Position control of a robotic manipulator using a Radial Basis Function Network and a simple vision system

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper describes a new practical approach for approximating the inverse kinematics of a manipulator using a RBFN (Radial Basis Function Network). In fact, there are several traditional methods based on the known geometry of the manipulator to calculate the relationship between the joint variable space and the world coordinate space. However, these traditional methods are impractical if the manipulator geometry cannot be easily determined, in a robot-vision system for example. Therefore, a neural network with its inherent learning ability can be an effective alternative solution for the inverse kinematics problem. In this paper, a training approach using the strict interpolation method combined with the LMS (Least Mean Square) is presented. The strict interpolation method with regularly spaced position training patterns in the workspace can produce an appropriate approximation of the inverse kinematic function. Additionally, the LMS algorithm can improve the approximate function iteratively through on-line training with arbitrary position patterns. The combination of these techniques can deal with variation in the set-up of the visual system used to measure the position of the manipulator in the workspace. To verify the performance of the proposed approach, a practical experiment has been performed using a Mitsubishi PA10-6CE manipulator observed by a webcam. All application programmes, such as robot servo control, neural network, and image processing tool, were written in C/C++ and run in a real robotic system. The experimental results prove that the proposed approach is effective. © 2008 IEEE.

Original languageEnglish
Title of host publication2008 IEEE International Symposium on Industrial Electronics, ISIE 2008
Pages1371-1376
Number of pages6
DOIs
Publication statusPublished - 2008
Event2008 IEEE International Symposium on Industrial Electronics - Cambridge, United Kingdom
Duration: 30 Jun 20082 Jul 2008

Conference

Conference2008 IEEE International Symposium on Industrial Electronics
Abbreviated titleISIE 2008
CountryUnited Kingdom
CityCambridge
Period30/06/082/07/08

Fingerprint

Radial basis function networks
Position control
Manipulators
Robotics
Inverse kinematics
Interpolation
Neural networks
Geometry
Application programs
Computer vision
Image processing
Robots
Experiments

Cite this

Bach, H. Dinh ; Dunnigan, Mathew Walter ; Reay, Donald Shewan. / Position control of a robotic manipulator using a Radial Basis Function Network and a simple vision system. 2008 IEEE International Symposium on Industrial Electronics, ISIE 2008. 2008. pp. 1371-1376
@inproceedings{c0c3f1c3e71c4148b83e12faaa9c7313,
title = "Position control of a robotic manipulator using a Radial Basis Function Network and a simple vision system",
abstract = "This paper describes a new practical approach for approximating the inverse kinematics of a manipulator using a RBFN (Radial Basis Function Network). In fact, there are several traditional methods based on the known geometry of the manipulator to calculate the relationship between the joint variable space and the world coordinate space. However, these traditional methods are impractical if the manipulator geometry cannot be easily determined, in a robot-vision system for example. Therefore, a neural network with its inherent learning ability can be an effective alternative solution for the inverse kinematics problem. In this paper, a training approach using the strict interpolation method combined with the LMS (Least Mean Square) is presented. The strict interpolation method with regularly spaced position training patterns in the workspace can produce an appropriate approximation of the inverse kinematic function. Additionally, the LMS algorithm can improve the approximate function iteratively through on-line training with arbitrary position patterns. The combination of these techniques can deal with variation in the set-up of the visual system used to measure the position of the manipulator in the workspace. To verify the performance of the proposed approach, a practical experiment has been performed using a Mitsubishi PA10-6CE manipulator observed by a webcam. All application programmes, such as robot servo control, neural network, and image processing tool, were written in C/C++ and run in a real robotic system. The experimental results prove that the proposed approach is effective. {\circledC} 2008 IEEE.",
author = "Bach, {H. Dinh} and Dunnigan, {Mathew Walter} and Reay, {Donald Shewan}",
year = "2008",
doi = "10.1109/ISIE.2008.4677070",
language = "English",
isbn = "1424416655",
pages = "1371--1376",
booktitle = "2008 IEEE International Symposium on Industrial Electronics, ISIE 2008",

}

Bach, HD, Dunnigan, MW & Reay, DS 2008, Position control of a robotic manipulator using a Radial Basis Function Network and a simple vision system. in 2008 IEEE International Symposium on Industrial Electronics, ISIE 2008. pp. 1371-1376, 2008 IEEE International Symposium on Industrial Electronics, Cambridge, United Kingdom, 30/06/08. https://doi.org/10.1109/ISIE.2008.4677070

Position control of a robotic manipulator using a Radial Basis Function Network and a simple vision system. / Bach, H. Dinh; Dunnigan, Mathew Walter; Reay, Donald Shewan.

2008 IEEE International Symposium on Industrial Electronics, ISIE 2008. 2008. p. 1371-1376.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Position control of a robotic manipulator using a Radial Basis Function Network and a simple vision system

AU - Bach, H. Dinh

AU - Dunnigan, Mathew Walter

AU - Reay, Donald Shewan

PY - 2008

Y1 - 2008

N2 - This paper describes a new practical approach for approximating the inverse kinematics of a manipulator using a RBFN (Radial Basis Function Network). In fact, there are several traditional methods based on the known geometry of the manipulator to calculate the relationship between the joint variable space and the world coordinate space. However, these traditional methods are impractical if the manipulator geometry cannot be easily determined, in a robot-vision system for example. Therefore, a neural network with its inherent learning ability can be an effective alternative solution for the inverse kinematics problem. In this paper, a training approach using the strict interpolation method combined with the LMS (Least Mean Square) is presented. The strict interpolation method with regularly spaced position training patterns in the workspace can produce an appropriate approximation of the inverse kinematic function. Additionally, the LMS algorithm can improve the approximate function iteratively through on-line training with arbitrary position patterns. The combination of these techniques can deal with variation in the set-up of the visual system used to measure the position of the manipulator in the workspace. To verify the performance of the proposed approach, a practical experiment has been performed using a Mitsubishi PA10-6CE manipulator observed by a webcam. All application programmes, such as robot servo control, neural network, and image processing tool, were written in C/C++ and run in a real robotic system. The experimental results prove that the proposed approach is effective. © 2008 IEEE.

AB - This paper describes a new practical approach for approximating the inverse kinematics of a manipulator using a RBFN (Radial Basis Function Network). In fact, there are several traditional methods based on the known geometry of the manipulator to calculate the relationship between the joint variable space and the world coordinate space. However, these traditional methods are impractical if the manipulator geometry cannot be easily determined, in a robot-vision system for example. Therefore, a neural network with its inherent learning ability can be an effective alternative solution for the inverse kinematics problem. In this paper, a training approach using the strict interpolation method combined with the LMS (Least Mean Square) is presented. The strict interpolation method with regularly spaced position training patterns in the workspace can produce an appropriate approximation of the inverse kinematic function. Additionally, the LMS algorithm can improve the approximate function iteratively through on-line training with arbitrary position patterns. The combination of these techniques can deal with variation in the set-up of the visual system used to measure the position of the manipulator in the workspace. To verify the performance of the proposed approach, a practical experiment has been performed using a Mitsubishi PA10-6CE manipulator observed by a webcam. All application programmes, such as robot servo control, neural network, and image processing tool, were written in C/C++ and run in a real robotic system. The experimental results prove that the proposed approach is effective. © 2008 IEEE.

UR - http://www.scopus.com/inward/record.url?scp=57849165142&partnerID=8YFLogxK

U2 - 10.1109/ISIE.2008.4677070

DO - 10.1109/ISIE.2008.4677070

M3 - Conference contribution

SN - 1424416655

SN - 9781424416653

SP - 1371

EP - 1376

BT - 2008 IEEE International Symposium on Industrial Electronics, ISIE 2008

ER -