Towards the robotic co-worker

Various human-friendly motion control methods were presented and analyzed. These are independently useful tools for numerous applications as they open up entirely new robot behaviors. However, due to their complex interrelationship in this chapter it is discussed how to integrate the presented methods into a more general hybrid state-based control architecture. Even though the focus is on robotic co-workers, the elaborated schemes are also applicable to service robots. The implementation of such a sensor-based robotic co-worker that brings robots closer to humans in industrial settings and achieve close cooperation is currently a challenging goal in robotics. Pioneering examples of intimate collaboration between human and robot, whose origin can be found in [10], are Intelligent Assist Devices (IADs), as the skill assist described in [18]. In 1983 a method was proposed at DLR for allowing immediate “programming by touch” of a robot through a force/torquesensor- ball [8], see Fig. 8.1 (left). Despite being a common vision in robotics the robotic co-worker has not become reality yet, as there are various open questions still to be answered. Apart from the control and safety aspects, the architectural level also poses significant challenges.