Lilik Anifah, Nurhayati, Puput Wanarti Rusimamto, Muhamad Syarifuddien Zuhrie, Haryanto
Much research has been conducted worldwide on the recognition and monitoring of punches in martial art sports during the training process. The performance of the punching movements can be accurately analyzed based on the collected data. The current study aims to classify the punches on a punching bag using K Means based on a data logging system. Its stages are hardware design, hardware implementation, hardware testing, learning, and testing. The FSR 402 sensor was used to measure the punching force. GY6500 MPU6500 was also utilized to identify and measure the reaction force of these punches. The data were collected utilizing K Means and were subsequently tested. The results revealed that the system exhibited good performance, proven by its accuracy of 93.6%, precision of 0.933, recall of 0.934, and F1 score of 0.934. Based on these results, it seems that the classification of right and left punch forces can be efficiently carried out. This system can help users analyze the punching bag training process, and thus improve their performance. © 2025, Dr D. Pylarinos. All rights reserved.
Department of Electrical Engineering, Faculty of Engineering, Universitas Negeri Surabaya, Indonesia; Department of Electrical Engineering, Faculty of Engineering, Universitas Trunojoyo Madura, Indonesia
