Desain Big Data Berbasis Artificial Intelligence Untuk Integrasi Dan Fusi Data AIS-SAR Dalam Maritime Situational Awareness

FIFI, FITRIAH (2026) Desain Big Data Berbasis Artificial Intelligence Untuk Integrasi Dan Fusi Data AIS-SAR Dalam Maritime Situational Awareness. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5018221090-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (18MB) | Request a copy

Abstract

Peningkatan aktivitas kapal di wilayah perairan memerlukan sistem pemantauan yang mampu mengintegrasikan data pelaporan kapal dan pengamatan satelit secara terstruktur. Penelitian ini merancang sistem Big Data berbasis Artificial Intelligence untuk integrasi dan fusi Automatic Identification System (AIS) dan Synthetic Aperture Radar (SAR) dalam mendukung Maritime Situational Awareness. Jalur AIS mencakup pemeriksaan atribut, pembentukan lintasan, estimasi posisi menggunakan Kalman Filter, pengujian parameter, dan penerapan Pipeline Analitik AIS berbasis FINAL_4_PIPELINE_MODELS.h5. Jalur SAR mengaitkan target dan metadata scene dengan keluaran YOLOv12 berupa kelas kapal, confidence score, dan bounding box. Kedua jalur diselaraskan berdasarkan ruang dan waktu, kemudian digabungkan melalui proses matching, integrasi, dan fusi. Implementasi mengolah 584 baris data AIS-SAR, 474 MMSI, dan 316 MMSI kandidat. Sistem menghasilkan 365 kandidat berbasis aturan, yaitu 304 spoofing, 41 go dark, dan 20 transshipment. Sebanyak dua kandidat dapat dihubungkan dengan keluaran model H5, sedangkan 363 kandidat berstatus not_available karena rangkaian AIS belum memenuhi kebutuhan 120 titik per MMSI atau 24 langkah per pasangan kapal. Pengujian tiga konfigurasi Kalman Filter menunjukkan RMSE terendah 192,04 km pada konfigurasi C, tetapi selisih terhadap konfigurasi dasar hanya 0,162%, sehingga parameter dasar tetap dipertahankan. Hasil akhir disajikan melalui dashboard dan dapat diteruskan melalui Telegram alert. Seluruh kandidat bersifat indikatif dan memerlukan verifikasi lanjutan.
Kata Kunci—AIS, Artificial Intelligence. ====================================================================================================================================
The increasing level of vessel activity in maritime areas requires a monitoring system capable of integrating vessel reporting data and satellite observations in a structured manner. This study designs an Artificial Intelligence-based Big Data system for integrating and fusing Automatic Identification System (AIS) and Synthetic Aperture Radar (SAR) data to support Maritime Situational Awareness. The system processes the data through AIS attribute inspection, vessel trajectory construction, position estimation using a Kalman Filter, SAR target preparation, spatial and temporal alignment, matching, and the integration of outputs from the AIS Analytical Pipeline based on FINAL_4_PIPELINE_MODELS.h5 and the YOLOv12-based SAR Vessel Detection Pipeline. The Kalman Filter generates estimated positions and residuals as supporting information without replacing the original AIS positions. Vessel activity candidates are generated based on AIS–SAR distance, Kalman–SAR distance, time difference, Kalman residuals, AIS update gaps, speed, course, and inter-vessel proximity. The implementation processed 584 AIS–SAR data records and produced 365 candidates, consisting of 304 spoofing candidates, 41 go dark candidates, and 20 transshipment candidates. Two candidates were linked to H5 model outputs, while 363 candidates were assigned not_available status because their AIS sequences did not satisfy the model input requirements. YOLOv12 outputs, including vessel class, confidence score, and bounding box, were linked to SAR metadata. The results are presented through a dashboard and can be forwarded via Telegram alerts. The system provides structured and traceable monitoring information, although all candidates and alerts remain indicative and require further verification.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	AIS, SAR, Big Data, Artificial Intelligence, Kalman Filter, data integration, Maritime Situational Awareness.
Subjects:	T Technology > T Technology (General)
Divisions:	Faculty of Marine Technology (MARTECH) > Naval Architecture and Shipbuilding Engineering > 36201-(S1) Undergraduate Thesis
Depositing User:	Fifi Fitriah
Date Deposited:	30 Jul 2026 01:10
Last Modified:	30 Jul 2026 01:10
URI:	http://repository.its.ac.id/id/eprint/140280

Actions (login required)

View Item