Wearable Device Untuk Klasifikasi Gerakan Pukulan Pada Olahraga Bulutangkis Menggunakan TinyML

Harizt, Luthfi Fajri (2024) Wearable Device Untuk Klasifikasi Gerakan Pukulan Pada Olahraga Bulutangkis Menggunakan TinyML. Diploma thesis, Universitas Andalas.

	Text (Cover & Abstrak) Cover dan Abstrak.pdf - Published Version Download (166kB)
	Text (BAB I) BAB I Pendahuluan.pdf - Published Version Download (130kB)
	Text (BAB V) BAB V Penutup.pdf - Published Version Download (45kB)
	Text (DAFTAR PUSTAKA) DAFTAR PUSTAKA.pdf - Published Version Download (113kB)
	Text (Skripsi Full Text) Draft Full_Harizt Luthfi Fajri.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

In badminton training and analysis, it's essential to assess a player's movements to help coaches analyze their playing patterns. Traditionally, this process involves manual counting or video recordings, which can be time-consuming and less precise. The aim of this research is to streamline this analysis process by utilizing Tiny Machine Learning on a microcontroller system, eliminating the need for internet connectivity and reducing costs while speeding up data processing. The designed system comprises a wearable device using Wemos Lolin32 microcontroller and an MPU6050 sensor to capture motion signals as input for processing. The trained model achieves a training accuracy of 94,5%. Additionally, the system can recognize user movements and clasify stroke types with an 80% with 91,2 % F-1 Score and 96,2 % accuracy using the trained model.

Actions (login required)

View Item