Sistem Deteksi Anomali Berbasis Machine Learning Untuk Identifikasi Kegagalan UAV

Satria, Bagja (2025) Sistem Deteksi Anomali Berbasis Machine Learning Untuk Identifikasi Kegagalan UAV. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5022211211_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (8MB) | Request a copy

Abstract

Deteksi anomali berbasis machine learning telah menjadi solusi yang efektif dalam mengidentifikasi potensi kegagalan pada Unmanned Aerial Vehicle (UAV). Penelitian ini bertujuan untuk mengembangkan sistem deteksi anomali pada data operasional UAV dan merancang mekanisme respons otomatis untuk meningkatkan keandalan dan keselamatan UAV, terutama dalam kondisi darurat. Sistem ini memanfaatkan transformasi data dari domain waktu ke domain frekuensi menggunakan Fast Fourier Transform (FFT) untuk mengidentifikasi pola spektral yang relevan. Hasil analisis kemudian digunakan untuk melatih model Artificial Neural Network (ANN), yang dirancang untuk mendeteksi anomali secara real-time dengan akurasi tinggi. Penelitian ini juga mencakup integrasi sistem deteksi anomali dan mekanisme respons otomatis ke dalam platform UAV jenis fixed-wing. Hasil yang diharapkan meliputi peningkatan kemampuan deteksi dini terhadap malfungsi UAV, mitigasi risiko kerusakan, serta kontribusi dalam pengembangan teknologi deteksi anomali berbasis machine learning untuk aplikasi UAV. Penelitian ini juga memberikan kontribusi akademik melalui pendekatan inovatif yang mengintegrasikan machine learning dan avionik UAV.
======================================================================================================================================
Machine learning-based anomaly detection has become an effective solution in identifying potential failures in Unmanned Aerial Vehicles (UAVs). This study aims to develop an anomaly detection system on UAV operational data and design an automatic response mechanism to improve the reliability and safety of UAVs, especially in emergency conditions. This system utilizes data transformation from the time domain to the frequency domain using Fast Fourier Transform (FFT) to identify relevant spectral patterns. The analysis results are then used to train an Artificial Neural Network (ANN) model, designed to detect real-time anomalies with high accuracy. This study also includes the integration of an anomaly detection system and automatic response mechanism into a fixed-wing UAV platform. Expected outcomes include improving the ability to detect early UAV malfunctions, mitigating the risk of damage, and contributing to the development of machine learning-based anomaly detection technology for UAV applications. This study also provides academic contributions through an innovative approach that integrates machine learning and UAV avionics.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	UAV, deteksi anomali, CNN, FFT, IMU, Raspberry Pi, UAV, anomaly detection, CNN, FFT, IMU, Raspberry Pi
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > QA Mathematics > QA336 Artificial Intelligence U Military Science > U Military Science (General) > UG Military Engineering > UG1242.D7 Unmanned aerial vehicles. Drone aircraft
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20201-(S1) Undergraduate Thesis
Depositing User:	Bagja Satria
Date Deposited:	24 Jul 2025 04:17
Last Modified:	24 Jul 2025 04:17
URI:	http://repository.its.ac.id/id/eprint/121091

Actions (login required)

View Item