Claim Missing Document
Check
Articles

Found 1 Documents
Search

ROBOT PENDETEKSI GAS BERACUN DENGAN MENGGUNAKAN METODE MULTI LAYER PERCEPTRON Gun Gun Maulana; Aris Budiyarto; Vera Suryani
Ensiklopedia of Journal Vol 2, No 2 (2020): Vol 2 No 2 Edisi 2 Januari 2020
Publisher : Lembaga Penelitian dan Penerbitan Hasil Penelitian Ensiklopedia

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (224.717 KB) | DOI: 10.33559/eoj.v2i2.411

Abstract

At present, there are a lot of harmful gases in the environment around us, especially in the industrial environment, such as radiological materials and toxic gases that can pollute the environment. Generally, gases can spread due to leakage. The gas leak was caused by several factors, such as mining accidents, war, and manufacturing accidents. Gas leakage incidents can cause huge losses to society and industry, both injury and financial problems. To reduce this risk we try to propose a toxic gas detection system with two modules used in running this system: (1) navigation module and (2) gas detection module. The navigation module is designed with a program to explore a room with a certain pattern. For movement and data collection will be carried out by the Microsoft Visual Studio interface software. The detection module is designed with Multi Layer Perceptron (MLP) artificial intelligence using the Back Propagation method which is directly applied to the robot, so that the robot is able to accurately and quickly identify gas obtained from the TGS sensor output. The sensors used are TGS 2602 for ammonia (NH3), TGS 2600 for carbon monoxide (CO), TGS 2610 for butane (C4H10), and TGS 2612 for methane (CH4). For this reason, this research is expected to implement electronic nose on mobile robots using the Multi Layer Perceptron method to effectively detect, identify, and confirm that dangerous gas sources have been found without direct human interaction