Advisor-enhanced real-time system control

D. W. Russell

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

The paper describes a hybrid signature-table controller that uses advisor logic to enhance its decision making and statistical-inference learning processes. The automaton reacts to changes in real state space with rapid-fire control decisions and uses timeouts or failures to initiate learning interludes. The post-processing algorithm causes the system to learn by adapting values in the control matrix with a process that utilizes computed decision strengths for each state and those of its advising neighbors. The familiar inverted pendulum system is used to highlight the effects of advisor logic on overall system performance.

Original languageEnglish (US)
Pages (from-to)977-983
Number of pages7
JournalControl Engineering Practice
Volume3
Issue number7
DOIs
StatePublished - Jul 1995

Fingerprint

Real time systems
Logic
Real-time
Inverted Pendulum
Pendulums
Statistical Inference
Learning Process
Post-processing
Automata
System Performance
Table
State Space
Fires
Signature
Decision making
Decision Making
Controller
Controllers
Processing
Learning

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Computer Science Applications
  • Electrical and Electronic Engineering
  • Applied Mathematics

Cite this

Russell, D. W. / Advisor-enhanced real-time system control. In: Control Engineering Practice. 1995 ; Vol. 3, No. 7. pp. 977-983.
@article{ac4ad2e517ba48d4b89aceb5a3043b56,
title = "Advisor-enhanced real-time system control",
abstract = "The paper describes a hybrid signature-table controller that uses advisor logic to enhance its decision making and statistical-inference learning processes. The automaton reacts to changes in real state space with rapid-fire control decisions and uses timeouts or failures to initiate learning interludes. The post-processing algorithm causes the system to learn by adapting values in the control matrix with a process that utilizes computed decision strengths for each state and those of its advising neighbors. The familiar inverted pendulum system is used to highlight the effects of advisor logic on overall system performance.",
author = "Russell, {D. W.}",
year = "1995",
month = "7",
doi = "10.1016/0967-0661(95)00080-E",
language = "English (US)",
volume = "3",
pages = "977--983",
journal = "Control Engineering Practice",
issn = "0967-0661",
publisher = "Elsevier Limited",
number = "7",

}

Advisor-enhanced real-time system control. / Russell, D. W.

In: Control Engineering Practice, Vol. 3, No. 7, 07.1995, p. 977-983.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Advisor-enhanced real-time system control

AU - Russell, D. W.

PY - 1995/7

Y1 - 1995/7

N2 - The paper describes a hybrid signature-table controller that uses advisor logic to enhance its decision making and statistical-inference learning processes. The automaton reacts to changes in real state space with rapid-fire control decisions and uses timeouts or failures to initiate learning interludes. The post-processing algorithm causes the system to learn by adapting values in the control matrix with a process that utilizes computed decision strengths for each state and those of its advising neighbors. The familiar inverted pendulum system is used to highlight the effects of advisor logic on overall system performance.

AB - The paper describes a hybrid signature-table controller that uses advisor logic to enhance its decision making and statistical-inference learning processes. The automaton reacts to changes in real state space with rapid-fire control decisions and uses timeouts or failures to initiate learning interludes. The post-processing algorithm causes the system to learn by adapting values in the control matrix with a process that utilizes computed decision strengths for each state and those of its advising neighbors. The familiar inverted pendulum system is used to highlight the effects of advisor logic on overall system performance.

UR - http://www.scopus.com/inward/record.url?scp=0029344325&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029344325&partnerID=8YFLogxK

U2 - 10.1016/0967-0661(95)00080-E

DO - 10.1016/0967-0661(95)00080-E

M3 - Article

AN - SCOPUS:0029344325

VL - 3

SP - 977

EP - 983

JO - Control Engineering Practice

JF - Control Engineering Practice

SN - 0967-0661

IS - 7

ER -