Kinematika Balik Menggunakan Neuro-Fuzzy Pada Manipulator Robot Denso

Rangkuti, Rika Puspitasari (2017) Kinematika Balik Menggunakan Neuro-Fuzzy Pada Manipulator Robot Denso. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 2215105046-Undergraduate_Theses.pdf] Text
2215105046-Undergraduate_Theses.pdf
Restricted to Repository staff only

Download (1MB) | Request a copy

Abstract

Manipulator robot adalah salah satu aplikasi robot yang paling banyak digunakan dalam bidang industri. Pergerakan dari manipulator robot dilakukan dengan menganalisa kinematika robot. Kinematika robot dapat diklasifikasikan menjadi dua jenis yaitu kinematika maju (forward kinematics) dan kinematika balik (inverse kinematics). Perhitungan forward kinematics akan menghasilkan posisi yang diinginkan oleh end-effector sedangkan inverse kinematics menghasilkan besar sudut pada masing-masing joint. Pada Tugas Akhir ini digunakan konsep dari forward kinematics dan inverse kinematics dalam mengontrol daerah kerja sebuah manipulator robot Denso 6-DOF. Permasalahan pada inverse kinematics lebih rumit dibandingkan dengan forward kinematics yang memiliki banyak solusi untuk memperoleh hasilnya, sehingga pada penelitian ini akan dilakukan penyelesaian inverse kinematics menggunakan neuro-fuzzy. Keluaran hasil dari inverse kinematics merupakan besar sudut dari masing – masing joint, hasil output ini akan diuji dengan perhitungan forward kinematics. Pada metode neuro-fuzzy akan dilakukan beberapa percobaan yaitu akan diberikan nilai laju pembelajaran yang berbeda-beda pada range (0.0001 – 0.0009) dan ketelitian error 0.00005, dengan laju pembelajaran = 0.0005 – 0.0009 memiliki nilai error jarak 0.0000554 m, laju pembelajaran = 0.0003 memiliki nilai error jarak 0.0000565 m dan laju pembelajaran = 0.0001 memiliki nilai error jarak 0.000052 m.
==========================================================================================
Robot Manipulator is one of the most widely of application robots used in industry. The movement of the robot manipulator is performed by analyzing the kinematics of the robot. The kinematics of the robot can be classified into two types, namely forward kinematics and inverse kinematics. The calculation of forward kinematics will produce the desired position by the end-effector while the inverse kinematics produces angle on each joint. In this Final Project used the concept of forward kinematics and inverse kinematics in controlling the work area of a Denso 6-DOF robot manipulator. The problem of inverse kinematics is more complicated than forward kinematics which has many solutions to obtain the result, so that in this research will be done inverse kinematics solution using neuro-fuzzy. The output of inverse kinematics is the angle of each joint, the output of which will be tested by forward kinematics calculation. On the method of neuro-fuzzy will do some experiments that will be given the value of learning rate in the range (0.0001 – 0.0009) and accuracy error 0.00005, with learning rate = 0.0005 – 0.0009 has a value of error distance 0.0000554 m, learning rate = 0.0003 has a value of error distance 0.0000565 m and learning rate = 0.0001 has a value of error distance 0.000052 m.

Item Type: Thesis (Undergraduate)
Additional Information: RSE 629.892 Ran k
Uncontrolled Keywords: manipulator robot Denso 6-DOF, forward kinematics, inverse kinematics, neuro-fuzzy
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science. EDP
Q Science > QA Mathematics > QA76 Computer software
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Q Science > QA Mathematics > QA248_Fuzzy Sets
T Technology > T Technology (General)
T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms
Divisions: Faculty of Industrial Technology > Electrical Engineering > 20201-(S1) Undergraduate Thesis
Depositing User: Rika Puspitasari Rangkuti
Date Deposited: 09 Oct 2017 04:21
Last Modified: 03 Jan 2018 04:06
URI: http://repository.its.ac.id/id/eprint/44010

Actions (login required)

View Item View Item