| Contract number: |
IST 004182 |
| Full name: |
Ambient Intelligence for the networked home environment |
| Type of Project: |
IP |
Project participants:
Philips Research - Philips Design - Philips Consumer Electronics
(the Netherlands), Fagor (Spain), France Telecom (France),
Fraunhofer IMS (Germany), Fraunhofer SIT (Germany), Ikerlan (Spain),
INRIA (France), Italdesign Giugiaro (Italy), SingularLogic (Greece),
Microsoft (Germany), Telin (the Netherlands), ICCS (Greece),
Telefónica I+D (Spain), University of Paderborn (Germany), VTT
(Finland)
For more information see Partners.
Budget
Total cost: 24 MEuro
Funding: 13 MEuro
Timeline
Start: September 2004
Duration 42 months
Problem description
Traditionally home automation, consumer electronics (CE), mobile
communications and personal computing (PC) were strictly separated
domains all having their own industrial players, with their own
business plans, standardization efforts and form factors. Examples
of devices in each domain are given in Figure 1 .

Figure 1 Examples of devices in the mobile, PC, CE and home
automation domains
Most industrial players that were strong in one domain have had a
hard time making inroads into another, except for a few “converged”
products such as Smartphone's or Mobile Digital Assistants that
combine aspects of the mobile and PC domains, or media centres that
combine aspects of the PC and CE domains. By introducing the
networked home, also called connected home, that is a home in which
several pieces of equipment are connected using a network, the
traditional separation of domains is no longer valid (see Figure 2
). Mobile devices can be taken with us, therefore extending our home
environment. By extended home environment we mean bringing what is
outside the home inside if we like that, as well as extending our
feeling of being at home when we, ourselves, are on the move. The
networked home offers great potential to improve people’s life, it
will allow simple and seamless access to content throughout the
home, it will allow much easier user interfaces even to program
simple devices, it can use context information to predict and assist
user actions, it can automatically detect and act upon situation
that affect the security or safety of the home users. This networked
home also represents a huge business opportunity, from selling
devices and infrastructure through to the software and services.
Already, the U.S. networking market is $1.4 billion [source:
Cahner’s In-Stat group]. Within Europe, even now, a connected
household spends on average 360 Euro a year on e-home-services [1],
excluding infrastructure, device and software cost.
Merging these traditionally separated domains within the home goes
quickly. In the year 2005 about 25 million households in Europe will
have some local networking [2]. This means that alone in Europe in
2005 the annual turnover for e-home-services will be some 10 billion
Euros a year. The US, clearly ahead in this area, already has 32
million home networks in use today; this is forecast to grow to 125
million home networks in 2006 [ 3 ]. The same source foresees that
these home networks will be entertainment centred. This is proof of
a fast merger between the traditional CE and PC markets. In 2006,
80% of all broadband households will also have home automation
systems installed [4]. This forecast is underlies the assumption
that an extra merger between home automation and home networking
will take place.

Figure 2 Example of home network in which PC, CE, mobile and home
automation domain are intertwined
Research today is typically performed for each consumer domain (home
automation, PC, CE, mobile) independently, whereas the user moves
continually between these domains. Creating consistent and
intelligent support between these domains needs new solutions. One
of the main challenges at this point in time is bringing existing
research activities and domain expertise together to realize the
needed total solutions. The networked home offers a huge potential
for the improvement of our daily lives and should be seen as an
important step towards Ambient Intelligence (AmI). The problem today
is that the huge potential of a networked home system in particular
and Ambient Intelligence in general is not fulfilled, as a
consistent framework for a home system is not in place. Therefore
end-users do not experience the benefits of home networking.
The main problems in home networking today are:
-
Installation and use of the devices, infrastructure and services
is not easy for the general public, together we call this the
usability of the system.
-
There are not enough attractive services to make it worthwhile to
buy and install a complete home system, i.e. the user benefits are
not clear enough.
That usability and attractive services are the main issues for user
acceptances of home networking is not only our opinion, see [ 5 ]
and [ 6 ], which raise the same issues for rapid user acceptance of
in-home networked systems but add a third important parameter namely
the cost of devices, infrastructure and services. It is thought [by
the same authors as above] that cost of the system is for a large
part determined by the amount of users, so rapid market penetration
will automatically elevate this issue.
In our view, to reach the goal of usability three technical aspects
are of major concern:
-
The user interface has to be user-friendly, familiar, easy to use,
robust against contradictory input, and it must respect the privacy
and security of its users.
-
Interoperability is necessary at all levels, despite the fact that
devices are built by different manufacturers and use different
communication standards and different hardware and software
platforms. It is unlikely that all of these are going to be reduced
to just one in the foreseeable future so agreement should be found
in middleware standards that provide functional interoperability at
a SW level. Standardisation of minimal interfaces of middleware
components, basic services and protocols is therefore necessary. So
far it has never been tried to offer a framework that leads to
interoperability over all four, the CE, PC, mobile and home
automation domains.
-
Automatic discovery of devices and services as well as service composability and upgradeability and self-administration are a
necessity for easy installation and use. Having an open middleware,
which is flexible enough to integrate and compose heterogeneous
services (heterogeneity including the base middleware relied upon by
the services), and high-level enough to enable service description,
recognition and composition up to the semantic level can achieve
this.
To make home systems much more attractive for end-users, the
benefits for the end-user of a combined home system must be evident,
so the attractive services should clearly offer a surplus over what
is offered by non-networked systems today. Most of these attractive
services will use knowledge of the world around the device like the
other devices in the system and the user. This is only possible in a
networked system. This gathering and use of this context and user
information leads to the development of new services for the
end-user that greatly enhance the attractiveness of the system for
the end-user. For example: use a display that is in the
neighbourhood to display information instead of on a small PDA
display, use the processing capabilities of the home server to do
speech recognition and communicating the results back to the camera
that has only small processing capabilities in order to combine
user-related and context-related information.
Within the Amigo project these services that specifically use the
knowledge of the world around the device and the user are called:
intelligent user services as they make the system ‘intelligent’ to
the end-user.
The Amigo project will provide solutions for the major problems that
are encountered in the use of home networking today. The project
aims to improve the usability of a home network by developing open,
standardized, interoperable middleware and improve the
attractiveness by developing interoperable intelligent user
services. The project will show the end-user usability and
attractiveness of such a home system by creating and demonstrating
prototype applications improving everyday life, addressing all vital
user aspects: home care and safety, home information and
entertainment, and extension of the home environment by means of
ambience sharing for advanced personal communication.
Technological and scientific objectives
To improve the usability and attractiveness of home networking for
the end-user Amigo’s main objective is to research and develop open,
standardized, interoperable middleware and intelligent user services
for the networked home environment, which offer users intuitive,
personalized and unobtrusive interaction by providing seamless
interoperability of services and applications.
In order to fulfil this main objective we have the following project
objectives:
-
Research and development of open and interoperable middleware for
a networked home system that combines home automation, CE, mobile
and PC functionalities that can be scaled up and down for use on any
hardware and software platform within these domains. The project
will describe the interfaces and protocols necessary to combine this
software with the rest of the terminal software.
-
Research and development of intelligent user services that
combine user interaction, user preferences and context awareness.
These services will keep the user in control and avoid cognitive
saturation of the user, respect user privacy and address security
issues.
-
Guarantee interoperability within this home system between
services even with the use of heterogeneous networking and diverse
devices. The project will contribute and adhere to, preferably open,
standards to enable true interoperability.
-
Guarantee automatic dynamic configuration of the networked home
system by developing high-level methods and concepts for dynamic
integration addressing autonomy, scalability and composability
aspects.
-
Verify the technical solutions with application prototypes and
related demonstrators, and get usability feedback by user
questionnaires, user tests and co-creative user evaluation. The
prototypes will focus on home care and safety, home information and
entertainment and the extended home environment. They will together
form a full demonstration of an Amigo home system in practice.

Figure 3 Example architecture of Amigo device
Figure 3 shows an example of the high-level architecture of an Amigo
device. It is highlighted to show those blocks and interfaces which
will be (further) developed within the Amigo project. The white
components shown are not the focus of the Amigo project and might or
might not be present in an Amigo device.
As well as defining the infrastructure and associated interfaces the
Amigo middleware consists of the following components:
The intelligent user services that Amigo will develop are:
-
Context aggregation and prediction
-
User modelling and profiling
-
Awareness and notification
-
Content selection and
-
User interface.
The above core middleware (components and infrastructure) and
intelligent user services will become available as open source
software together with architectural rules for everyone to use.
In addition, the Amigo project will deliver, prototype applications
to show the clear benefits for the end-user and technical proof of
working. This shall clearly prove the business rationale for
introducing the Amigo networked home system. The Amigo middleware
will be built to overcome diversity in e.g. hardware and software
platforms and networks. The intelligent user services will
incorporated in the Amigo system, though this does not mean that
they will be incorporated or run on all Amigo devices. Some devices
will not offer enough capabilities to do so.
References
-
Datamonitor Study “Digital Home 2003”
-
Gartner Dataquest
-
Networks in the home: analysis and forecasts (third edition),
Park associates
-
Strategis Group
|