Learning to Select and Generalize Striking Movements in Robot Table Tennis

Abstract

Learning new motor tasks autonomously from interac-tion with a human being is an important goal for both robotics and machine learning. However, when moving beyond basic skills, most monolithic machine learning approaches fail to scale. In this paper, we take the task of learning table tennis as an example and present a new framework which allows a robot to learn cooperative ta-ble tennis from interaction with a human. Therefore, the robot first learns a set of elementary table tennis hit-ting movements from a human teacher by kinesthetic teach-in, which is compiled into a set of dynamical sys-tem motor primitives (DMPs). Subsequently, the system generalizes these movements to a wider range of situa-tions using our mixture of motor primitives (MoMP) ap-proach. The resulting policy enables the robot to select appropriate motor primitives as well as to generalize be-tween them. Finally, the robot plays with a human table tennis partner and learns online to improve its behavior. We show that the resulting setup is capable of playing table tennis using an anthropomorphic robot arm. 1