Estimation of Unmeasured Golf Swing of Arm Based on the Swing Dynamics

Abstract

The design of wearable golf swing monitoring system is challenging because it has to resolve the trade-off between the convenience of wearability and the quantity of the data collected. If the swing dynamics are describable by a multi-segment mechanical model, the motion measurement of one segment could inform that of the others by mechanical relationship. In this study, we propose a method of estimating unmeasured arm swing data based on the dynamic characteristics of swing motion. Specifically, we hypothesized that the arm swing dynamics could be represented by rigid body dynamics with the specifiable joint toque profiles, which is formulated as a linear combination of joint kinematics. To simplify the complexity of the multi-joint 3D motion, we proposed a projection analysis of 3D motion data onto the two dimensional plane of the downswing. To validate the proposed model and the estimation methods, we collected the golf swing data of driver and iron 7 club of 12 subjects with various skill levels. Results showed that the subjects made swing plane very consistently during the repeated swing trials regardless the skill levels. The joint torque data were well represented by the linear combination of joint kinematics during the downswing.