Cognitive Control Technology for a Self-Optimizing Robot Based Assembly Cell

In the face of global competition there is a great danger for countries with high labor costs (e.g. Germany) to lose more and more production plants to low-wage countries. Almost inevitably there will be a relocation of after-sales services as well as of research and development. Eventually this will cause a significant decline of wealth. For this reason especially high-wage countries are always striving for higher productivity of production processes. On the other hand the products have to be of high-end quality to ensure an advantage in the market. Thus there is an obvious dilemma between planning-orientation and value-orientation which has to be resolved. This could possibly be obtained by shifting planning efforts to the runtime system and at the same time enabling the system to adapt to changing requests and circumstances. In order to get there, automation technology is definitely playing a key role in present-day highly automated production processes. Unfortunately classical automation technology has not been supporting this kind of self-organizing, self-controlling and self-optimizing behavior. This paper introduces an approach to make production systems more “intelligent” based on the idea of a cognitive control architecture. At first the motivation and the research vision are introduced followed by an outline of the research approach. As a concrete example of an application a robot based assembly cell is described. The methods used and insights gained so far are presented in the second part, followed by an outlook towards future activities.Copyright © 2008 by ASME