Abstract
The Organic Computing initiative uses life-like properties such as self-organisation, self-optimisation and self configuration towards building today's technical systems as flexible, robust, and adaptive systems. In a previous paper, we proposed a system for coordinating semi-autonomous agents under the framework of Organic Computing. It uses abstractions of observer and controller to add robustness and solve scheduling/allocation problems. In this context, the path planning and the observation of the agents were presented and also the detection of deviations in different situations was discussed. In this paper, we introduce control features of the system designed to deal with these types of deviations. That leads in turn to intervene in time when it is necessary, so that the system remains demonstrating robustness. Furthermore, this paper addresses the conflict between a central planning algorithm and the autonomy of the agents. A hybrid central/self-organizing multi-agent system is introduced solving this conflict.