A sensor-aided self coaching model for uncocking improvement in golf swing

Abstract

This paper describes an autonomous kinematic analysis platform for wrist angle measurement that is capable of evaluating a user's uncocking motion in his or her golf swing and providing instructional multimodal feedback to improve his or her skills. This uncocking motion, which is a characteristic movement of the wrist during the golf swing, is an important factor in achieving accurate ball hitting and long driving distances, but is difficult to measure. In order to efficiently compute the wrist angle for uncocking evaluation, we present a sensor-based intelligent Inertial Measurement Unit (IMU) agent that collects three-dimensional orientation data during the golf swing from two IMU sensors placed on the forearm and on the golf club. It accurately analyzes changes in wrist angle to detect uncocking throughout the sequence of golf swing motions. In this paper, we first introduce the design considerations based on the concept of the uncocking motion and explain the system architecture with the sensors used for quantitative measurement and qualitative feedback generation. Then, we illustrate the detailed algorithms for wrist angle computation, golf swing motion segmentation based on key pose detection, and uncocking evaluation. A multimodal feedback-based user interface for our system is also presented. Experimental results show that the proposed system has the ability to accurately calculate the wrist angle in real time and also that it can be applied to a practical self-coaching system to improve the uncocking motion.