Andreas Nüchter - Research

Mission

The research group on automation and machine vision started in August 2009. "Automation is the use of control systems, in concert with other applications of information technology, to control industrial machinery and processes, reducing the need for human intervention. [...] automation greatly reduces the need for human sensory and mental requirements as well. Processes and systems can also be automated" (source wikipedia:automation).
The performance of automated systems can be significantly improved by incorporating semantic information in the control. We are aiming at building bindings between raw sensor output, i.e., sub-symbolic data, to knowledge representations, i.e., symbols used by reasoning.

For the above stated challenge we develop appropriate methods, look at interesting applications and build smart systems (cf. Figure 1). A key sceintific issue of our research are sensor data processing methods and algorithms. For example, state of the art sensors such as 3D laser scanner or ToF cameras acquire reliably a huge amount of environmental data. Reasoning about fused/multimodal sensor data over short term, medium term and life-long periods are of interest. In addition active data acquisition and sensor based planning are key topics. Knowledge representations for hybrid control architectures for intelligent systems are subject of research.

Figure 1: Research in applications, methods and systems.

The following figure depicts the research I was recently involved (click to enlarge). For more details, please refer to my publication page.

Figure 2: Past achievements. (click to enlarge)