Integrated Eye Fatigue Recognition And Engine Stop Control System (EFREST) Pada Haul Dump Truck Pertambangan

Wandana, Akbar Krisna (2025) Integrated Eye Fatigue Recognition And Engine Stop Control System (EFREST) Pada Haul Dump Truck Pertambangan. Diploma thesis, Institut Teknologi Sepuluh Nopember.

Text Akbar Krisna Wandana_2042211073-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (13MB) | Request a copy

Abstract

Kecelakaan dump truck adalah kecelakaan yang rawan terjadi di lokasi penambangan, seperti tabrakan yang disebabkan oleh pengemudi yang tidak dapat melihat area blind spot dan/atau tertidur. Hal-hal seperti itu dapat dikurangi dengan membuat sistem keamanan yang cerdas. Setelah menganalisis masalah, meninjau desain prototipe, dan merekomendasikan spesifikasi komponen peralatan, dibuat sistem yang mengintegrasikan eye fatigue recognition system untuk mendeteksi pengemudi mengantuk dan engine stop control system yang dapat menghentikan laju truk saat terdeteksi jarak tidak aman. Prototipe sistem ini menggunakan 1 web cam yang diletakkan pada bagian depan atas cabin truk dengan 2 sensor ultrasonik HC-SR04 di depan dan belakang truk untuk mengatasi masalah blind spot truk. Kamera akan mendeteksi bukaan mata menggunakan Eye Aspect Ratio (EAR) dengan nilai minimal 0.2 sebagai indikasi pengemudi tertidur, yang kemudian fitur engine stop control pada truk akan bekerja, sehingga truk berhenti seketika. Sistem pendeteksian menggunakan machine learning berupa image processing dengan kontroller berupa Raspberry Pi 4. Sedangkan sensor ultrasonik HC-SR04 akan mendeteksi benda terlalu dekat (<6 cm), serta tebing atau jurang di depan dan belakang truk (>40 cm) maka dump truck juga akan berhenti. Dari pengujian sistem didapatkan bahwa eye fatigue recognition system memiliki nilai akurasi sebesar 94% dan respon aktuator berdasarkan input nilai EAR adalah 100% sesuai. Selain itu pada pendeteksi jarak aman menggunakan sensor ultrasonik 1 (bagian belakang truk) memiliki tingkat akurasi 98.3 %, dan sensor ultrasonik 2 (bagian depan truk) memiliki tingkat akurasi 98.8% dan respon aktuator adalah 100% sesuai. Disimpulkan bahwa rancangan alat memiliki 3 input data yaitu 1 web cam dan 2 ultrasonik yang terletak di depan dan belakang truk dengan aktuator berupa motor DC. Teknologi ini memiliki akurasi kinerja keseluruhan yakni 100% berdasarkan parameter pengujian terintegrasi yang telah dilakukan.
==================================================================================================================================
Dump truck accidents are accidents that are prone to occur at mining sites, such as collisions caused by drivers who cannot see blind spot areas and/or fall asleep. Such things can be reduced by creating an intelligent safety system. After analyzing the problem, reviewing the prototype design, and recommending equipment component specifications, a system was created that integrates an eye fatigue recognition system to detect driver fatigue while driving when the truck is in danger of an accident. This system prototype uses 1 web cam placed at the front top of the truck cabin with 2 HC-SR04 ultrasonic sensors at the front and back of the truck to overcome the truck blind spot problem. The camera will detect eye opening using Eye Aspect Ratio (EAR) with a minimum value of 0.2 as an indication that the driver is asleep, then the engine stop control feature on the truck will work, so that the truck stops instantly. The detection system uses machine learning in the form of image processing with a controller in the form of a Raspberry Pi 4. While the HC-SR04 ultrasonic sensor will detect objects too close (<6 cm), as well as cliffs or ravines in front and behind the truck (>40 cm), the dump truck will also stop. From system testing, it is found that the eye fatigue recognition system has an accuracy value of 94% and the actuator response based on the EAR value input is 100% appropriate. In addition, the safe distance detection using ultrasonic sensor 1 (rear of the truck) has an accuracy rate of 98.3%, and ultrasonic sensor 2 (front of the truck) has an accuracy rate of 98.8% and the actuator response based on the input distance value from ultrasonic 1 and 2 with the lower set point value of 6cm and the upper set point of 40cm is 100% appropriate. It is concluded that the tool design has 3 data inputs, namely 1 web cam and 2 ultrasonics located at the front and rear of the truck with actuators in the form of DC motors. This technology has an overall performance accuracy of 100% based on the integrated testing parameters that have been carried out.

Item Type:	Thesis (Diploma)
Uncontrolled Keywords:	Dump Truck, Safety, Machine Learning, Image Processing, Fatigue
Subjects:	T Technology > T Technology (General) T Technology > T Technology (General) > T55 Industrial Safety T Technology > TA Engineering (General). Civil engineering (General) > TA1573 Detectors. Sensors T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques. Image analysis--Data processing. T Technology > TA Engineering (General). Civil engineering (General) > TA1650 Face recognition. Optical pattern recognition.
Divisions:	Faculty of Vocational > Civil Infrastructure Engineering (D4)
Depositing User:	Akbar Krisna Wandana
Date Deposited:	14 Mar 2025 06:50
Last Modified:	14 Mar 2025 06:50
URI:	http://repository.its.ac.id/id/eprint/118963

Actions (login required)

View Item