“Ambient Intelligence” implies intelligence that is all around us. It is a developing technology that will increasingly make our everyday environment sensitive and responsive to our presence. It is a potential future in which we will be surrounded by intelligent objects and in which the environment will recognize the presence of persons and will respond to it in an undetectable manner. In an AmI environment people are surrounded with networks of embedded intelligent devices that can sense their state, anticipate, and perhaps adapt to their needs.

Ambient Intelligence builds on three recent key technologies:
• Ubiquitous Computing
• Ubiquitous Communication
• Intelligent User Interfaces.
Some of these concepts are barely a decade old and this reflects on the focus of current implementations of AmI.
Ubiquitous Computing means integration of microprocessors into everyday objects like furniture, clothing, white goods, toys, even paint.
Ubiquitous Communication enables these objects to communicate with each other and the user by means of ad-hoc and wireless networking.
An Intelligent User Interface enables the inhabitants of the AmI environment to control and interact with the environment in a natural (voice, gestures) and personalized way .



• Sensors, vision, and networks
• Mobile and pervasive computing
• Human-centered interfaces
• Artificial Intelligence
• Robotics
• Multi-agents
• Societal applications and implication
• Internet of Things


AmI systems can be deployed in many possible environments. Below we describe some of these environments in order to better illustrate the scope of the idea.
Scenario 1: An instance of the concept of Ambient Intelligence is a Smart Home. Here an AmI specification may include the following details. The meaningful environment is the house, including the backyard and a portion of the front door as these areas also have sensors. Objects are plants, furniture, and so on.There are also multiple sensors in S, movement sensors,pull cord switch, smoke detector, doorbell detector,pressure pad, plus switch sensors for taps, a cooker and a TV. In addition, there is a set of actuators, as the taps, cooker and TV also have the capacity to be turned on and off without human assistance.
Scenario 2: Let us consider a specific room of ahospital as the environment, whit a patient monitored for health and security reasons. Objects in the environment are furniture, medical equipment, specificelements of the room like a toilet and a window. Interactors in this environment will be the patient, relativesand carers (e.g., nurses and doctors). Sensors can be movement sensors and wrist band detectors for identifying who is entering or leaving the room and who is approaching specific areas like a window or the toilet. Actuators can be microphones within the toilet to
interact with the patient in an emergency. Contexts of interest can be “the patient has entered the toilet and
has not returned after 20 minutes” or “frail patient left the room.
Scenario 3: Assume a central underground coordination station is equipped with location sensors to track the location of each unit in real-time. Based on the time needed to connect two locations with sensors, the system can also predict the speed of each unit.Examples of objects in this environment are tracks and stations. Interactors are trains, drivers and command centre officers. Sensors are used for identification purposes based on ID signals sent from the train.Other signals can be sent as well, e.g., emergency status. Actuators will be signals coordinating the flow of trains and messages that can be delivered to each unit in order to regulate their speed and the time they have to spend at a stop. Contexts of interest can be “delays” or “stopped train”. One interaction rule can be “if line blocked ahead and there are intermediate stops describe the situation to passengers”.
Scenario 4: Lets assume a school where students are monitored to best advise on balancing their learning experience. The objects within a classroom or play ground are tables and other available elements. The interactors are students and teachers. The sensors will identify who is using what scientific kit and that in turn will allow monitoring of how long students are involved with a particular experiment. Actuators can be recommendations delivered to wristwatch-like personalized displays. Contexts of interest can be “student has been with a single experimentation kit for too long” or “student has not engaged in active experimentation”.
The first context will trigger a rule “if student has been interacting with one single kit for more than 20 minutes advise the student to try the next experiment available” whilst the second one can send a message to a tutor, such as “if student S has not engaged for more than 5 minutes with an experiment then tutor has to encourage and guide S”.
Scenario 5: When a fire brigade has to act then the environment can be a city or a neighborhood. Streets can be equipped with sensors to measure passage of traffic within the areas through which the fire brigade truck might go through in order to reach the place where the emergency is located. Objects here will be streets and street junctions. Interactors will be cars. Actuators can be traffic lights as they can help speed the fire brigade through. A context will be a fire occurring at peak time with a number of alternative streets to be used. An interaction rule can be “if all streets are busy, use traffic lights to hold traffic backfrom the vital passage to be used”.
Scenario 6: If a production line is the environment then different sensors can track the flow of items at critical bottlenecks in the system and the system can compare the current flow with a desired benchmark. Decision makers can then take decisions on how to proceed and how to react to the arrival of new materials and to upcoming demands. Different parts of the plant can be de/activated accordingly. Similarly, sensors can provide useful information on places where there has been a problem and the section has stopped production, requiring a deviation in flow. Objects here are transportation belts and elements being manufactured whilst actuators are the different mechanisms dis/allowing the flow of elements at particular places. A context can be “a piece of system requiring maintenance” and a related interaction rule can be “if section A becomes unavailable then redirect the flow of objects through alternative paths”.


AmI has a strong emphasis on forcing computing to make an effort to reach and serve humans. This may sound the obvious expectation from computing systems but the reality is that so far humans have to do the effort to specialize themselves in order to enjoy the advantages of computing. It is expected that enforcing this requirement at the core of the area will constitute a major driving force and a turning point in the history of computer science. The technological infrastructure seems to be continuously evolving in that direction, and there is a fruitful atmosphere on all sides involved: normal users/consumers of technology, technology generators, technology providers and governmental institutions, that this paradigm shift is needed and feasible. Still, achieving that capability is far from easy and certainly is not readily available at the moment. The short history of computer science is full of problems which turned to be harder than expected and there is plenty of examples of important systems that crashed.
The very fact that makes AmI systems strong can be also their more serious weakness. If humans are put at the centre of the system and made more dependant on the technological environment (we called this an e-bubble), reliability on that e-bubble will be at the level of safety critical systems. Since these systems are autonomous and proactive,
predictability and reliability should not be underestimated if we want the environments where we live and to be helpful and safe.