Structure variable multi-sensoric supervisory control of human interactive robots

A new structure variable NF based supervisory control concept for advanced multi-sensory robots (humanoid robots) is presented. It relies on a reliable identification of the actual process phases or fault cases. A dynamic reconfiguration of optimal situation-specific sub-controllers is activated for each process phase and fault scenario identified by means of a heuristic fuzzy logic mode selector. Initial results with its implementation within the framework of the on-going German Humanoid Robot Project is also reported.