Pengembangan Simulasi Evakuasi Gempa Bumi Berbasis Augmented Reality Menggunakan YOLO11 dengan Fitur Wayfinding Multi-Lantai

Rachmawati, Putri Meyliya (2026) Pengembangan Simulasi Evakuasi Gempa Bumi Berbasis Augmented Reality Menggunakan YOLO11 dengan Fitur Wayfinding Multi-Lantai. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5025221062-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (10MB) | Request a copy

Abstract

Gempa bumi merupakan bencana alam yang berpotensi menimbulkan korban jiwa, baik akibat dampak langsung maupun bahaya sekunder seperti benda yang roboh atau pecah. Edukasi prosedur berlindung di bawah meja dan evakuasi, masih banyak disampaikan melalui metode konvensional yang kurang interaktif. Penelitian ini mengembangkan aplikasi simulasi gempa bumi berbasis Android yang memanfaatkan teknologi Augmented Reality (AR) untuk mendeteksi objek bahaya sekunder serta memandu proses evakuasi multi-lantai tanpa bergantung pada sistem GPS. Aplikasi dibangun menggunakan Unity 6 dengan AR Foundation dan ARCore, serta model deteksi objek YOLO11 yang dilatih untuk mengenali empat kelas objek, yaitu Bookshelf, Glass, Table, dan Window. Model dikonversi ke format ONNX dan dijalankan melalui Unity Inference Engine untuk inferensi secara real-time. Fitur wayfinding multi-lantai diimplementasikan menggunakan QR Code sebagai titik referensi spasial dan AR Anchor untuk menjaga stabilitas posisi penanda evakuasi. Pengujian dilakukan melalui tiga metode untuk mengevaluasi fungsionalitas, efektivitas edukasi, dan kegunaan aplikasi. Pengujian Black Box terhadap 30 skenario uji menghasilkan tingkat keberhasilan rata-rata 97,75%. Evaluasi efektivitas edukasi menggunakan metode Normalized Gain (N-Gain) menghasilkan rata-rata 0,48 yang termasuk kategori Sedang. Pengujian kegunaan menggunakan System Usability Scale (SUS) menghasilkan rata-rata skor 81,17 yang termasuk kategori Grade A dengan rentang persentil 90-95. Hasil tersebut menunjukkan bahwa aplikasi yang dikembangkan berhasil mencapai tujuan penelitian sebagai media simulasi dan edukasi tanggap gempa bumi yang fungsional, efektif, dan mudah digunakan.
==========================================================================================================================================
Earthquakes are natural disasters that can cause casualties due to both direct impacts and secondary hazards, such as falling or shattered objects. Education regarding sheltering under tables and evacuation procedures is still commonly delivered through conventional methods that are less interactive. This study develops an Android-based earthquake simulation application that utilizes Augmented Reality (AR) technology to detect secondary hazard objects and guide users through multi-floor evacuation without relying on a Global Positioning System (GPS). The application was developed using Unity 6 with AR Foundation and ARCore, and employs the YOLO11 object detection model trained to recognize four object classes: Bookshelf, Glass, Table, and Window. The model was converted into ONNX format and executed through the Unity Inference Engine for real-time inference. The multi-floor wayfinding feature was implemented using QR Codes as spatial reference points and AR Anchors to maintain the stability of evacuation markers. The application was evaluated through three methods to assess its functionality, educational effectiveness, and usability. Black-box testing across thirty test scenarios achieved an average success rate of 97.75%. Educational effectiveness was evaluated using the Normalized Gain (N-Gain) method, resulting in an average score of 0.48, which falls into the Medium category. Usability testing using the System Usability Scale (SUS) yielded an average score of 81.17, corresponding to Grade A with a percentile rank of 90-95. These results indicate that the developed application successfully fulfills its objectives as a functional, effective, and user-friendly earthquake preparedness simulation and educational tool.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Android, Augmented Reality, Deteksi Bahaya Sekunder, Simulasi Gempa Bumi, Wayfinding Multi-Lantai, Android, Augmented Reality, Earthquake Simulation, Multi-Floor Wayfinding, Secondary Hazard Detection
Subjects:	Q Science > QA Mathematics > QA76.9 Computer algorithms. Virtual Reality. Computer simulation.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Putri Meyliya Rachmawati
Date Deposited:	23 Jul 2026 15:19
Last Modified:	23 Jul 2026 15:19
URI:	http://repository.its.ac.id/id/eprint/137161

Actions (login required)

View Item