Embedded Adaptive Fuzzy Controller Based on Reinforcement Learning for DC Motor with Flexible Shaft

A. Aziz Khater; Mohammad El-Bardini; Nabila M. El-Rabaie

doi:10.1007/s13369-015-1752-4

Embedded Adaptive Fuzzy Controller Based on Reinforcement Learning for DC Motor with Flexible Shaft

A. Aziz Khater^*
, Mohammad El-Bardini
, Nabila M. El-Rabaie

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

20 Scopus citations

Abstract

In this paper, we propose an adaptive fuzzy controller in which the scaling factors of the input/output membership functions are adapted in the real time using a Reinforcement Q-Learning algorithm based on a proposed reward function. The proposed controller is implemented practically using an Arduino DUE board to control a DC motor with flexible shaft. The practical results show that the performance of the proposed controller is significantly improved compared with the other controllers. Also, the results show better performance over a wide range of the measurement errors and load disturbances.

Original language	English
Pages (from-to)	2389-2406
Number of pages	18
Journal	Arabian Journal for Science and Engineering
Volume	40
Issue number	8
DOIs	https://doi.org/10.1007/s13369-015-1752-4
State	Published - 22 Aug 2015
Externally published	Yes

Bibliographical note

Publisher Copyright:
© 2015, King Fahd University of Petroleum & Minerals.

Keywords

Arduino DUE board
DC motor
Fuzzy PID (FPID) controller
H-bridge
Incremental encoder
Reinforcement Q-learning

ASJC Scopus subject areas

General

Access to Document

10.1007/s13369-015-1752-4

Cite this

@article{f00e0a629c37486e8e3292305f664e20,

title = "Embedded Adaptive Fuzzy Controller Based on Reinforcement Learning for DC Motor with Flexible Shaft",

abstract = "In this paper, we propose an adaptive fuzzy controller in which the scaling factors of the input/output membership functions are adapted in the real time using a Reinforcement Q-Learning algorithm based on a proposed reward function. The proposed controller is implemented practically using an Arduino DUE board to control a DC motor with flexible shaft. The practical results show that the performance of the proposed controller is significantly improved compared with the other controllers. Also, the results show better performance over a wide range of the measurement errors and load disturbances.",

keywords = "Arduino DUE board, DC motor, Fuzzy PID (FPID) controller, H-bridge, Incremental encoder, Reinforcement Q-learning",

author = "\{Aziz Khater\}, A. and Mohammad El-Bardini and El-Rabaie, \{Nabila M.\}",

note = "Publisher Copyright: {\textcopyright} 2015, King Fahd University of Petroleum \& Minerals.",

year = "2015",

month = aug,

day = "22",

doi = "10.1007/s13369-015-1752-4",

language = "English",

volume = "40",

pages = "2389--2406",

journal = "Arabian Journal for Science and Engineering",

issn = "2193-567X",

publisher = "Springer Nature",

number = "8",

}

TY - JOUR

T1 - Embedded Adaptive Fuzzy Controller Based on Reinforcement Learning for DC Motor with Flexible Shaft

AU - Aziz Khater, A.

AU - El-Bardini, Mohammad

AU - El-Rabaie, Nabila M.

PY - 2015/8/22

Y1 - 2015/8/22

N2 - In this paper, we propose an adaptive fuzzy controller in which the scaling factors of the input/output membership functions are adapted in the real time using a Reinforcement Q-Learning algorithm based on a proposed reward function. The proposed controller is implemented practically using an Arduino DUE board to control a DC motor with flexible shaft. The practical results show that the performance of the proposed controller is significantly improved compared with the other controllers. Also, the results show better performance over a wide range of the measurement errors and load disturbances.

AB - In this paper, we propose an adaptive fuzzy controller in which the scaling factors of the input/output membership functions are adapted in the real time using a Reinforcement Q-Learning algorithm based on a proposed reward function. The proposed controller is implemented practically using an Arduino DUE board to control a DC motor with flexible shaft. The practical results show that the performance of the proposed controller is significantly improved compared with the other controllers. Also, the results show better performance over a wide range of the measurement errors and load disturbances.

KW - Arduino DUE board

KW - DC motor

KW - Fuzzy PID (FPID) controller

KW - H-bridge

KW - Incremental encoder

KW - Reinforcement Q-learning

UR - https://www.scopus.com/pages/publications/84937597171

U2 - 10.1007/s13369-015-1752-4

DO - 10.1007/s13369-015-1752-4

M3 - Article

AN - SCOPUS:84937597171

SN - 2193-567X

VL - 40

SP - 2389

EP - 2406

JO - Arabian Journal for Science and Engineering

JF - Arabian Journal for Science and Engineering

IS - 8

ER -

Embedded Adaptive Fuzzy Controller Based on Reinforcement Learning for DC Motor with Flexible Shaft

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this