The goal of our research is to advance the state of the art in sensor-based control of (in the first place) industrial robot controllers. Our expertise lies mainly with force and touch sensing, but more and more integration with 3D visual servoing is required. This talk will give an overview of the two major research tracks during the last four years: (i) Bayesian estimation for 6D robot tasks with force and vision sensors, and (ii) the development of a large-scale, hard realtime and open source software framework to support intelligent robotic systems and advanced machine tools.