Artificial Neural Network based dynamic modelling of indigenous pneumatic muscle actuators

Prashant K. Jamwal, Sheng Quan Xie

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Robots are increasingly becoming popular medical devices, helping surgeons and practitioners as surgical, rehabilitation or service robots. Robots have been proved commendable in working together with patients and practitioners to achieve the common goal of well-being. Apart from high power to weight ratio and accuracy, robots are expected to be safe and flexible. At The University of Auckland we had earlier developed robots for ankle joint and lower limb rehabilitation using McKibben pneumatic muscle actuators (PMA) which were safe and flexible. However, these actuators had larger response time and hysteresis apart from compromised actuation limits. As a result of our further research we have been able to develop inhouse pleated PMA (PPMA) in our laboratory which show improved response time with low hysteresis. The newly developed actuators have larger actuation as well. In order to cope with the non-linear and transient nature of these actuators, this paper further proposes a new Artificial Neural Network (ANN) based approach. To optimize ANN model parameters a hybrid approach combing back propagation (BP) algorithm with Modified Genetic Algorithm (MGA) is developed. Results show that the hybrid approach is able to model the PPMA behaviour closely.

Original languageEnglish
Title of host publicationProceedings of 2012 8th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications, MESA 2012
Pages190-195
Number of pages6
DOIs
Publication statusPublished - 2012
Externally publishedYes
Event2012 8th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications, MESA 2012 - Suzhou, China
Duration: Jul 8 2012Jul 10 2012

Other

Other2012 8th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications, MESA 2012
CountryChina
CitySuzhou
Period7/8/127/10/12

Fingerprint

Pneumatics
Muscle
Actuators
Robots
Neural networks
Patient rehabilitation
Hysteresis
Backpropagation algorithms
Genetic algorithms

Keywords

  • Artificial Neural Networks
  • Dynamic Modelling
  • Pleated Pneumatic Muscle Actuators

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Mechanical Engineering

Cite this

Jamwal, P. K., & Xie, S. Q. (2012). Artificial Neural Network based dynamic modelling of indigenous pneumatic muscle actuators. In Proceedings of 2012 8th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications, MESA 2012 (pp. 190-195). [6275560] https://doi.org/10.1109/MESA.2012.6275560

Artificial Neural Network based dynamic modelling of indigenous pneumatic muscle actuators. / Jamwal, Prashant K.; Xie, Sheng Quan.

Proceedings of 2012 8th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications, MESA 2012. 2012. p. 190-195 6275560.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Jamwal, PK & Xie, SQ 2012, Artificial Neural Network based dynamic modelling of indigenous pneumatic muscle actuators. in Proceedings of 2012 8th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications, MESA 2012., 6275560, pp. 190-195, 2012 8th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications, MESA 2012, Suzhou, China, 7/8/12. https://doi.org/10.1109/MESA.2012.6275560
Jamwal PK, Xie SQ. Artificial Neural Network based dynamic modelling of indigenous pneumatic muscle actuators. In Proceedings of 2012 8th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications, MESA 2012. 2012. p. 190-195. 6275560 https://doi.org/10.1109/MESA.2012.6275560
Jamwal, Prashant K. ; Xie, Sheng Quan. / Artificial Neural Network based dynamic modelling of indigenous pneumatic muscle actuators. Proceedings of 2012 8th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications, MESA 2012. 2012. pp. 190-195
@inproceedings{4ae4ff07f0f1497bbfc407366ebfd87e,
title = "Artificial Neural Network based dynamic modelling of indigenous pneumatic muscle actuators",
abstract = "Robots are increasingly becoming popular medical devices, helping surgeons and practitioners as surgical, rehabilitation or service robots. Robots have been proved commendable in working together with patients and practitioners to achieve the common goal of well-being. Apart from high power to weight ratio and accuracy, robots are expected to be safe and flexible. At The University of Auckland we had earlier developed robots for ankle joint and lower limb rehabilitation using McKibben pneumatic muscle actuators (PMA) which were safe and flexible. However, these actuators had larger response time and hysteresis apart from compromised actuation limits. As a result of our further research we have been able to develop inhouse pleated PMA (PPMA) in our laboratory which show improved response time with low hysteresis. The newly developed actuators have larger actuation as well. In order to cope with the non-linear and transient nature of these actuators, this paper further proposes a new Artificial Neural Network (ANN) based approach. To optimize ANN model parameters a hybrid approach combing back propagation (BP) algorithm with Modified Genetic Algorithm (MGA) is developed. Results show that the hybrid approach is able to model the PPMA behaviour closely.",
keywords = "Artificial Neural Networks, Dynamic Modelling, Pleated Pneumatic Muscle Actuators",
author = "Jamwal, {Prashant K.} and Xie, {Sheng Quan}",
year = "2012",
doi = "10.1109/MESA.2012.6275560",
language = "English",
isbn = "9781467323475",
pages = "190--195",
booktitle = "Proceedings of 2012 8th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications, MESA 2012",

}

TY - GEN

T1 - Artificial Neural Network based dynamic modelling of indigenous pneumatic muscle actuators

AU - Jamwal, Prashant K.

AU - Xie, Sheng Quan

PY - 2012

Y1 - 2012

N2 - Robots are increasingly becoming popular medical devices, helping surgeons and practitioners as surgical, rehabilitation or service robots. Robots have been proved commendable in working together with patients and practitioners to achieve the common goal of well-being. Apart from high power to weight ratio and accuracy, robots are expected to be safe and flexible. At The University of Auckland we had earlier developed robots for ankle joint and lower limb rehabilitation using McKibben pneumatic muscle actuators (PMA) which were safe and flexible. However, these actuators had larger response time and hysteresis apart from compromised actuation limits. As a result of our further research we have been able to develop inhouse pleated PMA (PPMA) in our laboratory which show improved response time with low hysteresis. The newly developed actuators have larger actuation as well. In order to cope with the non-linear and transient nature of these actuators, this paper further proposes a new Artificial Neural Network (ANN) based approach. To optimize ANN model parameters a hybrid approach combing back propagation (BP) algorithm with Modified Genetic Algorithm (MGA) is developed. Results show that the hybrid approach is able to model the PPMA behaviour closely.

AB - Robots are increasingly becoming popular medical devices, helping surgeons and practitioners as surgical, rehabilitation or service robots. Robots have been proved commendable in working together with patients and practitioners to achieve the common goal of well-being. Apart from high power to weight ratio and accuracy, robots are expected to be safe and flexible. At The University of Auckland we had earlier developed robots for ankle joint and lower limb rehabilitation using McKibben pneumatic muscle actuators (PMA) which were safe and flexible. However, these actuators had larger response time and hysteresis apart from compromised actuation limits. As a result of our further research we have been able to develop inhouse pleated PMA (PPMA) in our laboratory which show improved response time with low hysteresis. The newly developed actuators have larger actuation as well. In order to cope with the non-linear and transient nature of these actuators, this paper further proposes a new Artificial Neural Network (ANN) based approach. To optimize ANN model parameters a hybrid approach combing back propagation (BP) algorithm with Modified Genetic Algorithm (MGA) is developed. Results show that the hybrid approach is able to model the PPMA behaviour closely.

KW - Artificial Neural Networks

KW - Dynamic Modelling

KW - Pleated Pneumatic Muscle Actuators

UR - http://www.scopus.com/inward/record.url?scp=84867453551&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84867453551&partnerID=8YFLogxK

U2 - 10.1109/MESA.2012.6275560

DO - 10.1109/MESA.2012.6275560

M3 - Conference contribution

SN - 9781467323475

SP - 190

EP - 195

BT - Proceedings of 2012 8th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications, MESA 2012

ER -