Directional Forgetting for Stable Co-Adaptation in Myoelectric Control

Research output: Contribution to journalArticleScientificpeer-review

Standard

Directional Forgetting for Stable Co-Adaptation in Myoelectric Control. / Yeung, Dennis; Farina, Dario; Vujaklija, Ivan.

In: Sensors (Basel, Switzerland), Vol. 19, No. 9, 2203, 01.05.2019.

Research output: Contribution to journalArticleScientificpeer-review

Harvard

APA

Vancouver

Author

Bibtex - Download

@article{c5d3c866d8104233b64c1b2ffe30eba2,
title = "Directional Forgetting for Stable Co-Adaptation in Myoelectric Control",
abstract = "Conventional myoelectric controllers provide a mapping between electromyographic signals and prosthetic functions. However, due to a number of instabilities continuously challenging this process, an initial mapping may require an extended calibration phase with long periods of user-training in order to ensure satisfactory performance. Recently, studies on co-adaptation have highlighted the benefits of concurrent user learning and machine adaptation where systems can cope with deficiencies in the initial model by learning from newly acquired data. However, the success remains highly dependent on careful weighting of these new data. In this study, we proposed a function driven directional forgetting approach to the recursive least-squares algorithm as opposed to the classic exponential forgetting scheme. By only discounting past information in the same direction of the new data, local corrections to the mapping would induce less distortion to other regions. To validate the approach, subjects performed a set of real-time myoelectric tasks over a range of forgetting factors. Results show that directional forgetting with a forgetting factor of 0.995 outperformed exponential forgetting as well as unassisted user learning. Moreover, myoelectric control remained stable after adaptation with directional forgetting over a range of forgetting factors. These results indicate that a directional approach to discounting past training data can improve performance and alleviate sensitivities to parameter selection in recursive adaptation algorithms.",
keywords = "co-adaptation, directional forgetting, electromyography, myoelectric control, upper-limb prostheses",
author = "Dennis Yeung and Dario Farina and Ivan Vujaklija",
note = "| openaire: EC/H2020/687795/EU//INPUT",
year = "2019",
month = "5",
day = "1",
doi = "10.3390/s19092203",
language = "English",
volume = "19",
journal = "Sensors",
issn = "1424-8220",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "9",

}

RIS - Download

TY - JOUR

T1 - Directional Forgetting for Stable Co-Adaptation in Myoelectric Control

AU - Yeung, Dennis

AU - Farina, Dario

AU - Vujaklija, Ivan

N1 - | openaire: EC/H2020/687795/EU//INPUT

PY - 2019/5/1

Y1 - 2019/5/1

N2 - Conventional myoelectric controllers provide a mapping between electromyographic signals and prosthetic functions. However, due to a number of instabilities continuously challenging this process, an initial mapping may require an extended calibration phase with long periods of user-training in order to ensure satisfactory performance. Recently, studies on co-adaptation have highlighted the benefits of concurrent user learning and machine adaptation where systems can cope with deficiencies in the initial model by learning from newly acquired data. However, the success remains highly dependent on careful weighting of these new data. In this study, we proposed a function driven directional forgetting approach to the recursive least-squares algorithm as opposed to the classic exponential forgetting scheme. By only discounting past information in the same direction of the new data, local corrections to the mapping would induce less distortion to other regions. To validate the approach, subjects performed a set of real-time myoelectric tasks over a range of forgetting factors. Results show that directional forgetting with a forgetting factor of 0.995 outperformed exponential forgetting as well as unassisted user learning. Moreover, myoelectric control remained stable after adaptation with directional forgetting over a range of forgetting factors. These results indicate that a directional approach to discounting past training data can improve performance and alleviate sensitivities to parameter selection in recursive adaptation algorithms.

AB - Conventional myoelectric controllers provide a mapping between electromyographic signals and prosthetic functions. However, due to a number of instabilities continuously challenging this process, an initial mapping may require an extended calibration phase with long periods of user-training in order to ensure satisfactory performance. Recently, studies on co-adaptation have highlighted the benefits of concurrent user learning and machine adaptation where systems can cope with deficiencies in the initial model by learning from newly acquired data. However, the success remains highly dependent on careful weighting of these new data. In this study, we proposed a function driven directional forgetting approach to the recursive least-squares algorithm as opposed to the classic exponential forgetting scheme. By only discounting past information in the same direction of the new data, local corrections to the mapping would induce less distortion to other regions. To validate the approach, subjects performed a set of real-time myoelectric tasks over a range of forgetting factors. Results show that directional forgetting with a forgetting factor of 0.995 outperformed exponential forgetting as well as unassisted user learning. Moreover, myoelectric control remained stable after adaptation with directional forgetting over a range of forgetting factors. These results indicate that a directional approach to discounting past training data can improve performance and alleviate sensitivities to parameter selection in recursive adaptation algorithms.

KW - co-adaptation

KW - directional forgetting

KW - electromyography

KW - myoelectric control

KW - upper-limb prostheses

UR - http://www.scopus.com/inward/record.url?scp=85066852774&partnerID=8YFLogxK

U2 - 10.3390/s19092203

DO - 10.3390/s19092203

M3 - Article

VL - 19

JO - Sensors

JF - Sensors

SN - 1424-8220

IS - 9

M1 - 2203

ER -

ID: 34640946