Thematic issue on contribution of Artificial Intelligence to Ambient Intelligence

Abstract

∗ Corresponding author. E-mail: anind@cs.cmu.edu. Since its inception, the vision of Ambient Intelligence (AmI) holds great promise in permeating everyday life thereby changing the nature of almost every human activity [6]. In the AmI paradigm, intelligent computation will be invisibly embedded into our everyday environments through a pervasive transparent infrastructure (consisting of a multitude of sensors, actuators, processors and networks), which is capable of recognizing, responding and adapting to people and activity in a seamless and unobtrusive way. AmI promises to advance the quality and trustworthiness of a huge number of applications in all domains of human activity, including, for example, health care, education, entertainment, administration, transportation, construction, and the use of energy resources [1]. Early research efforts concentrated on the conception of innovative scenarios using the AmI vision and the development of the system architectures, software and technology required to realize them. Early architectures adopted either a bottom-up approach, leading to the development of systems consisting of networked objects having digitally augmented functionality, or a middle-out approach, focusing on the design of middleware and the development of services on top. Nevertheless, during these early years, important problems, such as heterogeneity, discovery, transparency, context representation and adaptation were encountered, but the restricted scope of the systems required solutions specific to particular settings [3,8,12,13,19]. Hence, although Artificial Intelligence techniques were adopted, they were employed to solve specific problems within a well defined context [4,5,16]; some of these early architectures have found their way in commercially available components for smart homes or communicating information devices (mobile phones, personalized medical devices, media capture and processing devices, etc.). Nowadays, a new generation of AmI technology is gradually becoming available, as large scale complex architectures and systems are being developed to support an ever increasing number of domains of human activity. Such systems are often described as Ambient Ecologies, offering a rich set of components and interfaces, having a dynamic and adaptive structure and implying that people, as well as other goalpursuing actors (agents, avatars, robots, etc.) are regarded as being their integral part [9]. Complexity results from the large number of active nodes, services and interactions, the vast number of data and information, the multitude of locations and the increasing number of users and tasks that have to be supported. In addition, as people become familiar with the AmI paradigm, human-centred requirements become more important; these include adaptation, consistency, trustworthiness, security, efficiency, multimodal interaction and controlled transparency. In addition, an important body of research is dedicated to the design of biologically inspired AmI systems, capable of exhibiting cognitive abilities, such Journal of Ambient Intelligence and Smart Environments 1 (2009) 207–209 DOI 10.3233/AIS-2009-0029 IOS Press