Intelligent environment

Intelligent Environments (IE) are spaces with embedded systems and information and communication technologies creating interactive spaces that bring computation into the physical world and enhance occupants experiences. 'Intelligent environments are spaces in which computation is seamlessly used to enhance ordinary activity. One of the driving forces behind the emerging interest in highly interactive environments is to make computers not only genuine user-friendly but also essentially invisible to the user'. Intelligent Environments (IE) are spaces with embedded systems and information and communication technologies creating interactive spaces that bring computation into the physical world and enhance occupants experiences. 'Intelligent environments are spaces in which computation is seamlessly used to enhance ordinary activity. One of the driving forces behind the emerging interest in highly interactive environments is to make computers not only genuine user-friendly but also essentially invisible to the user'. IEs describe physical environments in which information and communication technologies and sensor systems disappear as they become embedded into physical objects, infrastructures, and the surroundings in which we live, travel and work. The goal here is to allow computers to take part in activities never previously involved and allow people to interact with computers via gesture, voice, movement, and context. The idea of having an artificial intelligence capable of managing an environment, recollect data, and respond in consequence is older than we would expect. In the novel 2001: A Space Odyssey from 1968, long before the microcomputers revolution, you have the fictional character HAL 9000, a computer capable of controlling the different sensors and systems of the environment and using them as extensions of itself. The character Proteus from the 1973 novel Demon Seed also portrays the same characteristics of an artificial intelligence controlling an embedded environment. By the time these two novels were released, the idea of a computer controlling the environment that surrounds us was not broadly accepted by the community since both characters played the role of evil machines whose only objectives included the control over humans. The term 'Intelligent environments' is a concept and expression originally created by Peter Droege for his homonymous Elsevier publication of 1997, a project that commenced in 1986: Intelligent environments - Spatial Aspects of the Information Revolution https://www.sciencedirect.com/book/9780444823328/intelligent-environments. The 1986 project was his winning entry into the Campus City Kawasaki competition in Japan, seeking to apply the benefits of information technology and advanced telecommunications to an entire city, its societal empowerment, trans-industrial prosperity and, above all, its environmental redemption. It is not until 1991 with the introduction of ubiquitous computing by Mark Weiser when we start seeing an inclination from the scientific community to study the area of computing outside of the typical machine with a keyboard and a screen. It became something that could be potentially implemented into anything that surrounds us, proposing casual access to computing to any user. In 1996 Hashimoto Laboratory at the University of Tokyo developed the first research on intelligent spaces. J.-H. Lee and H. Hashimoto designed a room with a homemade three-dimensional tracking sensor and mobile robots, all this connected to a network. The idea was for the robots to support the person in the room with different tasks with the help of vision cameras and computer sets, becoming one of the first intelligent environment. At first, intelligent spaces were designed with the only objective to help people with physical tasks. Robots included in the room would help people to grab objects as well as support people with disabilities to do certain jobs. This idea started shifting into the concept we have today of intelligent environments, not only an environment to support people but also robots. The intelligent space became a platform that extends the censorial capacity of anything connected to it. If we start designing products, either software or hardware around this intelligent environments, the effort needed to complete all kinds of tasks would be drastically reduced. Practical implementation of intelligent environments implies the solution of many challenges. Pervasive computing systems embedded in IE need to be proactive and to accomplish this, it is crucial that systems can track and determine the users' intent. The challenge here is finding that action that supposedly will help the user rather than hinder him. As of right now, algorithms behind the intelligent environments are constantly being reworked by the simple method of trial and error in artificial environments. It is not until a programmer can see an accurate enough level of prediction for the product to become commercialized. The degree of accuracy of intelligent environments depends on the task they want to accomplish. Some simple actions that do not substantially affect the user can admit more failures in the predictions than other functions that hold more responsibility. Still, there are always actions that cannot be fully predicted by the IE and needs some input from the user to be completed. One of the most significant challenges as of right now is determining which are those actions that are required for user input and how to create algorithms capable of eliminating that input so that the usability of the systems improves. By the other hand, pro-activity of such environments has to be handled very carefully. Pervasive computing systems are supposed to be minimally intrusive and at the same time be capable of taking decisions that will help users. One way to achieve that is making those systems capable of modifying their behavior based on the user's state and surroundings. Here again, some challenges arise: What are the required data and information that a system needs to be context-aware? How frequently should that information be measured and consulted without hurting system performance? The goal is to create an IE capable of reacting fast and accurate to the needs and inputs of the user so it would be unnecessary for the sensors to record information that will not help the algorithms make the correct action to what is happening. Recognizing important data and filtering the environment to search for the appropriate place to obtain it results in a great challenge.