalize and operationalize in the context of Ambient Intelligence with a dedicated event-driven software architecture. We also propose extended digital proprioception, by means of which devices and environments can access supplementary information about themselves from other sources, beyond their internal sensing capabilities. We use the latter concept to propose extended human proprioception enabled by the conjoint operation of smart devices and environments. Our contributions enable a new way to conceptualize interactions in smart environments by designing user experiences mediated by spatial communication interfaces where physical space integrates interaction.
Chapter
We report insights into the preferences of people with motor impairments to use smart wearables to access applications and services of ambient intelligence environments. We highlight preferences for smartglasses and the delivery of notifications, smartwatches and health applications, smartwatches and control of smart homes, and for smart rings and bracelets for making payments and using public interactive systems. We also report results from a correlation analysis indicating that people with higher disability levels prefer smart earbuds and rings to smartwatches, smartglasses, and smart bracelets. Our findings are useful to inform applications at the intersection of ambient intelligence and wearable computing to increase access to smart environments for users with motor impairments and limited mobility.
Article
The progression to next generation networks is replete with abundant co-existing technologies. To comply with the always best connected paradigm, several vertical handover decision approaches have been proposed in literature, using advanced techniques and tools. This paper discusses the application of soft computing techniques in the vertical handover decision-making process with emphasis on the state-of-the-art techniques. For a comprehensive evaluation, the algorithms are classified into three sets based on the soft computing technique used, namely fuzzy logic, machine learning, and evolutionary algorithms, and representative handover algorithms in each group are discussed. These papers are categorized in a well-defined structure to bring out their contribution, to underline the pretermitted notions, and to bring forth the emerging issues for future research. This paper summarizes the soft computing concepts and reviews its applications in candidate network selection, QoE enhancement, and reducing the unnecessary handovers.
Article
Full-text available
The motivation for working up this report was to provide the parties involved or interested in Ubicom activities with information about opportunities and enablers in the area, and to contribute Tekes in defining research topics for the current and future research programs dealing with Ubicom. The main approach in the report is from the application system viewpoint. The main emphasis of the technical enablers is in short-range wireless communication technologies. The report deals with the evolution and different approaches to the Ubicom concept around the world, possible application areas and scenarios with general requirements concerning them, key technical enablers with an estimation of their applicability and development during the next few years, aspects for the planning and evaluation of the business models of application systems, and recommendations for the future research activities and leading commercial applications. The main suggestions of this report for the Ubicom research are: The research should be interdisciplanary, that is, it should combine human sciences, usability studies, business and economic research as well as legislation and ethical issues with the technology research, which is the basic enabler of the advancements in this area. Furthermore, both the application and the technology viewpoints must be combined. The core technology research should address short-range wireless communication, middleware, reasoning and smart algorithms, usability issues, as well as sensors and actuator technologies.
Article
A vision of future daily life is explored in Ambient Intelligence (AmI). It contains the assumption that intelligent technology should disappear into our environment to bring humans an easy and entertaining life. The mental, physical, methodical invisibility of AmI will have an effect on the relation between design and use activities of both users and designers. Especially the ethics discussions of AmI, privacy, identity and security are moved into the foreground. However in the process of using AmI, it will go beyond these themes. The infiltration of AmI will cause the construction of new meanings of privacy, identity and security because the "visible" acting of people will be preceded, accompanied and followed by the invisible and visible acting of the AmI technology and their producers. A question in this paper is: How is it possible to create critical transformative rooms in which doubting will be possible under the circumstances that autonomous 'intelligent agents' surround humans? Are humans in danger to become just objects of artificial intelligent conversations? Probably the relation between mental, physical, methodical invisibility and visibility of AmI could give answers.
Article
Today, approximately 10 percent of the world's population is over the age of 60; by 2050 this proportion will have more than doubled. Moreover, the greatest rate of increase is amongst the "oldest old," people aged 85 and over. While many older adults remain healthy and productive, overall this segment of the population is subject to physical and cognitive impairment at higher rates than younger people. This article surveys new technologies that incorporate artificial intelligence techniques to support older adults and help them cope with the changes of aging, in particular with cognitive decline.
Article
A research project on "ambient intelligence" recently launched by NTT Communication Science Laboratories aims to envision a new lifestyle made possible by communication science. Research and development of "ambient intelligence" should bridge the boundaries between technological fields and thus cover the entire field of communication science, rather than be limited to specific fields. Besides performing the basic R&D, we are striving to get the concept established and conducting design and publicity activates related to ambient intelligence in a comprehensive and strategic way. This article introduces this new project.