Micshael, Zeens Imanuel Faisan (2026) Implementasi Smart Trash Bin Menggunakan Esp32-Cam Dan Arduino Uno Untuk Deteksi Jenis Sampah Berbasis Iot. S1 Teknik Informatika thesis, STMIK Widya Cipta Dharma.
![[img]](https://repository.wicida.ac.id/style/images/fileicons/text.png) |
Text
2243105-S1- Jurnal.pdf Download (729kB) |
|
Text
2243105-S1-Teknik Informatika.pdf Restricted to Repository staff only Download (2MB) | Request a copy |
Abstract
Integrasi teknologi otomatisasi pada sistem manajemen limbah menjadi aspek krusial dalam meningkatkan efisiensi proses pemilahan sampah secara presisi. Penelitian ini bertujuan untuk merancang dan mengimplementasikan purwarupa Smart Trash Bin berbasis Internet of Things (IoT) yang terintegrasi dengan mekanisme konveyor sabuk untuk mengklasifikasikan limbah menjadi kategori logam, plastik, dan organik secara otomatis, di mana seluruh proses penelitian dan pengujian sistem dilaksanakan di lingkungan kampus STMIK Widya Cipta Dharma . Metode pengembangan sistem yang digunakan adalah paradigma prototyping yang meliputi tahapan communication, quick plan, modeling quick design, construction of prototype, serta deployment, delivery, dan feedback . Arsitektur sistem menerapkan pendekatan komputasi terdistribusi guna mengoptimalisasi pemrosesan data dan mencegah terjadinya hambatan aliran informasi (bottleneck). Dalam skema ini, Arduino Uno difungsikan sebagai pengendali utama aktuator mekanis dan pembaca sensor fisik (induktif dan ultrasonik). Modul ESP32-CAM didedikasikan untuk penangkapan citra visual, sementara ESP32 Wrover berperan sebagai pusat komunikasi data menuju platform Blynk. Pemrosesan identifikasi sampah organik dilakukan menggunakan model kecerdasan buatan yang dilatih melalui platform web Teachable Machine. Hasil pengujian sistem menunjukkan bahwa purwarupa mampu beroperasi secara stabil dan akurat. Pada tahap filtrasi awal, sensor fisik mencatatkan tingkat keberhasilan sebesar 99% dalam mendeteksi material logam dan botol plastik. Implementasi kecerdasan buatan berhasil memverifikasi sampah organik dengan tingkat kepercayaan antara 90% hingga 98% melalui mekanisme penguncian data (locking) selama 3 detik. Algoritma keamanan batas waktu (timeout) 15 detik terbukti efektif dalam mengambil alih instruksi untuk mengarahkan objek yang tidak teridentifikasi menuju wadah plastik sebagai prosedur keamanan sistem. Selain itu, fitur pemantauan berbasis IoT sukses mentransmisikan data kapasitas dan notifikasi ke aplikasi Blynk secara real-time dengan tingkat keberhasilan mutlak 100%. ============================================================ The integration of automation technology into waste management systems has become a crucial aspect of enhancing the efficiency of precision waste sorting processes. This research aims to design and implement an Internet of Things (IoT)-based Smart Trash Bin prototype integrated with a belt conveyor mechanism to automatically classify waste into metal, plastic, and organic categories, where all research and system testing processes were conducted at the STMIK Widya Cipta Dharma campus environment. The software development method employed is the prototyping paradigm, encompassing the stages of communication, quick plan, modeling quick design, construction of prototype, as well as deployment, delivery, and feedback. The system architecture utilizes a distributed computing approach to optimize data processing and prevent potential information flow bottlenecks. In this framework, Arduino Uno functions as the primary mechanical actuator controller and physical sensor reader (inductive and ultrasonic), the ESP32-CAM module is dedicated to visual image capture, while the ESP32 Wrover acts as the data communication center for the Blynk platform. Organic waste identification is processed using an artificial intelligence model trained through the Teachable Machine web platform. System testing results indicate that the prototype operates stably and accurately. In the initial filtration stage, physical sensors achieved a 99% success rate in detecting metal and plastic bottles. The artificial intelligence implementation successfully verified organic waste with a confidence level ranging from 90% to 98% through a 3-second data locking mechanism. The 15-second timeout security algorithm proved highly effective in executing fail-safe instructions to direct unidentified objects to the plastic container as a system security procedure. Furthermore, the IoT-based monitoring feature successfully transmitted real-time capacity updates and notifications to the Blynk application with a 100% success rate.
| Item Type: | Thesis (S1 Teknik Informatika) |
|---|---|
| Additional Information: | Pembimbing 1 : Ahmad Fajri, S.Kom., M.Kom Pembimbing 2 : Presa Taruna Oliver, S.Kom., M.T. |
| Uncontrolled Keywords: | Smart Trash Bin, Internet of Things, Arduino Uno, ESP32-CAM, Konveyor, Komputasi Terdistribusi |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science |
| Divisions: | Teknik Informatika |
| Depositing User: | Mr Micshael Zeens Imanuel Faisan |
| Date Deposited: | 21 May 2026 03:08 |
| Last Modified: | 21 May 2026 03:08 |
| URI: | http://repository.wicida.ac.id/id/eprint/6311 |
Actions (login required)
 |
View Item |
