International Journals of Advanced Research in Computer Science and Software Engineering Research Article February ISSN: 2277-128X (Volume-8, Issue-2) a 2018 Ambient Intelligence: Principles, Current Trends, Future Directions Manas Kumar Yogi * K. Chandrasekhar Bhanu Prakash Asst. Prof., CSE Department, Asst. Prof., CSE Department, B.Tech II Year Student, CSE Pragati Engineering College, Pragati Engineering College, Department, Pragati Engineering Andhra Pradesh, India Andhra Pradesh, India College, Andhra Pradesh, India [email protected] [email protected] [email protected] Abstract— In this paper we have presented the involved principles, trends, future directions of ambient intelligence. in the first section we have elucidated concept of ambient intelligence with the prevalent need of intelligent communication with the help of designing knowledgeable entities. We have presented the design process of I-blocks with its inherent merits. we have also discussed the various design concepts of ambient intelligence objects. Finally, we presented the current research directions to motivate societal needs of human beings. Keywords— ambient, Computer Vision, intelligence, smart, ubiquitous I. INTRODUCTION This introductory paper describes Ambient Intelligence (AmI) from the perspectives of researchers working in the field of Artificial Intelligence and Computer Vision. It is for the reader to get acquainted with some of the ideas .That will be explored in greater detail in the following content .Ambient Intelligence is a term that was introduced by the European community to identify a paradigm to equip environments with advanced technology and computing to create an ergonomic space for the occupant user. Here the term ergonomic is used in a broad sense, encompassing both better living environment, secure space, but also an active, almost living space around us; capable of aiding us with daily chores and professional duties. Later on in this paper you will be able to see examples of enhanced homes for the elderly, intelligent buildings, devices built for education and entertainment and conventional visual surveillance systems, easily portable to other domains of application, such as the training of professionals. The AmI paradigm can be realized only through a number of technologies, all involving modern computing hardware and software. In particular, an AmI system requires the use of distributed sensors and actuators to create a pervasive technological layer, able to interact transparently with a user, either passively by observing and trying to interpret what the user actions and intentions are, but also actively, by learning the preferences of the user and adapting the system parameters to improve the quality of life and work of the occupant. The Essex approach-The Department of Computer Science at Essex University carries out research in the field of AmI. Their approach is focused on the implementation of AmI as indoors smart environments. In particular, state of the art Artificial Intelligence techniques are employed in the implementation of a futuristic IntelligentDormitory (iDorm). The following contents will describe their approach and some of the employed technology. The main idea here is to illustrate a concrete example of AmI put into practice with success. Later on in this paper you will be able to understand how Computer Vision could enhance the iDorm and typical AmI enabled smart environments. II. CURRENT TRENDS Ambient Intelligence: the contribution of different disciplines Ambient Intelligence represents a vision of the future where people are surrounded by electronic artifacts and environments, sensitive and responsive. Ambientintelligence technologies are expected to combine concepts of ubiquitouscomputing and intelligent systems putting humans in the centre of technologicaldevelopments. This represents a long-term objective for European researchbringing together researchers across multiple disciplines like computer science, electronics and mechanical engineering, design, architecture, social sciences,software engineering. Key concepts of ambient intelligence are: A. Ubiquitous Computing: that is wired, wireless and ad-hoc networkingthat exploit highly portable or else numerous, very-low-cost computingdevices; discovery mechanisms, software architectures, system integrationand prototyping, portable devices; © www.ijarcsse.com, All Rights Reserved Page | 43 Yogi et al., International Journal of Advanced Research in Computer Science and Software Engineering 8(2) ISSN(E): 2277-128X, ISSN(P): 2277-6451, pp. 43-50 B. Context Awareness: sensors, tracking and positioning, smart devices, wearable, models of context of use, software architectures for multiplatform interfaces; C. Intelligence: learning algorithms, user profiling, personalization and adaptively ,autonomous intelligence, agent based user interfaces; and D. Natural user-system interaction: ambient interfaces, multimodal interaction, innovative interaction styles and concepts. Appreciation of the social interactions of objects in environments, and the cultural values the new environments contribute to determine. At a general level, the interest for Ambient Systems is driven by both technology, design and user orientation. Such systems should allow to achieve interoperability of devices and integration of new services through available tools; and to orchestrate devices and entities to support and enhance human activities. To achieve this, systems have to be so designed as to support a range of human activities, and be intimately inter-twined with physical settings, consisting of spaces, places and everyday objects and materials. These physical settings and devices share some important characteristics: ubiquity, transparency, intelligence, furthermore they can be personalized, they are adaptive and anticipatory. Ubiquity refers to a situation in which we are surrounded by a multitude of interconnected embedded systems. Transparency indicates that the surrounding systems are invisible and moved into the background of our surroundings. Intelligence refers to the fact that the digital surroundings exhibit some form of adaptation, e.g. by recognizing the people that live/interact in these surroundings in order to adapt themselves to them, learn from their behaviour, and possibly show emotion. Furthermore, as said above devices for ambient intelligence should be personalisable,adaptive and anticipatory, characteristics that may receive a fundamental contribution from modern artificial intelligence. Investigation and understanding of adaptation does not only lead to the possibility of creating adaptive devices, but may also be the corner- stone for creating devices that can be personalised. If devices are to be personalised, they should be able to change to the individual users need, i.e. they should adapt since they should change to fit different conditions. Also the anticipatory characteristic of ambient intelligence devices may demand adaptation, since anticipation demands the ability of a system to internally extrapolate future interactions with the surrounding environment, and hence adapt in an internal system. The extrapolation can be viewed as an adaptation process (a change in the internal system to fit the future condition). In this light, we view modern artificial intelligence as a most valuable field for supporting ambient intelligence. But modern artificial intelligence, or more precisely embodied artificial intelligence ,may also play an even more crucial role in the creation of ambient Intelligence. Again, looking at the characteristics of ambient intelligence, we find that devices should be built into our natural surroundings and we should be able to interact with these devices in a natural way. Embodied artificial intelligence research ,tells us that when creating physical entities with adaptive characteristics it is often advantageous to try to find the right balance between control, hardware, material, and energy use, and that this balance may be found through a bottom-up approach. The bottom-up approach is characterized by initially creating a minimal system with a basic behaviour, and step-by-step add components only as they become necessary for new, more advanced behaviours built on top of each other. We believe that embodied artificial intelligence research is one of the disciplines that may provide some answers on how to create ambient intelligence in different domains. In the development of ambient intelligence, we may also find inspiration from the development and research in tangible interfaces, since also within this field, research is focused on the development of seamless interaction with physical information processing systems. Tangible interfaces may represent a valid opportunity for the development of novel interactive technologies that can overcome the limitation of the current computer-based technologies constrained to screen, mouse and keyboard interaction. Several research groups developed haptic and tangible interfaces, and today, the many approaches to tangible interfaces differ in implementation and focus, while, at the same time, sharing certain main characteristics. This is in particular the feature that characterizes most of the existing prototypes of tangible interfaces. These physical objects are mainly interfaces either to a digital counterpart (e.g. navigational systems) or to a physical target system (e.g. use of tangible devices as a control systems of other physical or digital worlds). An