Stabilization of an inverted pendulum by a layered neural network

Minoru Sekiguchi, Tamami Sugasaka, Ryo Kurazume

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

This paper discusses the stabilization of an inverted pendulum by a neural network. The inverted pendulum is a typical example of an unstable and nonlinear system. Fujitsu has developed a self-learning method that enables a neural network to control an inverted pendulum by using heuristic rules. A system based on this method has been developed which generates teaching data for inversion by trial and error. Then, the neural network learns the teaching data, and acquires the control rules after a period of trial and error. The method proved its effectiveness by stably inverting a pendulum in an experimental inverted pendulum system after a period of trial and error.

Original languageEnglish
Pages (from-to)278-285
Number of pages8
JournalFujitsu Scientific and Technical Journal
Volume29
Issue number3
Publication statusPublished - Sep 1993
Externally publishedYes

Fingerprint

Pendulums
Stabilization
Neural networks
Teaching
Nonlinear systems

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering

Cite this

Stabilization of an inverted pendulum by a layered neural network. / Sekiguchi, Minoru; Sugasaka, Tamami; Kurazume, Ryo.

In: Fujitsu Scientific and Technical Journal, Vol. 29, No. 3, 09.1993, p. 278-285.

Research output: Contribution to journalArticle

Sekiguchi, Minoru ; Sugasaka, Tamami ; Kurazume, Ryo. / Stabilization of an inverted pendulum by a layered neural network. In: Fujitsu Scientific and Technical Journal. 1993 ; Vol. 29, No. 3. pp. 278-285.
@article{139e6bea633d4a0d9a6af8b205e44683,
title = "Stabilization of an inverted pendulum by a layered neural network",
abstract = "This paper discusses the stabilization of an inverted pendulum by a neural network. The inverted pendulum is a typical example of an unstable and nonlinear system. Fujitsu has developed a self-learning method that enables a neural network to control an inverted pendulum by using heuristic rules. A system based on this method has been developed which generates teaching data for inversion by trial and error. Then, the neural network learns the teaching data, and acquires the control rules after a period of trial and error. The method proved its effectiveness by stably inverting a pendulum in an experimental inverted pendulum system after a period of trial and error.",
author = "Minoru Sekiguchi and Tamami Sugasaka and Ryo Kurazume",
year = "1993",
month = "9",
language = "English",
volume = "29",
pages = "278--285",
journal = "Fujitsu Scientific and Technical Journal",
issn = "0016-2523",
publisher = "Fujitsu Ltd",
number = "3",

}

TY - JOUR

T1 - Stabilization of an inverted pendulum by a layered neural network

AU - Sekiguchi, Minoru

AU - Sugasaka, Tamami

AU - Kurazume, Ryo

PY - 1993/9

Y1 - 1993/9

N2 - This paper discusses the stabilization of an inverted pendulum by a neural network. The inverted pendulum is a typical example of an unstable and nonlinear system. Fujitsu has developed a self-learning method that enables a neural network to control an inverted pendulum by using heuristic rules. A system based on this method has been developed which generates teaching data for inversion by trial and error. Then, the neural network learns the teaching data, and acquires the control rules after a period of trial and error. The method proved its effectiveness by stably inverting a pendulum in an experimental inverted pendulum system after a period of trial and error.

AB - This paper discusses the stabilization of an inverted pendulum by a neural network. The inverted pendulum is a typical example of an unstable and nonlinear system. Fujitsu has developed a self-learning method that enables a neural network to control an inverted pendulum by using heuristic rules. A system based on this method has been developed which generates teaching data for inversion by trial and error. Then, the neural network learns the teaching data, and acquires the control rules after a period of trial and error. The method proved its effectiveness by stably inverting a pendulum in an experimental inverted pendulum system after a period of trial and error.

UR - http://www.scopus.com/inward/record.url?scp=0027663878&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0027663878&partnerID=8YFLogxK

M3 - Article

VL - 29

SP - 278

EP - 285

JO - Fujitsu Scientific and Technical Journal

JF - Fujitsu Scientific and Technical Journal

SN - 0016-2523

IS - 3

ER -