Quantitative object motion prediction by an ART2 and Madaline combined neural network: Concepts and experiments

Qiuming Zhu, Ahmed Y. Tawfik

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

A neural network that combines the structures of an ART2 network and a Madaline network is applied to the prediction of object motions in dynamic environments. By the use of an ART2 network, the system extracts a set of coherent motion patterns from a sequence of noise-corrupted input signals. This process is done in an unsupervised learning mode. The system then uses the extracted patterns to direct the quantitative prediction of the future motion states of the object through a Madaline network. A push-forward shift-register keeps track of the object motion profile in real time, and provides input signals to both networks. The method does not require any assumption of mathematical models for the object motions, and is applicable to a variety of situations, such as environments with cooperating robots, where the object motions exhibit quite irregular and vigorous variations.

Original languageEnglish (US)
Pages (from-to)569-578
Number of pages10
JournalEngineering Applications of Artificial Intelligence
Volume8
Issue number5
DOIs
StatePublished - Oct 1995

Fingerprint

Neural networks
Unsupervised learning
Shift registers
Experiments
Robots
Mathematical models

Keywords

  • ART2 neural network
  • Madaline network
  • Object motion prediction
  • quantitative prediction

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Artificial Intelligence
  • Electrical and Electronic Engineering

Cite this

@article{c799ece981bb4c83a28dc2ecd330c8fc,
title = "Quantitative object motion prediction by an ART2 and Madaline combined neural network: Concepts and experiments",
abstract = "A neural network that combines the structures of an ART2 network and a Madaline network is applied to the prediction of object motions in dynamic environments. By the use of an ART2 network, the system extracts a set of coherent motion patterns from a sequence of noise-corrupted input signals. This process is done in an unsupervised learning mode. The system then uses the extracted patterns to direct the quantitative prediction of the future motion states of the object through a Madaline network. A push-forward shift-register keeps track of the object motion profile in real time, and provides input signals to both networks. The method does not require any assumption of mathematical models for the object motions, and is applicable to a variety of situations, such as environments with cooperating robots, where the object motions exhibit quite irregular and vigorous variations.",
keywords = "ART2 neural network, Madaline network, Object motion prediction, quantitative prediction",
author = "Qiuming Zhu and Tawfik, {Ahmed Y.}",
year = "1995",
month = "10",
doi = "10.1016/0952-1976(95)00034-X",
language = "English (US)",
volume = "8",
pages = "569--578",
journal = "Engineering Applications of Artificial Intelligence",
issn = "0952-1976",
publisher = "Elsevier Limited",
number = "5",

}

TY - JOUR

T1 - Quantitative object motion prediction by an ART2 and Madaline combined neural network

T2 - Concepts and experiments

AU - Zhu, Qiuming

AU - Tawfik, Ahmed Y.

PY - 1995/10

Y1 - 1995/10

N2 - A neural network that combines the structures of an ART2 network and a Madaline network is applied to the prediction of object motions in dynamic environments. By the use of an ART2 network, the system extracts a set of coherent motion patterns from a sequence of noise-corrupted input signals. This process is done in an unsupervised learning mode. The system then uses the extracted patterns to direct the quantitative prediction of the future motion states of the object through a Madaline network. A push-forward shift-register keeps track of the object motion profile in real time, and provides input signals to both networks. The method does not require any assumption of mathematical models for the object motions, and is applicable to a variety of situations, such as environments with cooperating robots, where the object motions exhibit quite irregular and vigorous variations.

AB - A neural network that combines the structures of an ART2 network and a Madaline network is applied to the prediction of object motions in dynamic environments. By the use of an ART2 network, the system extracts a set of coherent motion patterns from a sequence of noise-corrupted input signals. This process is done in an unsupervised learning mode. The system then uses the extracted patterns to direct the quantitative prediction of the future motion states of the object through a Madaline network. A push-forward shift-register keeps track of the object motion profile in real time, and provides input signals to both networks. The method does not require any assumption of mathematical models for the object motions, and is applicable to a variety of situations, such as environments with cooperating robots, where the object motions exhibit quite irregular and vigorous variations.

KW - ART2 neural network

KW - Madaline network

KW - Object motion prediction

KW - quantitative prediction

UR - http://www.scopus.com/inward/record.url?scp=0029388880&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029388880&partnerID=8YFLogxK

U2 - 10.1016/0952-1976(95)00034-X

DO - 10.1016/0952-1976(95)00034-X

M3 - Article

AN - SCOPUS:0029388880

VL - 8

SP - 569

EP - 578

JO - Engineering Applications of Artificial Intelligence

JF - Engineering Applications of Artificial Intelligence

SN - 0952-1976

IS - 5

ER -