Deteksi Marka Jalan Untuk Pengendalian Otomatis Miniatur Mobil

Adiatmaja, Gagatsatya (2021) Deteksi Marka Jalan Untuk Pengendalian Otomatis Miniatur Mobil. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 07211640000027-Undergraduate_Thesis.pdf] Text
07211640000027-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 April 2023.

Download (5MB) | Request a copy

Abstract

Kendaraan sudah menjadi sesuatu yang tidak bisa dipisahkan dari kehidupan manusia. Faktor penting dalam berkendara adalah keamanan dan kenyamanan. Namun ternyata tingkat kecelakaan masih tinggi. Penyebabnya sebagian besar dikarenakan kesalahan manusia itu sendiri atau human error.
Salah satu solusi untuk mengurangi tingginya tingkat kecelakaan dan meningkatkan kenyamanan dalam berkendara adalah dengan membuat teknologi autonomous car atau mobil yang dikendalikan secara otomatis. Banyak riset dan penelitian mengenai teknologi tersebut tetapi belum terealisasikan dimana perlu dikembangkan metode untuk mengendalikan mobil secara otomatis. Sebagian besar penelitian yang dilakukan di Indonesia terkait autonomous car belum diuji cobakan secara real-time. Padahal uji coba secara real-time sangat perlu dilakukan untuk mengetahui apakah metode tersebut memang dapat diterapkan pada mobil.
Oleh sebab itu pada penelitian ini diusulkan deteksi marka jalan untuk pengendalian otomatis miniatur mobil. Penelitian ini menggunakan metode Residual Neural Network 18 (ResNet18) untuk mengolah data citra marka jalan. Data citra tersebut dilatih pada proses training untuk menghasilkan model yang dapat membedakan garis marka jalan dan badan jalan sehingga autonomous car dapat mempertahankan lajunya diantara dua garis marka tersebut secara otomatis. Metode tersebut diujicobakan pada minatur mobil sehingga diharapkan pada beberapa tahun kedepan mobil dapat terkendali otomatis sepenuhnya.
======================================================================================================
Vehicles have become something that can not be
separated from human life. Important factors in driving are safety and comfort. However, the accident rate is still high. The most common cause of traffic accidents is human error. One of the solutions to reduce accident rates and increase driving comfort is to make an autonomous car or a car which controlled automatically. There is a lot of research on this technology but it has not been realized where it is necessary to develop a method to control a car automatically. Most of the research conducted in Indonesia on autonomous cars has not been tested in real-time. Real-time testing is necessary to see if the method can be applied to cars. Therefore, this research proposes road detection for automatic prototype cars. This study uses the Residual Neural Network 18 (ResNet18) method to process road-marking image data. The image data is placed in training to produce a model that can distinguish road markings and road bodies so that autonomous cars can maintain their speed between the two markings automatically. This method was tested on a prototype car so it is hoped that in the next few years the car can be controlled automatically.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Jetson Nano, marka jalan, autonomous car, Resnet Jetson Nano, road mark, autonomous car, resnet
Subjects: T Technology > TA Engineering (General). Civil engineering (General) > TA1573 Detectors. Sensors
T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques
T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL152.8 Vehicles, Remotely piloted. Autonomous vehicles.
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Computer Engineering > 90243-(S1) Undergraduate Thesis
Depositing User: Gagatsatya Adiatmaja
Date Deposited: 12 Mar 2021 06:50
Last Modified: 12 Mar 2021 06:50
URI: http://repository.its.ac.id/id/eprint/84142

Actions (login required)

View Item View Item