KENDALI AC BERBASIS GESTURE TANGAN MENGGUNAKAN LSTM PADA PERANGKAT EDGE

Alhabsyi, Ali Akbar (2026) KENDALI AC BERBASIS GESTURE TANGAN MENGGUNAKAN LSTM PADA PERANGKAT EDGE. Other thesis, Institut Teknologi Sepuluh nopember.

Text 5024221005-Undergraduate_Thesis.pdf Download (22MB)

Abstract

Perkembangan teknologi smart home mendorong hadirnya inovasi dalam metode interaksi antara manusia dan perangkat elektronik. Air Conditioner (AC) merupakan salah satu perangkat rumah tangga yang penting, namun hingga kini pengendaliannya masih didominasi oleh remote infrared (IR) konvensional. Metode ini memiliki berbagai keterbatasan, seperti ketergantungan pada line of sight, risiko kerusakan atau kehilangan remote, serta kurang fleksibel untuk lingkungan rumah pintar modern. Penelitian ini mengusulkan perancangan sistem kendali AC berbasis gesture tangan dengan memanfaatkan model Long Short-Term Memory (LSTM). Kamera digunakan untuk menangkap gerakan tangan, kemudian diproses dengan MediaPipe untuk mengekstraksi koordinat landmark. Data sekuensial hasil ekstraksi diproses menggunakan model LSTM yang dijalankan pada Raspberry Pi sebagai perangkat edge computing, sehingga sistem dapat bekerja secara mandiri tanpa memerlukan server eksternal. Hasil klasifikasi gesture diteruskan ke ESP32 yang dilengkapi dengan modul infrared transmitter untuk mengirimkan perintah kendali ke AC. Dengan pendekatan ini, pengguna dapat menyalakan, mematikan, dan mengatur AC hanya dengan menggunakan gesture tangan. Sistem yang dirancang diharapkan mampu bekerja secara real-time dengan akurasi tinggi, serta memberikan pengalaman interaksi yang lebih alami, praktis, dan mendukung implementasi smart home.
====================================================================================================================================
The development of smart home technology has driven innovation in human-device interaction methods. Air conditioners (AC) are among the most essential household appliances, yet their control is still dominated by conventional infrared (IR) remotes. This approach has several limitations, such as dependence on line of sight, risk of damage or loss of the remote, and limited flexibility for modern smart home environments. This study proposes the design of an AC control system based on hand gesture recognition using a Long Short-Term Memory (LSTM) model. A camera is used to capture hand movements, which are then processed using MediaPipe to extract landmark coordinates. The resulting sequential data are processed by an LSTM model running on a Raspberry Pi as an edge computing device, enabling the system to operate independently without an external server. The recognized gestures are then forwarded to an ESP32 equipped with an infrared transmitter to send control commands to the AC. With this approach, users can turn the AC on, turn it off, and adjust its settings simply by using hand gestures. The proposed system is expected to operate in real time with high accuracy while providing a more natural, practical, and flexible interaction experience to support smart home implementation.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	gesture tangan, LSTM, Raspberry Pi, ESP32, infrared, edge computing, smart home, kendali AC. gesture tangan, LSTM, Raspberry Pi, ESP32, infrared, edge computing, smart home, kendali AC.
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > QA Mathematics > QA76 Computer software Q Science > QA Mathematics > QA76.6 Computer programming. Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) Q Science > QC Physics > QC457 Infrared technology. T Technology > T Technology (General) T Technology > T Technology (General) > T57.5 Data Processing T Technology > TK Electrical engineering. Electronics Nuclear engineering T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5103.2 Wireless communication systems. Two way wireless communication T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7878 Electronic instruments
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Computer Engineering > 90243-(S1) Undergraduate Thesis
Depositing User:	Ali Akbar Alhabsyi
Date Deposited:	09 Jul 2026 07:06
Last Modified:	09 Jul 2026 07:06
URI:	http://repository.its.ac.id/id/eprint/134459

Actions (login required)

View Item