Self-powered intelligent badminton racket for machine learning-enhanced real-time training monitoring

Abstract

Intelligent sensing technology exerts a crucial role in badminton training: by capturing the behavior of athletes, the technology can effectively promote the enhancement of motor skills and performance. However, sports sensors that are multifunctional, real-time, and convenient remain an ongoing challenge. This study designs a self-powered intelligent badminton racket (SIBR) with machine learning-based triboelectric/piezoelectric effects. The silver paste coating method is employed for constructing customized electrodes, thereby forming triboelectric sensing array on the badminton strings, which enables hitting position monitoring. Meanwhile, flexible piezoelectric films with a specific shape are embedded in the hand glue; thus, the grip posture is identified. These sensing arrays can directly convert mechanical signals into electrical signals for achieving zero power consumption. In addition, the study integrates a wireless module for signal acquisition and transmission at the bottom of the racket handle, which ensures real-time sensor monitoring based on normal usage. The collected multi-channel data obtained from the SIBR is utilized for machine learning, achieving an accuracy of hitting position that can reach 95.0%. SIBR provides a powerful reference for badminton training and unfolds a new path and direction for badminton sports monitoring.