Den här sidan är utskriven från Högskolan i Halmstads webbplats (www.hh.se). Texten uppdaterades senast den 2017-04-25. Besök webbplatsen om du vill vara säker på att läsa den senaste versionen.
The goal with artificial intelligence (AI) research and development is to construct systems that behave intelligently. Today it is common to assume that human experts define the task to be performed, what data should be collected, how should it be represented, and what metrics to use for performance evaluation. This means that these systems are designed or programmed, which leads to them breaking when the context changes.
Our aim with Aware Intelligent Systems research is to approach the construction of systems that can do life-long learning; systems that require less supervision and can handle surprising situations. In order to do so, the systems must become more aware and able to learn on their own, to handle events that are unknown at the time of design. Our research focuses on creation of systems that, as autonomously as possible, can construct knowledge from real life data capturing the interaction with the environment.
Those goals are match contemporary societal challenges, and we collaborate with many industrial partners. For example, recent developments in wearable sensors has inspired a vision of personalised health; modern energy production is becoming more volatile, diverse and distributed; transport efficiency depends on better maintenance and monitoring solutions. All those areas require novel solutions that build upon available data and require autonomous knowledge creation.
The research questions we explore include selecting what data to collect and how to find general and robust representations; how to do (semi-)autonomous deviation detection, dealing with concept drift and seasonal variations; how to associate events from different data sources; is it possible to explain why certain things have happened.
Aware systems research is a systems science, i.e., there are many interconnected parts and the results need to address several aspects, tying them together. To enable this, we build demonstrators to showcase what this means, with sets of tools for all levels.
The Technology Area is responsible for carrying through and developing courses within artificial intelligens, image analysis, learning systems, mechatronical systems, signals and systems, and control theory.