Perancangan Penyelesaian Kinematika Balik Pada Posisi Dan Orientasi End Effector Robot Open Manipulator-X Dengan Metode Neural Network

Syauqi, Muhammad Yavi Marsa (2023) Perancangan Penyelesaian Kinematika Balik Pada Posisi Dan Orientasi End Effector Robot Open Manipulator-X Dengan Metode Neural Network. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 07111940000221-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (3MB) | Request a copy

Abstract

Robot manipulator merupakan bukti dari perkembangan teknologi otomasi di era 4.0 saat ini, Robot yang memiliki 4 degree of freedom (DoF) dalam pergerakannya merupakan salah satu sistem yang biasa digunakan untuk mengefisiensi dan mengurangi resiko kecelakaan pada dunia Industri. Namun, pada penyelesaian kinematika balik robot manipulator terdapat permasalahan dimana terdapat multi solution dalam mendapatkan nilai sudut keluaran pada setiap joint. Oleh sebab itu, Penelitian ini bertujuan untuk menyelesaikan permasalahan kinematika balik tersebut menggunakan metode neural network menggunakan Robot Open Manipulator-X. Penelitian ini dilakukan dengan pengambilan dataset dengan kinematika maju, perancangan model neural network dengan melakukan pengujian pada jumlah neuron, hidden layer, dan learning rate, serta pembuktian dengan pengujian titik dari hasil neural network pada robot melalui simulasi pada gazebo dan real plant. Hasil dari peneletian ini yaitu metode neural network dapat menemukan single solution sebagai penyelesaian kinematika balik untuk Robot Open Manipulator-X dengan rata-rata error posisi dan orientasi sebesar (0.36 ± 0.06) cm dan (3.224±2.942)° pada simulasi, serta (0.37 ± 0.06) cm dan (3.224±2.942)° pada real plant.
====================================================================================================================================
Manipulator robot is evidence of the technological advancement in automation in the current era of Industry 4.0. A robot with 4 degrees of freedom (DoF) in its movement is commonly used to improve efficiency and reduce the risk of accidents in the industrial world. However, there is a problem in solving the inverse kinematics of robot manipulators where multiple solutions exist in obtaining the angle values for each joint. Therefore, this research aims to solve the inverse kinematics problem using a neural network method with the Robot Open Manipulator-X. The research involves collecting a dataset with forward kinematics, designing a neural network model by testing various configurations such as the number of neurons, hidden layers, and learning rate, and validating the results through point testing of the neural network's output on the robot using simulations in Gazebo and real plant environments. The result of this research is that the neural network method can find a single solution as the inverse kinematics solution for the Robot Open Manipulator-X with an average position error of (0.36 ± 0.06) cm and orientation error of (3.224±2.942)° in simulation, as well as (0.37 ± 0.06) cm and (3.224±2.942)° in the real plant.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Kinematika balik, Robot Manipulator, Neural network
Subjects:	Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models. T Technology > TJ Mechanical engineering and machinery > TJ211 Robotics. T Technology > TJ Mechanical engineering and machinery > TJ211.4 Robot motion
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20201-(S1) Undergraduate Thesis
Depositing User:	Muhammad Yavi Marsa Syauqi
Date Deposited:	26 Jul 2023 04:21
Last Modified:	26 Jul 2023 04:21
URI:	http://repository.its.ac.id/id/eprint/98932

Actions (login required)

View Item