Ambient Intelligence Lab
The Ambient Intelligence Lab (AIL) is an avant-garde research lab at Carnegie Mellon University. As a part of the federal-funded Cylab, the lab innovates technologies in cognitive agent networks, video intelligence, 3D surface analysis, visual data mining, and digital human modeling. The lab is developing the next generation computing and networking systems with innovative frameworks.
What is Ambient Intelligence?
As the volume and complexity of information grows exponentially, information-overload becomes a common problem in our life. We find an increasing demand for intelligent systems to navigate databases, spot anomalies, and extract patterns from seemingly disconnected numbers, words, images and voices. Ambient Intelligence (AmI) is an emerging paradigm for knowledge discovery, which originally emerged as a design language for invisible computing and smart environments. Since its introduction in the late 1990's, AmI has matured and evolved, having inspired the development of new concepts for information processing, as well as multi-disciplinary fields including computer science, interaction design, mobile computing, and cognitive science.
In a broad sense, Ambient Intelligence is perceptual interaction, involving common sense, serendipity, analogy, insight, sensory fusion, anticipation, aesthetics and emotion all modalities that we take for granted in human interaction but have normally been considered out of reach in the computational world. We discover knowledge through the windows of our senses: sight, sound, smell, taste and touch, which not only describe the nature of physical reality but also connect us to it. Our knowledge is shaped by the fusion of multidimensional information sources: shape, color, time, distance, direction, balance, speed, force, similarity, likelihood, intent and truth. Ambient Intelligence is not only interaction but also perception. We do not simply acquire knowledge but rather construct it with hypotheses and feedback. Many difficult discovery problems become solvable through interaction with perceptual interfaces that enhance human strengths and compensate for human weaknesses to extend discovery capabilities. For example, people are much better than machines at detecting patterns in a visual scene, while machines are better at detecting errors in streams of numbers.