Arlean, Tegar Wangi (2017) Kinematika Balik Manipulator Robot Denso Dengan Metode Neural Network. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
2215105051-Undergraduate_Theses.pdf - Published Version Download (6MB) | Preview |
Abstract
Manipulator robot adalah mekanik elektronik dengan cara kerja menyerupai lengan manusia. Manipulator robot adalah peralatan yang sering digunakan di dalam bidang industri robot dan tersusun dari sendi (joint) , link , dan end-effector. Terdapat dua jenis sendi robot, yaitu sendi putar (revolute joint) dan sendi geser (prismatic joint). Kinematika maju (forward kinematics) dan kinematika balik (inverse kinematics) merupakan konsep perhitungan daerah kerja dari manipulator robot. Daerah kerja dari forward kinematics berupa ruang joint, sedangkan daerah kerja inverse kinematics berupa ruang cartesian. Forward kinematics dan inverse kinematics diterapkan pada manipulator robot Denso 6-DOF. Perhitungan forward kinematics menghasilkan posisi yang di inginkan oleh end-effetor dari masukan berupa nilai semua joint, sedangkan inverse kinematics menghasilkan nilai untuk setiap joint dari masukan berupa posisi end-effector. Untuk menyelesaikan permasalahan inverse kinematics yang mempunyai banyak solusi dibandingkan dengan forward kinematics, maka digunakan metode neural network. Pada inverse kinematics neural network dilakukan pengujian pola persegi pada ketelitian 0,00005 dan learning rate 0,00025 dengan jumlah 150 neuron menghasilkan error jarak x 1,49e-05 m , error jarak y 1,06e-05 m, dan error jarak z 2,93e-05 m.
====================================================================
Robot manipulator is an electronic mechanic by means of work resembling a human arm. Robot manipulators are commonly used tools in the robot industry and are composed of joints, links, and end-effector. There are two types of robotic joints, namely the revolute joint and the prismatic joint. Forward kinematics and inverse kinematics are the calculation concepts of the working area of the robot manipulator. The working area of the forward kinematics is a joint space, while the inverse kinematics working area is cartesian space. Forward kinematics and inverse kinematics are applied to the Denso 6-DOF robot manipulator. The calculation of forward kinematics produces the desired position by the end-effetor of the inputs of the value of all joints, whereas the inverse kinematics yields the value for each joint of the input as the end-effector position. To solve the inverse kinematics problem which has many solutions compared with forward kinematics, then the neural network method is used. In the inverse kinematics neural network, a square pattern test of 0.00005 and learning rate 0.00025 with 150 neurons resulted in x 1.49e-05 m distance error, 1.06e-05 m distance error, and z-distance error of 2.93e-05 m.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSE 629.892 Arl k |
| Uncontrolled Keywords: | manipulator robot, Denso, 6-DOF, forward kinematics, inverse kinematics, neural network. |
| 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) 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: | Tegar Wangi Arlean |
| Date Deposited: | 25 Sep 2017 07:39 |
| Last Modified: | 06 Mar 2019 03:25 |
| URI: | http://repository.its.ac.id/id/eprint/44020 |
Actions (login required)
 |
View Item |