Motion analysis of interaction mode control using principal component analysis

Recently, robot application has been widely used not only in industry but also human society. Hereafter, in order to extend the range of work and kinds of motion in human society, it is needed to think about what human is and what human motion is. In conventional method for analysis of human motion, visual-based approach has widely researched. However, force adjustment is important information for many kinds of task such as processing technology and surgical operation. In addition, in order to acquire the advanced motion for robots, feature amount of advanced motion is needed to analyze. First step of motion analysis based on position and force information, the condition which has theoretical value is conducted. This paper proposes motion analysis method for interaction mode control systems using principal component analysis(PCA). Using the proposal, the dominant component is directly estimated from the motion information. To confirm the effectiveness of the method, motion data abstracted by automated control is used. Validity of the proposal is confirmed by experiment of interaction mode control. Experimental results in this paper are compared with the theoretical value.