Sistem Auto Docking Pada Service Robot Menggunakan Persepsi Visual

Firmansyah, Riza Agung (2015) Sistem Auto Docking Pada Service Robot Menggunakan Persepsi Visual. Masters thesis, Institut Technology Sepuluh Nopember.

[img]
Preview
Text
2212204014-Master Thesis.pdf - Published Version

Download (1MB) | Preview

Abstract

Sistem auto docking adalah sebuah sistem pada sebuah robot yang berfungsi untuk melakukan pengisian baterai secara otomatis. Sistem ini dijalankan saat robot mendeteksi tegangan kerja minimal. Robot melakukan pengisian baterai pada sebuah docking station. Sistem auto docking ini dibangun dengan menggunakan persepsi visual berbasis local binary pattern (LBP) histogram matching agar robot mampu mendeteksi docking station. Setelah pendeteksian berhasil, dilanjutkan dengan ekstraksi fitur untuk mendapatkan posisi dan orientasi docking station. Selanjutnya posisi dan orientasi dijadikan sebagai informasi masukan fuzzy logic controller untuk menjalankan robot. Robot menjalankan sistem auto docking menggunakan persepsi visual saat tegangan baterai dibawah 23.6 volt dan menghentikan pengisian baterai saat tegangan baterai diatas 26.4 volt. Pengisian baterai dilakukan dalam waktu 135 menit. Sistem auto docking mampu bekerja dengan tingkat akurasi 86.7 % dan optimal pada rentang luminasi 116 lux hingga 395 lux. Robot melakukan penyambungan konektor dengan rata-rata waktu 53.78 detik dari jarak 450 cm dengan akurasi penyambungan konektor mencapai 84%. ===================================================================================================== Auto docking system is a system on a robot that has a function to make the battery charging automatically. This system is executed when the robot detects a minimum working voltage. Robot perform battery charging on a docking station. Auto docking system was built using visual perception based on local binary pattern (LBP) histogram matching to detect a docking station. After docking station detected, then feature extraction is performed to get the posistion and orientation of the docking station. Position and orientation is used as fuzzy logic controller input to move the robot. Robot executed the auto docking system using visual perception when the battery voltage is below 23.6 volts and finish charging the battery when voltage is over 26.4 volts. Battery charging is done within 135 minutes. Auto docking system using visual perception has an accuration 86.7% and optimally work at luminance 116 lux up to 395 lux. The robot is able to perform splicing connector with an average time of 53.78 seconds from a distance of 450 cm with an accuracy of 84%.

Item Type: Thesis (Masters)
Additional Information: RTE 629.892 Fir s
Uncontrolled Keywords: service robot, auto docking, persepsi visual, local binary pattern, histogram matching, fuzzy logic controller.
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions: Faculty of Industrial Technology > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User: Mr. Tondo Indra Nyata
Date Deposited: 06 Nov 2018 03:21
Last Modified: 06 Nov 2018 03:21
URI: https://repository.its.ac.id/id/eprint/59938

Actions (login required)

View Item View Item