Linguistic Decision Making For Robot Route Learning

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Machine learning enables the creation of a nonlinearmapping that describes robot-environment interaction, whilecomputing linguistics make the interaction transparent. In thispaper, we develop a novel application of a linguistic decision treefor a robot route learning problem by dynamically deciding therobot’s behaviour, which is decomposed into atomic actions in thecontext of a specified task.We examine the real-time performanceof training and control of a Linguistic Decision Tree, and explorethe possibility of training a machine learning model in anadaptive system without dual CPUs for parallelisation of trainingand control. A quantified evaluation approach is proposed, anda score is defined for the evaluation of a model’s robustnessregarding the quality of training data. Compared with thenon-linear system identification NARMAX model structure withoffline parameter estimation, the linguistic decision tree modelwith online LID3 learning achieves much better performance,robustness and reliability.
LanguageEnglish
Pages203-215
JournalIEEE Transactions on Neural Networks and Learning Systems
Volume25
Issue number1
DOIs
Publication statusPublished - Jan 2014

Fingerprint

Linguistics
Decision making
Robots
Decision trees
Learning systems
Model structures
Parameter estimation
Program processors
Linear systems
Identification (control systems)

Keywords

  • Linguistic decision tree
  • Task decomposition
  • Atomic action
  • Dynamic behaviour decision
  • Robot route learning.

Cite this

@article{0aad7521bdf249778e2753658dfbe656,
title = "Linguistic Decision Making For Robot Route Learning",
abstract = "Machine learning enables the creation of a nonlinearmapping that describes robot-environment interaction, whilecomputing linguistics make the interaction transparent. In thispaper, we develop a novel application of a linguistic decision treefor a robot route learning problem by dynamically deciding therobot’s behaviour, which is decomposed into atomic actions in thecontext of a specified task.We examine the real-time performanceof training and control of a Linguistic Decision Tree, and explorethe possibility of training a machine learning model in anadaptive system without dual CPUs for parallelisation of trainingand control. A quantified evaluation approach is proposed, anda score is defined for the evaluation of a model’s robustnessregarding the quality of training data. Compared with thenon-linear system identification NARMAX model structure withoffline parameter estimation, the linguistic decision tree modelwith online LID3 learning achieves much better performance,robustness and reliability.",
keywords = "Linguistic decision tree, Task decomposition, Atomic action, Dynamic behaviour decision, Robot route learning.",
author = "Hongmei He and TM McGinnity and SA Coleman and B Gardiner",
year = "2014",
month = "1",
doi = "10.1109/TNNLS.2013.2258037",
language = "English",
volume = "25",
pages = "203--215",
number = "1",

}

Linguistic Decision Making For Robot Route Learning. / He, Hongmei; McGinnity, TM; Coleman, SA; Gardiner, B.

Vol. 25, No. 1, 01.2014, p. 203-215.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Linguistic Decision Making For Robot Route Learning

AU - He, Hongmei

AU - McGinnity, TM

AU - Coleman, SA

AU - Gardiner, B

PY - 2014/1

Y1 - 2014/1

N2 - Machine learning enables the creation of a nonlinearmapping that describes robot-environment interaction, whilecomputing linguistics make the interaction transparent. In thispaper, we develop a novel application of a linguistic decision treefor a robot route learning problem by dynamically deciding therobot’s behaviour, which is decomposed into atomic actions in thecontext of a specified task.We examine the real-time performanceof training and control of a Linguistic Decision Tree, and explorethe possibility of training a machine learning model in anadaptive system without dual CPUs for parallelisation of trainingand control. A quantified evaluation approach is proposed, anda score is defined for the evaluation of a model’s robustnessregarding the quality of training data. Compared with thenon-linear system identification NARMAX model structure withoffline parameter estimation, the linguistic decision tree modelwith online LID3 learning achieves much better performance,robustness and reliability.

AB - Machine learning enables the creation of a nonlinearmapping that describes robot-environment interaction, whilecomputing linguistics make the interaction transparent. In thispaper, we develop a novel application of a linguistic decision treefor a robot route learning problem by dynamically deciding therobot’s behaviour, which is decomposed into atomic actions in thecontext of a specified task.We examine the real-time performanceof training and control of a Linguistic Decision Tree, and explorethe possibility of training a machine learning model in anadaptive system without dual CPUs for parallelisation of trainingand control. A quantified evaluation approach is proposed, anda score is defined for the evaluation of a model’s robustnessregarding the quality of training data. Compared with thenon-linear system identification NARMAX model structure withoffline parameter estimation, the linguistic decision tree modelwith online LID3 learning achieves much better performance,robustness and reliability.

KW - Linguistic decision tree

KW - Task decomposition

KW - Atomic action

KW - Dynamic behaviour decision

KW - Robot route learning.

U2 - 10.1109/TNNLS.2013.2258037

DO - 10.1109/TNNLS.2013.2258037

M3 - Article

VL - 25

SP - 203

EP - 215

IS - 1

ER -