Deep Neural Network Untuk Visual Lokalisasi Mobil Otonom Di Lingkungan Kampus Its

Prajna, Arnold (2022) Deep Neural Network Untuk Visual Lokalisasi Mobil Otonom Di Lingkungan Kampus Its. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 07111840000104-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (4MB)

Abstract

Intelligent Car (I-Car) ITS merupakan prototipe mobil otonom dimana salah satu metode lokalisasinya yang utama didapatkan melalui pembacaan data GPS, namun keakuratan pembacaan GPS dipengaruhi oleh tersedianya informasi dari satelit-satelit GPS, dimana informasi tersebutseringkali tergantung oleh kondisi tempat pada saat itu, seperti cuaca, bentuk, dan keramaian (dense) dari suatu dataran sehingga terkadang informasi lokasi tersebut tidak tersedia secara akurat. Melalui latar belakang permasalahan ini, solusi yang ditawarkan untuk mengatasi ketidaktersedianya informasi GPS yang akurat tersebut adalah lokalisasi I-Car ITS berdasarkan data visual kamera omnidireksional melalui rekognisi lingkungan sekitar kampus ITS menggunakan Deep Neural Network. Proses dari rekognisi adalah dengan mengambil data koordinat GPS untuk dijadikan sebagai titik acuan output pada saat kamera omnidireksional mengambil citra-citra lingkungan sekitar. Uji coba visual lokalisasi dilakukan pada lingkungan ITS dengan total 200 koordinat GPS, dimana setiap koordinat GPS mewakili satu kelas sehingga terdapat 200 kelas untuk klasifikasi. Setiap koordinat/kelas memiliki citra training sebanyak 96 citra. Kondisi ini dicapai untuk kecepatan kendaraan sebesar 20 km/jam, dengan kecepatan akuisisi citra dari omnidireksional kamera sebesar 30 fps. Hasil akurasi lokalisasi visual diperoleh rata-rata sebesar 45-50% dengan arsitektur CNN - AlexNet. Hasil pengujian tersebut diperoleh dengan penggunaan parameter learning rate sebesar 0.00001, augmentasi data, dan teknik DropOut untuk mencegah terjadinya Overfitting dan meningkatkan kestabilan akurasi.
==============================================================================================================================
Intelligent Car (I-Car) ITS is an autonomous car prototype where one of the main localization methods is obtained through reading GPS data, but the accuracy of GPS readings is influenced by the availability of information from GPS satellites, where the information often depends on the conditions of the place at that time, such as weather, shape, and density of a land so that sometimes the location information is not available accurately. Against this background, the solution offered to overcome the unavailability of accurate GPS information is ITS I-Car localization based on omnidirectional camera visual data through environmental recognition around the ITS campus using Deep Neural Network. The process of recognition is to take GPS coordinate data to be used as an output reference point when the omnidirectional camera takes images of the surrounding environment. Visual localization trials were carried out in the ITS environment with a total of 200 GPS coordinates, where each GPS coordinate represents one class so that there are 200 classes for classification. Each coordinate/class has 96 training images. This condition is achieved for a vehicle speed of 20 km/h, with an image acquisition speed of 30 fps from the omnidirectional camera. The result of visual localization accuracy is 45-50% with CNN - AlexNet architecture. The test results were obtained by using a learning rate parameter of 0.00001, data augmentation, and the Drop Out technique to prevent overfitting and improve accuracy stability.

Item Type:	Thesis (Other)
Additional Information:	RSE 629.892 Pra d-1 2022
Uncontrolled Keywords:	Mobil Otonom. Visual Lokalisasi. Autonomous Car. Visual Localization.
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20201-(S1) Undergraduate Thesis
Depositing User:	Mr. Marsudiyana -
Date Deposited:	12 Jun 2026 06:41
Last Modified:	12 Jun 2026 06:41
URI:	http://repository.its.ac.id/id/eprint/133772

Actions (login required)

View Item