Setiawan, M. Hendy (2025) Implementasi Single Board Computer untuk Akuisisi Data dan Komputasi Gas Array Sensor pada Unmanned Surface Vehicle. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5019211076-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5019211076-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (7MB) | Request a copy |
Abstract
Kebocoran kilang minyak mencemari perairan dan melepaskan Volatile Organic Compounds (VOC) yang membahayakan kesehatan manusia serta merusak ekosistem. Monitoring otomatis diperlukan untuk mendeteksi kontaminasi tanpa risiko paparan langsung. Penelitian ini mengimplementasikan single board computer pada Unmanned Surface Vehicle (USV) untuk deteksi gas berbahaya dan pemetaan area kontaminasi di perairan. Sistem menggunakan sensor gas array (TGS 2600, TGS 2602, dan TGS 2610) yang terintegrasi dengan Raspberry Pi untuk pemrosesan data. Algoritma machine learning Random Forest mengklasifikasi lima jenis gas: alkohol, pertalite, dexlite, ammonia, dan toluene. Sistem navigasi menggunakan Pixhawk dengan GPS untuk pemetaan koordinat real-time. Pengujian dilakukan dalam tiga kondisi: dry run (indoor), semi indoor (kolam), dan outdoor (Danau 8 ITS). Hasil menunjukkan sistem berhasil mengidentifikasi kelima gas dengan tingkat kepercayaan tinggi dalam kondisi terkontrol. Namun, performa menurun seiring kompleksitas lingkungan dengan tingkat keberhasilan 80% (semi indoor) dan 60% (outdoor). Ammonia dan toluene gagal terdeteksi dalam kondisi terbuka karena reaktivitas kimia dan degradasi fotokimia. Penelitian membuktikan implementasi single board computer pada USV efektif untuk monitoring pencemaran udara di perairan dalam kondisi terkontrol, namun memerlukan pengembangan lebih lanjut untuk aplikasi outdoor yang stabil.
=======================================================================================================================================
Oil refinery leakages contaminate water bodies and release Volatile Organic Compounds (VOC) that endanger human health and damage ecosystems. Automatic monitoring is required to detect contamination without direct exposure risks. This research implements a single board computer on an Unmanned Surface Vehicle (USV) for hazardous gas detection and contamination area mapping in water bodies. The system utilizes a gas sensor array (TGS 2600, TGS 2602, and TGS 2610) integrated with Raspberry Pi for data processing. A Random Forest machine learning algorithm classifies five types of gases: alcohol, pertalite, dexlite, ammonia, and toluene. The navigation system employs Pixhawk with GPS for real-time coordinate mapping. Testing was conducted under three conditions: dry run (indoor), semi-indoor (pool), and outdoor (ITS Lake 8). Results demonstrate that the system successfully identified all five gases with high confidence levels under controlled conditions. However, performance degraded with environmental complexity, achieving success rates of 80% (semi-indoor) and 60% (outdoor). Ammonia and toluene failed to be detected in open conditions due to chemical reactivity and photochemical degradation. This research proves that implementing a single board computer on USV is effective for air pollution monitoring in water bodies under controlled conditions, but requires further development for stable outdoor applications.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Machine Learning, Pencemaran Udara, Single Board Computer, Sensor Gas Array, Unmanned Surface Vehicle, Air Pollution, Machine Learning, Single Board Computer, Gas Sensor Array, Unmanned Surface Vehicle |
| Subjects: | V Naval Science > VM Naval architecture. Shipbuilding. Marine engineering V Naval Science > VM Naval architecture. Shipbuilding. Marine engineering > VM365 Remote submersibles. Autonomous vehicles. |
| Divisions: | Faculty of Marine Technology (MARTECH) > Marine Engineering > 36202-(S1) Undergraduate Thesis |
| Depositing User: | M. Hendy Setiawan |
| Date Deposited: | 04 Aug 2025 06:41 |
| Last Modified: | 04 Aug 2025 06:41 |
| URI: | http://repository.its.ac.id/id/eprint/126968 |
Actions (login required)
 |
View Item |