IDENTIFIKASI JALAN SEBAGAI INPUT SISTEM KENDALI KEMUDI MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK (CNN) PADA AUTONOMOUS ELECTRIC VEHICLE

HERMAWAN, MARKUS and Suprapto, Bhakti Yudho (2021) IDENTIFIKASI JALAN SEBAGAI INPUT SISTEM KENDALI KEMUDI MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK (CNN) PADA AUTONOMOUS ELECTRIC VEHICLE. Undergraduate thesis, Sriwijaya University.

	Text RAMA_20201_03041281722033.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (9MB) | Request a copy
	Text RAMA_20201_03041281722033_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (5MB) | Request a copy
Preview	Text RAMA_20201_03041281722033_0011027503_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (2MB) | Preview
	Text RAMA_20201_03041281722033_0011027503_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (303kB) | Request a copy
	Text RAMA_20201_03041281722033_0011027503_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (377kB) | Request a copy
	Text RAMA_20201_03041281722033_0011027503_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
	Text RAMA_20201_03041281722033_0011027503_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (123kB) | Request a copy
	Text RAMA_20201_03041281722033_0011027503_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (126kB) | Request a copy
	Text RAMA_20201_03041281722033_0011027503_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy

Abstract

Developments in the field of computer vision and robotics continue to be encouraged, especially autonomous electric vehicles to overcome traffic accidents due to human error. Currently, the method used is not real-time. So, in this study, an autonomous electric vehicle to be able to follow a predetermined route by identifying the road using the Convolutional Neural Network (CNN) as an input to the steering control system. The training model applied in this test uses the Stochastic Gradient Descent 150 epoch optimizer because it has a smaller loss value of 0.6133 and has a higher accuracy rate of 0.7743 when compared to the training model using the Adam optimizer. Testing using the model is carried out in 2 tests, namely simulation testing and real-time testing. In simulation testing, from 15 trials conducted, the percentage of success was 93.333%, with only one class classification error and 100% success rate for testing data transmission from the system to the tool while in real-time testing, the autonomous electric vehicle managed to follow a predetermined route accurately. However, the autonomous electric vehicle has not succeeded in avoiding the object in front of it due to the lack of precise steering mechanics and also the lack of variation in training data from various conditions that may be passed by the autonomous electric vehicle.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Autonomous Electric Vehicle, CNN, Identifikasi Jalan
Subjects:	T Technology > T Technology (General) > T59.5 Automation T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7885-7895 Computer engineering. Computer hardware
Divisions:	03-Faculty of Engineering > 20201-Electrical Engineering (S1)
Depositing User:	Markus Hermawan
Date Deposited:	02 Aug 2021 06:30
Last Modified:	02 Aug 2021 06:30
URI:	http://repository.unsri.ac.id/id/eprint/51100

Actions (login required)

View Item