Extracting Motion Skills from Expert's Proficient Operation Records Using Recurrent Neural Network

Abstract This paper presents a method or extracting motion control sKills using recurrent neural networks called Elman networks. We made some experiments using the raw data acquired while a human performs a simple task of fetching objects by stretching and folding his/her arm, and demonstrate that the network can learn invariant features of the generalized motion concepts, classify the motion by referring to self-organized skill performer's intentions, and understand a task structure of the observed human bodily motion. These capabilities are essential dor macihne intelligence to establishing the human-robot shared autonomy, a new style of human-machine collaboration proposed in the area of pie robotics.