Cognitive Neuroinformatics

In the course of evolution biological systems have developed cognitive and intelligent abilities which are still more efficient and powerful than today’s technical systems. Seeing, hearing, tactile perception as well as localisation in and exploration of spatial environments are some examples of such cognitive abilities.

The process of learning and subsequent decision-making are examples of more human-like cognitive capacities. First, we try to understand and formalize these abilities, such that we can map them to theoretical approaches. Finally, we seek to transfer them into intelligent technical systems. The resultant systems combine low level cognitive abilities like pattern recognition with h igher level cognitive capacities like knowledge representation, reasoning, and planning.

Our research includes further developments in theories of soft computing concerning automated learning, processing and representation of uncertain knowledge (e.g. Dempster-Shafer Theory). Further research interests are the development of adaptive and parallel inference processes, which are based on the principle of information gain. In addition, we investigate and model the fusion of multisensory information, numerosity, natural scene statistics, SLAM (simultaneous localisation and mapping), attentional processes, and deep learning. The areas of application are bio inspired systems, assistive systems for elderly people and autonomous deep space navigation.