Napitupulu, Ezekiel Walfred Ebenezer Pangihutan (2025) Autonomous Drone untuk Deteksi dan Tracking Objek Menggunakan YOLOv8. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5024211038-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5024211038-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (13MB) | Request a copy |
Abstract
Penelitian ini mengusulkan pengembangan sistem autonomous drone berbasis pemrosesan citra digital yang dapat mendeteksi serta mengikuti target menggunakan model deteksi objek YOLOv8. Sistem ini dirancang sebagai solusi untuk meningkatkan efisiensi dan keberlanjutan dalam proses pengiriman paket, khususnya pada tahapan last-mile delivery. Drone yang digunakan adalah DJI Ryze Tello, yang dilengkapi dengan kamera onboard dan dikendalikan menggunakan pustaka Python djitellopy. Proses deteksi objek dilakukan secara real-time, dan sistem dirancang untuk mampu membedakan arah target (kiri, kanan, atau belakang) menggunakan klasifikasi visual. Dataset diperoleh melalui akuisisi video dengan sudut pandang bervariasi, kemudian dilabeli dan dilatih menggunakan YOLOv8. Pengujian dilakukan dengan variasi jarak untuk menilai akurasi dan tingkat keyakinan (confidence) model dalam klasifikasi objek "Car Toy Left" dan "Car Toy Right". Hasil menunjukkan bahwa sistem mampu mendeteksi dan mengikuti target dengan akurasi 100% pada jarak 2-3 meter, dan menurun menjadi 90% untuk "Car Toy Right" dan 85% untuk "Car Toy Left" pada jarak 4 meter. Pengujian kontrol pergerakan drone menunjukkan bahwa kontrol umpan balik memberikan respons yang lebih stabil namun lebih lambat dibandingkan dengan kontrol tanpa umpan balik. Penelitian ini memberikan kontribusi terhadap pengembangan sistem drone yang adaptif, efisien, dan mendukung otomatisasi logistik yang ramah lingkungan.
========================================================================================================================
This research proposes the development of an autonomous drone system based on digital image processing, capable of detecting and tracking targets using the YOLOv8 object detection model. The system is designed to enhance the efficiency and sustainability of package delivery, particularly in last-mile logistics. The DJI Ryze Tello drone is employed, equipped with an onboard camera and controlled via Python using the djitellopy library. Real-time object detection is implemented, and the system is capable of classifying the target's orientation (left, right, or behind) through visual classification. The dataset was acquired using multi-angle video capture, labeled accordingly, and trained using YOLOv8. Evaluation was conducted under varying distances to assess the model's classification accuracy, drone response time, and confidence for "Car Toy Left" and "Car Toy Right" labels. Results show that the system successfully detects and tracks targets with 100% accuracy at distances of 2-3 meters, decreasing to 90% for "Car Toy Right" and 85% for "Car Toy Left" at 4 meters. Testing of the drone's movement control revealed that feedback control provides more stable but slower responses compared to non-feedback control. This research contributes to the development of an adaptive, efficient, and environmentally friendly drone-based logistics automation system.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Autonomous Drone, YOLOv8, Pemrosesan Citra Digital, Object Tracking, Digital Image Processing. |
| Subjects: | Q Science > QA Mathematics > QA336 Artificial Intelligence Q Science > QA Mathematics > QA76.6 Computer programming. Q Science > QA Mathematics > QA76.9.U83 Graphical user interfaces. User interfaces (Computer systems)--Design. T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL776 .N67 Quadrotor helicopters--Automatic control U Military Science > UG1242 Drone aircraft--Control systems. (unmanned vehicle) |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Computer Engineering > 90243-(S1) Undergraduate Thesis |
| Depositing User: | Ezekiel Walfred Ebenezer Pangihutan Napitupulu |
| Date Deposited: | 23 Jun 2025 02:27 |
| Last Modified: | 23 Jun 2025 02:27 |
| URI: | http://repository.its.ac.id/id/eprint/119180 |
Actions (login required)
 |
View Item |