Predicted information gain and convolutional neural network for prediction of gait periods using a wearable sensors network

Uriel Martinez Hernandez; Adrian Rubio-Solis

doi:10.1109/RO-MAN50785.2021.9515395

Predicted information gain and convolutional neural network for prediction of gait periods using a wearable sensors network

Uriel Martinez Hernandez, Adrian Rubio-Solis

University College London

Research output: Chapter or section in a book/report/conference proceeding › Chapter in a published conference proceeding

70 Downloads (Pure)

Abstract

This work presents a method for recognition of walking activities and prediction of gait periods using wearable sensors. First, a Convolutional Neural Network (CNN) is used to recognise the walking activity and gait period. Second, the output of the CNN is used by a Predicted Information Gain (PIG) method to predict the next most probable gait period while walking. The output of these two processes are combined to adapt the recognition accuracy of the system. This adaptive combination allows us to achieve an optimal recognition accuracy over time. The validation of this work is performed with an array of wearable sensors for the recognition of level-ground walking, ramp ascent and ramp descent, and prediction of gait periods. The results show that the proposed system can achieve accuracies of 100% and 99.9% for recognition of walking activity and gait period, respectively. These results show the benefit of having a system capable of predicting or anticipating the next information or event over time. Overall, this approach offers a method for accurate activity recognition, which is a key process for the development of wearable robots capable of safely assist humans in activities of daily living

Original language	English
Title of host publication	2021 30th IEEE International Conference on Robot and Human Interactive Communication, RO-MAN 2021
Publisher	IEEE
Pages	1132-1137
Number of pages	6
ISBN (Electronic)	978-1-6654-0492-1
ISBN (Print)	978-1-6654-4637-2
DOIs	https://doi.org/10.1109/RO-MAN50785.2021.9515395
Publication status	Published - 31 Dec 2021

Publication series

Name	2021 30th IEEE International Conference on Robot and Human Interactive Communication, RO-MAN 2021

Bibliographical note

Funding Information:
*This work was supported by the Royal Society Research Grants for the ‘Touching and feeling the immersive world’ project (RGS/R2/192346). Uriel is with the inte-R-action Lab and the Centre for Autonomous Robotics (CENTAUR), Department of Electronic and Electrical Engineering, University of Bath, UK [email protected] Adrian is with the Faculty of Mathematical and Physical Sciences, University College London, London, UK [email protected]

Publisher Copyright:
© 2021 IEEE.

ASJC Scopus subject areas

Human-Computer Interaction
Communication
Artificial Intelligence

Access to Document

10.1109/RO-MAN50785.2021.9515395

RO-MAN_2021_camera_ready.PDF
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.
Accepted author manuscript, 2.96 MB

Cite this

Martinez Hernandez, U., & Rubio-Solis, A. (2021). Predicted information gain and convolutional neural network for prediction of gait periods using a wearable sensors network. In 2021 30th IEEE International Conference on Robot and Human Interactive Communication, RO-MAN 2021 (pp. 1132-1137). Article 9515395 (2021 30th IEEE International Conference on Robot and Human Interactive Communication, RO-MAN 2021). IEEE. https://doi.org/10.1109/RO-MAN50785.2021.9515395

Predicted information gain and convolutional neural network for prediction of gait periods using a wearable sensors network. / Martinez Hernandez, Uriel; Rubio-Solis, Adrian.
2021 30th IEEE International Conference on Robot and Human Interactive Communication, RO-MAN 2021. IEEE, 2021. p. 1132-1137 9515395 (2021 30th IEEE International Conference on Robot and Human Interactive Communication, RO-MAN 2021).

Research output: Chapter or section in a book/report/conference proceeding › Chapter in a published conference proceeding

Martinez Hernandez, U & Rubio-Solis, A 2021, Predicted information gain and convolutional neural network for prediction of gait periods using a wearable sensors network. in 2021 30th IEEE International Conference on Robot and Human Interactive Communication, RO-MAN 2021., 9515395, 2021 30th IEEE International Conference on Robot and Human Interactive Communication, RO-MAN 2021, IEEE, pp. 1132-1137. https://doi.org/10.1109/RO-MAN50785.2021.9515395

Martinez Hernandez U, Rubio-Solis A. Predicted information gain and convolutional neural network for prediction of gait periods using a wearable sensors network. In 2021 30th IEEE International Conference on Robot and Human Interactive Communication, RO-MAN 2021. IEEE. 2021. p. 1132-1137. 9515395. (2021 30th IEEE International Conference on Robot and Human Interactive Communication, RO-MAN 2021). Epub 2021 Aug 23. doi: 10.1109/RO-MAN50785.2021.9515395

Martinez Hernandez, Uriel ; Rubio-Solis, Adrian. / Predicted information gain and convolutional neural network for prediction of gait periods using a wearable sensors network. 2021 30th IEEE International Conference on Robot and Human Interactive Communication, RO-MAN 2021. IEEE, 2021. pp. 1132-1137 (2021 30th IEEE International Conference on Robot and Human Interactive Communication, RO-MAN 2021).

@inproceedings{83642db6a0cb43fbb6bf763f2d9a431a,

title = "Predicted information gain and convolutional neural network for prediction of gait periods using a wearable sensors network",

abstract = "This work presents a method for recognition of walking activities and prediction of gait periods using wearable sensors. First, a Convolutional Neural Network (CNN) is used to recognise the walking activity and gait period. Second, the output of the CNN is used by a Predicted Information Gain (PIG) method to predict the next most probable gait period while walking. The output of these two processes are combined to adapt the recognition accuracy of the system. This adaptive combination allows us to achieve an optimal recognition accuracy over time. The validation of this work is performed with an array of wearable sensors for the recognition of level-ground walking, ramp ascent and ramp descent, and prediction of gait periods. The results show that the proposed system can achieve accuracies of 100% and 99.9% for recognition of walking activity and gait period, respectively. These results show the benefit of having a system capable of predicting or anticipating the next information or event over time. Overall, this approach offers a method for accurate activity recognition, which is a key process for the development of wearable robots capable of safely assist humans in activities of daily living",

author = "{Martinez Hernandez}, Uriel and Adrian Rubio-Solis",

note = "Funding Information: *This work was supported by the Royal Society Research Grants for the {\textquoteleft}Touching and feeling the immersive world{\textquoteright} project (RGS/R2/192346). Uriel is with the inte-R-action Lab and the Centre for Autonomous Robotics (CENTAUR), Department of Electronic and Electrical Engineering, University of Bath, UK [email protected] Adrian is with the Faculty of Mathematical and Physical Sciences, University College London, London, UK [email protected] Publisher Copyright: {\textcopyright} 2021 IEEE.",

year = "2021",

month = dec,

day = "31",

doi = "10.1109/RO-MAN50785.2021.9515395",

language = "English",

isbn = "978-1-6654-4637-2",

series = "2021 30th IEEE International Conference on Robot and Human Interactive Communication, RO-MAN 2021",

publisher = "IEEE",

pages = "1132--1137",

booktitle = "2021 30th IEEE International Conference on Robot and Human Interactive Communication, RO-MAN 2021",

address = "USA United States",

}

TY - GEN

T1 - Predicted information gain and convolutional neural network for prediction of gait periods using a wearable sensors network

AU - Martinez Hernandez, Uriel

AU - Rubio-Solis, Adrian

N1 - Funding Information: *This work was supported by the Royal Society Research Grants for the ‘Touching and feeling the immersive world’ project (RGS/R2/192346). Uriel is with the inte-R-action Lab and the Centre for Autonomous Robotics (CENTAUR), Department of Electronic and Electrical Engineering, University of Bath, UK [email protected] Adrian is with the Faculty of Mathematical and Physical Sciences, University College London, London, UK [email protected] Publisher Copyright: © 2021 IEEE.

PY - 2021/12/31

Y1 - 2021/12/31

N2 - This work presents a method for recognition of walking activities and prediction of gait periods using wearable sensors. First, a Convolutional Neural Network (CNN) is used to recognise the walking activity and gait period. Second, the output of the CNN is used by a Predicted Information Gain (PIG) method to predict the next most probable gait period while walking. The output of these two processes are combined to adapt the recognition accuracy of the system. This adaptive combination allows us to achieve an optimal recognition accuracy over time. The validation of this work is performed with an array of wearable sensors for the recognition of level-ground walking, ramp ascent and ramp descent, and prediction of gait periods. The results show that the proposed system can achieve accuracies of 100% and 99.9% for recognition of walking activity and gait period, respectively. These results show the benefit of having a system capable of predicting or anticipating the next information or event over time. Overall, this approach offers a method for accurate activity recognition, which is a key process for the development of wearable robots capable of safely assist humans in activities of daily living

AB - This work presents a method for recognition of walking activities and prediction of gait periods using wearable sensors. First, a Convolutional Neural Network (CNN) is used to recognise the walking activity and gait period. Second, the output of the CNN is used by a Predicted Information Gain (PIG) method to predict the next most probable gait period while walking. The output of these two processes are combined to adapt the recognition accuracy of the system. This adaptive combination allows us to achieve an optimal recognition accuracy over time. The validation of this work is performed with an array of wearable sensors for the recognition of level-ground walking, ramp ascent and ramp descent, and prediction of gait periods. The results show that the proposed system can achieve accuracies of 100% and 99.9% for recognition of walking activity and gait period, respectively. These results show the benefit of having a system capable of predicting or anticipating the next information or event over time. Overall, this approach offers a method for accurate activity recognition, which is a key process for the development of wearable robots capable of safely assist humans in activities of daily living

UR - http://www.scopus.com/inward/record.url?scp=85115054723&partnerID=8YFLogxK

U2 - 10.1109/RO-MAN50785.2021.9515395

DO - 10.1109/RO-MAN50785.2021.9515395

M3 - Chapter in a published conference proceeding

SN - 978-1-6654-4637-2

T3 - 2021 30th IEEE International Conference on Robot and Human Interactive Communication, RO-MAN 2021

SP - 1132

EP - 1137

BT - 2021 30th IEEE International Conference on Robot and Human Interactive Communication, RO-MAN 2021

PB - IEEE

ER -

Predicted information gain and convolutional neural network for prediction of gait periods using a wearable sensors network

Abstract

Publication series

Bibliographical note

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this