Following human guidance to cooperatively carry a large object

摘要: Carrying a large object like table is task that cannot be solved by single robot or human, but requires two workers, for example one human and robot. For this human-robot cooperation, the must perceive synchronize with its motion. It also to carry. In paper, we present an approach uses arm compliance follow guidance on fast time scale moves base restore nominal position arms. perceiving object, acquire model of it using RGBD camera match current measurements. This real-time pose estimate suitable approaching grasping it, as well detection lifting lowering ground again. We evaluate our in lab experiments has anthropomorphic upper body omnidirectional base. report successful public demonstration @Home league at RoboCup 2011.