DEVELOPMENT OF A REAL-TIME PLASTIC WASTE DETECTION SYSTEM ON REMOTELY OPERATED UNDERWATER VEHICLE USING YOLO-BASED REMOTE INFERENCE

QOSHIROTU THORFI GIBRAN YUSUF, . (2026) DEVELOPMENT OF A REAL-TIME PLASTIC WASTE DETECTION SYSTEM ON REMOTELY OPERATED UNDERWATER VEHICLE USING YOLO-BASED REMOTE INFERENCE. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.

	Text Qoshirotu Thorfi Gibran Yusuf_Skripsi_compressed-COVER.pdf Download (1MB)
	Text Qoshirotu Thorfi Gibran Yusuf_Skripsi_compressed-BAB 1.pdf Download (487kB)
	Text Qoshirotu Thorfi Gibran Yusuf_Skripsi_compressed-BAB 2.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text Qoshirotu Thorfi Gibran Yusuf_Skripsi_compressed-BAB 3.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text Qoshirotu Thorfi Gibran Yusuf_Skripsi_compressed-BAB 4.pdf Restricted to Registered users only Download (3MB) | Request a copy
	Text Qoshirotu Thorfi Gibran Yusuf_Skripsi_compressed-BAB 5.pdf Restricted to Registered users only Download (547kB) | Request a copy
	Text Qoshirotu Thorfi Gibran Yusuf_Skripsi_compressed-DAFTAR PUSTAKA.pdf Download (548kB)
	Text Qoshirotu Thorfi Gibran Yusuf_Skripsi_compressed-LAMPIRAN.pdf Restricted to Registered users only Download (588kB) | Request a copy

Abstract

QOSHIROTU THORFI GIBRAN YUSUF. Development of a Real-TimePlastic Waste Detection System on Remotely Operated Underwater Vehicle Using YOLO-Based Remote Inference, Undergraduate Thesis, Physics Study Program, Faculty of Mathematics and Natural Sciences, Universitas Negeri Jakarta, June 2026. Plastic waste pollution in underwater environments has become a majorenvironmental issue that requires effective monitoring and detection systems. Remotely Operated Underwater Vehicles (ROUVs) integrated with computer vision technology offer a promising solution for underwater monitoring applications. However, implementing real-time object detection directly on embedded devices such as Raspberry Pi 4 presents computational limitations, especially when handling simultaneous vehicle control and live YOLO inference. This research proposes a distributed ROUV system architecture for real-time underwater plastic waste detection and remote vehicle operation using the YOLO11n model in ONNX format. In the proposed system, Raspberry Pi 4 functions as the onboard controller for video acquisition and communication handling, while the main backend server and YOLO inference are executed on a laptop to reduce computational load. Communication between the system components is implemented using WebSocket and UART protocols, alongside a web-based dashboard for live monitoring and remote control. Experimental results show that the system achieves an average YOLO processing time of 54 ms, total latency of 92.8 ms, and average FPS of 18.7 FPS. The results indicate that the proposed architecture successfully improves real-time performance and enables stable live underwater plastic waste detection with responsive remote vehicle control.

Item Type:	Thesis (Sarjana)
Additional Information:	1). Dr.Mutia Delina, M.Si. ; 2). Riser Fahdiran, M.Si.
Subjects:	Sains > Fisika
Divisions:	FMIPA > S1 Fisika
Depositing User:	Qoshirotu Thorfi Gibran Yusuf .
Date Deposited:	12 Aug 2026 03:46
Last Modified:	12 Aug 2026 03:46
URI:	http://repository.unj.ac.id/id/eprint/70120

Actions (login required)

View Item