Ebook: Advances in Ambient Intelligence

Since the early days of Ambient Intelligence I have been following at close range and with great curiosity the various developments and achievements that have been reported in the open literature on this novel paradigm for the disappearing computer. To obtain a proper feeling for the speed at which the AmI concept has been developing itself, I have been monitoring its Web presence over the past five years by registering on a monthly basis the number of hits Google returns upon the controlled search for the key words “Ambient Intelligence”. After a wavering start with a grand total of only a few hundred hits the number ramped up almost exponentially for a number of consecutive years. Since the beginning of 2006 the growth started to slow down to follow a more or less linear increase leading to current value of about 550,000 hits.

The incredible speedup of the AmI Web presence can be attributed to the publication of the seminal document by the Society and Technology Advisory Group's (ISTAG). It explains the AmI concept on the basis of four AmI scenarios and through this remarkably simple approach the issue attracted an enormous amount of attention. Furthermore, the European Commission follow up on the ISTAG research recommendations by launching the sixth Frame Work Programme in IST using the concept of Ambient Intelligence as its central research theme. This obviously has largely contributed to the enormous growth in the number of studies into the AmI concept.

Evidently, the evolution in the number of Google hits reflects the typical development of a novel paradigm such as Ambient Intelligence. After a conception phase in which the concept is being shaped by a small community of truly convicted early adopters it spreads rapidly to become a leading theme for a large and broad audience of researchers. After the hype the growth slows down again to normal proportions and this is also the phase in which the concept is judged on its true merits. It needs reflection to get rooted; it needs criticism to separate the wheat from the chaff, and it needs reinvention to become robust. It simply needs time to become truly meaningful. To a certain extent this also can be viewed as an existential phase as it reveals whether there is any substance to the concept at all and whether it is truly capable of inspiring researchers to come up with new ideas and glue them together to become a fully integrated disruptive computing paradigm that is meaningful to society providing its participants with maximum benefit.

The Second Workshop on Artificial Intelligence Techniques for Ambient Intelligence (AITAmI'07) has contributed to this purification process as it provided a critical forum for the discussion of the recent developments in the area. The present volume, which resulted from the workshop, does even more so contribute, as the editors have succeeded in compiling a most remarkable collection of chapters. Some chapters are extended version of the best papers presented at the workshop, others are high-quality invited contributions that provide critical reviews of specific developments. The chapters reflect a high scholarship and expertise of the contributing authors and the book is definitely a mandatory reading for anyone who is professionally active in this promising field of Ambient Intelligence.

It goes without saying that it is my strong belief that Ambient Intelligence will be become an influential new computing paradigm and I am confident that the present volume will contribute to the steady growth of the number of AmI Google hits as many other researchers will feel the need to refer to many of its most inspiring chapters.

Emile Aarts, Scientific Program Manager, Philips Research, Eindhoven, The Netherlands, August 2007

Information processing technologies should be able to support people in every aspect of their everyday life, with information processing units embedded in the environment which communicate with portable or wearable personal devices. The keywords are “here, now and me” rather than “wherever, whenever, whoever”.

We want to add one more element to the vision, the concept of situatedness. Ubiquitous computing and ambient intelligence put stress on having sensors in the environment to sense the context of the user. However, I define situation not as something to be sensed or computed, but something to be used as it is. We should not design a system that acts on the sense-represent-compute-act cycle. Rather, we should design a system that can set up the context appropriately.

Cyber Assist project was conducted under the above philosophy. In this chapter, we overview the project under the light of ambient intelligence.

Ambient Intelligence as a vision for future information systems is currently receiving more and more attention in research and industry. In this article, we investigate how related technology domains, such as Artificial Intelligence, Multi-Agent Systems and the Semantic Web can provide methods and IT infrastructures to support Ambient Intelligence middleware and applications. By means of three case studies ranging from personal assistance to megacity management we identify functional and non-functional requirements. Based on this study we propose AmIRA as an Ambient Intelligence Reference Architecture with a focus on methods, architectures and components. We illustrate the instantiation of AmIRA by describing the implementation of three industrial applications related to the previously presented case studies.

The basic goal of context-aware systems is to make software aware of the environment and to adapt to their changing context. For that purpose, the core problem is to have a powerful context model. While significative formalizations have been proposed, context models are either expressed through logical formalisms or with ontology-based approaches. The major problem with all that approaches is that they suffer from the chronic insufficiency of first-order logic to cope with dynamic change and especially, to solve the frame problem. Therefore, building context-aware software is a complex task due to a lack of appropriate formal models in dynamic environments. In this paper, we propose a model which combines the strengths of both approaches while trying not to carry their specific weaknesses into the resulting formal framework. For this purpose, the formal model relies both on a knowledge representation with ontologies and on a logical reasoning with Dependent Record Types (DRT) based on Intuitionistic Type Theory and the Curry-Howard isomorphism. This logic modelling aims to be applied to any kind of process-based applications.

This work deals with a distributed knowledge representation and active information fusion system suitable for use in Ambient Intelligence applications. The architecture – which is modeled as a multi-agent system – exploits a sound formal model to relate numerical data to symbolic representations, thus being able to reason about predicates, situations and contexts. In particular, agents collaborate to perform an intelligent multi-sensor data fusion according to the guidance of an active classification layer. Experimental results performed both in simulation and in a real set-up are discussed with respect to a number of implications and future directions related to the system approach.

We are interested in situations where multiple uncoordinated non-professional programmer end-users want to exploit the Ambient Intelligence (AmI) infrastructure on their own, without calling in embedded systems programmers, in order to support their daily activities. Our goal is allowing them to achieve this objective through on-the-fly creation and execution of high-level programs that we call uQueries (for user-defined or ubiquitous queries). The key challenge then is to support the cost-effective and stepwise development of uQuery engines – systems for end-user programming and execution of uQueries. We present a meta-level architecture that addresses this issue by leveraging Artificial Intelligence methods that make possible the separation of uQuery representation and reasoning concerns from those of their effective execution through model-to-code transformation. We show that (1) interconnections between ambient devices may be dynamically specified as control flows between high-level descriptions of their primitive functionality, (2) specifications may be elaborated by concurrent, uncoordinated end-users through a Web interface, and (3) they may be automatically distributed and concurrently executed on ambient devices as a system of mobile agents. We have created a prototype of this architecture, the Ambiance Platform, which has allowed experimental validation of the approach using an application scenario proposed in the state-of-the-art of relevant research areas. This experience led us to identify important issues to be explored, including dynamic and seamless integration of sensor and actuator nodes into the system. Furthermore, opportunities exist for significant performance and resource use optimization, for instance by integrating learning mechanisms into uQuery specification, transformation and execution.

The XPod system aims to integrate awareness of human activity and multimedia preferences to produce an adaptive system that plays contextually appropriate media. The XPod project introduces a “smart” music player that learns its user's preferences and activity, and tailors its music selections accordingly. We have experimented with various physiological sensing platforms to measure a user's physiological state. These devices are able to monitor a number of variables to determine the user's levels of activity and motion to predict what music is appropriate at that time. The XPod user trains the player to understand what music is preferred and under what conditions. After training, XPod can predict the desirability of a song given the user's physical state. XPod learns a users listening preferences from the interaction with the user and from collaborative filtering

Busy people would benefit from ambient, adaptive tools for reminding them what they have to do, depending on various contextual parameters. We are developing an adaptive and expressive agent, which learns when and how to notify users about self-assigned tasks and events. In this paper, we focus on two crucial issues for such a system: the selection of the input data set, and the design of an appropriate mechanism to get user feedback without being too intrusive. We describe in particular the inputs that encapsulate the current context of the agent: relative temporal distances, historical information about reminders and categories, and the context of both the user and the hosting device.

Media Arts have acquired informal expertise in the design of spaces and events shared by sentient systems and people. Because of this, Media Arts should be of interest to investigators as they begin to seriously consider social interactions as an integral part of the Ambient Intelligence research agenda.

We have seen the future of ubiquitous computing, and it's gloomy. While the various technologies keep advancing at a breakneck pace, and half of human race are already carrying computing systems with them, the goal of transparently integrating all those devices and networks to usable ubiquitous systems seems to be receding further away every day. We are seeing around us major shortcomings in terms of usability, interoperability, and security, and expect that the situation in integrated systems will only get worse. The next decade will bring wonderful advances in individual devices, services and technologies, but the overall picture is not as rosy. In this chapter, we discuss the drivers of this undesirable development, while also observing some signs of more positive trends.