Spreadsheet-based neural networks modelling and simulation for training and predicting inverse kinematics of robot arm

Khairul Annuar Abdullah, Zuriati Yusof, Riza Sulaiman

Research output: Contribution to journalArticle

Abstract

This paper is proposed to solve the inverse kinematic (IK) problem of two-degree-of-freedom planar robot arm using neural networks (NN). Several NN model designs of distinct hidden neurons based on the sum of square error function of joint angle are developed and trained with generalised reduced gradient algorithm. The paper is also intended to demonstrate the modelling process of feed-forward NN topology in spreadsheet environment. The spreadsheet functions as INDEX, SUMPRODUCT, EXP, and SUMSQ; the utilities as name manager, data validation, data table, ActiveX controls, answer report, and charts; and the add-in Solver are utilised to develop the models. With the input parameters of link lengths and end-effector position and orientation, two models with the structures 5-12-1 and 5-10-1 are discovered best-capable in predicting first and second joint angles respectively. This NNbased IK technique contributes significantly to the optimal motion control of robot arm for quality processing and assembly tasks.

Original languageEnglish
Pages (from-to)218-243
Number of pages26
JournalInternational Journal of Computer Aided Engineering and Technology
Volume10
Issue number3
DOIs
Publication statusPublished - 1 Jan 2018

Fingerprint

Inverse kinematics
Spreadsheets
Robots
Neural networks
Feedforward neural networks
Degrees of freedom (mechanics)
Motion control
End effectors
Neurons
Managers
Topology
Processing

Keywords

  • Feed-forward neural networks
  • Generalised reduced gradient algorithm
  • Inverse kinematics
  • Multiple linear regression
  • Robot arm
  • Spreadsheet modelling and simulation

ASJC Scopus subject areas

  • Software
  • Engineering(all)
  • Computer Science Applications

Cite this

Spreadsheet-based neural networks modelling and simulation for training and predicting inverse kinematics of robot arm. / Abdullah, Khairul Annuar; Yusof, Zuriati; Sulaiman, Riza.

In: International Journal of Computer Aided Engineering and Technology, Vol. 10, No. 3, 01.01.2018, p. 218-243.

Research output: Contribution to journalArticle

@article{89f47a1f8aa8450d9de9971808e41cd5,
title = "Spreadsheet-based neural networks modelling and simulation for training and predicting inverse kinematics of robot arm",
abstract = "This paper is proposed to solve the inverse kinematic (IK) problem of two-degree-of-freedom planar robot arm using neural networks (NN). Several NN model designs of distinct hidden neurons based on the sum of square error function of joint angle are developed and trained with generalised reduced gradient algorithm. The paper is also intended to demonstrate the modelling process of feed-forward NN topology in spreadsheet environment. The spreadsheet functions as INDEX, SUMPRODUCT, EXP, and SUMSQ; the utilities as name manager, data validation, data table, ActiveX controls, answer report, and charts; and the add-in Solver are utilised to develop the models. With the input parameters of link lengths and end-effector position and orientation, two models with the structures 5-12-1 and 5-10-1 are discovered best-capable in predicting first and second joint angles respectively. This NNbased IK technique contributes significantly to the optimal motion control of robot arm for quality processing and assembly tasks.",
keywords = "Feed-forward neural networks, Generalised reduced gradient algorithm, Inverse kinematics, Multiple linear regression, Robot arm, Spreadsheet modelling and simulation",
author = "Abdullah, {Khairul Annuar} and Zuriati Yusof and Riza Sulaiman",
year = "2018",
month = "1",
day = "1",
doi = "10.1504/IJCAET.2018.10011625",
language = "English",
volume = "10",
pages = "218--243",
journal = "International Journal of Computer Aided Engineering and Technology",
issn = "1757-2657",
publisher = "Inderscience Publishers",
number = "3",

}

TY - JOUR

T1 - Spreadsheet-based neural networks modelling and simulation for training and predicting inverse kinematics of robot arm

AU - Abdullah, Khairul Annuar

AU - Yusof, Zuriati

AU - Sulaiman, Riza

PY - 2018/1/1

Y1 - 2018/1/1

N2 - This paper is proposed to solve the inverse kinematic (IK) problem of two-degree-of-freedom planar robot arm using neural networks (NN). Several NN model designs of distinct hidden neurons based on the sum of square error function of joint angle are developed and trained with generalised reduced gradient algorithm. The paper is also intended to demonstrate the modelling process of feed-forward NN topology in spreadsheet environment. The spreadsheet functions as INDEX, SUMPRODUCT, EXP, and SUMSQ; the utilities as name manager, data validation, data table, ActiveX controls, answer report, and charts; and the add-in Solver are utilised to develop the models. With the input parameters of link lengths and end-effector position and orientation, two models with the structures 5-12-1 and 5-10-1 are discovered best-capable in predicting first and second joint angles respectively. This NNbased IK technique contributes significantly to the optimal motion control of robot arm for quality processing and assembly tasks.

AB - This paper is proposed to solve the inverse kinematic (IK) problem of two-degree-of-freedom planar robot arm using neural networks (NN). Several NN model designs of distinct hidden neurons based on the sum of square error function of joint angle are developed and trained with generalised reduced gradient algorithm. The paper is also intended to demonstrate the modelling process of feed-forward NN topology in spreadsheet environment. The spreadsheet functions as INDEX, SUMPRODUCT, EXP, and SUMSQ; the utilities as name manager, data validation, data table, ActiveX controls, answer report, and charts; and the add-in Solver are utilised to develop the models. With the input parameters of link lengths and end-effector position and orientation, two models with the structures 5-12-1 and 5-10-1 are discovered best-capable in predicting first and second joint angles respectively. This NNbased IK technique contributes significantly to the optimal motion control of robot arm for quality processing and assembly tasks.

KW - Feed-forward neural networks

KW - Generalised reduced gradient algorithm

KW - Inverse kinematics

KW - Multiple linear regression

KW - Robot arm

KW - Spreadsheet modelling and simulation

UR - http://www.scopus.com/inward/record.url?scp=85044353275&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85044353275&partnerID=8YFLogxK

U2 - 10.1504/IJCAET.2018.10011625

DO - 10.1504/IJCAET.2018.10011625

M3 - Article

VL - 10

SP - 218

EP - 243

JO - International Journal of Computer Aided Engineering and Technology

JF - International Journal of Computer Aided Engineering and Technology

SN - 1757-2657

IS - 3

ER -