Using Different Combinations of Body-Mounted IMU Sensors to Estimate Speed of Horses-A Machine Learning Approach

Hamed Darbandi; F.M. Serra Braganca; Berend Jan van der Zwaag; john P. voskamp; Annik  Imogen Gmel; Eyrun Halla Haraldsdóttir; Paul Havinga

doi:10.3390/s21030798

Using Different Combinations of Body-Mounted IMU Sensors to Estimate Speed of Horses-A Machine Learning Approach

Hamed Darbandi, F.M. Serra Braganca, Berend Jan van der Zwaag, john P. voskamp, Annik Imogen Gmel, Eyrun Halla Haraldsdóttir, Paul Havinga

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

Abstract

Speed is an essential parameter in biomechanical analysis and general locomotion research. It is possible to estimate the speed using global positioning systems (GPS) or inertial measurement units (IMUs). However, GPS requires a consistent signal connection to satellites, and errors accumulate during IMU signals integration. In an attempt to overcome these issues, we have investigated the possibility of estimating the horse speed by developing machine learning (ML) models using the signals from seven body-mounted IMUs. Since motion patterns extracted from IMU signals are different between breeds and gaits, we trained the models based on data from 40 Icelandic and Franches-Montagnes horses during walk, trot, tölt, pace, and canter. In addition, we studied the estimation accuracy between IMU locations on the body (sacrum, withers, head, and limbs). The models were evaluated per gait and were compared between ML algorithms and IMU location. The model yielded the highest estimation accuracy of speed (RMSE = 0.25 m/s) within equine and most of human speed estimation literature. In conclusion, highly accurate horse speed estimation models, independent of IMU(s) location on-body and gait, were developed using ML.

Original language	English
Article number	798
Pages (from-to)	1-12
Number of pages	12
Journal	ACS Sensors
Volume	21
Issue number	3
DOIs	https://doi.org/10.3390/s21030798
Publication status	Published - 1 Feb 2021

Bibliographical note

Funding Information:
This study was partly funded by EFRO OP-Oost (project ?Paardensprong?), and the Swiss Federal Office for Agriculture (contract number 627001325). The authors wish to thank Sigridur Bjornsdottir and V?kingur Gunnarsson from H?lar University College, Elin Hernlund and Marie Rhodin from Swedish University of Agricultural Sciences, Michael Weishaupt and Marie-Theres Dittmann from University of Zurich, and all the animal caretakers that helped with the data collection.

Funding Information:
Funding: This study was partly funded by EFRO OP-Oost (project “Paardensprong”), and the Swiss Federal Office for Agriculture (contract number 627001325).

Publisher Copyright:
© 2021 by the authors. Licensee MDPI, Basel, Switzerland.

Copyright:
Copyright 2021 Elsevier B.V., All rights reserved.

Keywords

Breed
Feature extraction
Gait
Inertial measurement unit
Machine learning

Access to Document

10.3390/s21030798Licence: CC BY

sensors-21-00798Final published version, 15.5 MB

Cite this

@article{6faa34c55bb9467192c5b932f62e8442,

title = "Using Different Combinations of Body-Mounted IMU Sensors to Estimate Speed of Horses-A Machine Learning Approach",

abstract = "Speed is an essential parameter in biomechanical analysis and general locomotion research. It is possible to estimate the speed using global positioning systems (GPS) or inertial measurement units (IMUs). However, GPS requires a consistent signal connection to satellites, and errors accumulate during IMU signals integration. In an attempt to overcome these issues, we have investigated the possibility of estimating the horse speed by developing machine learning (ML) models using the signals from seven body-mounted IMUs. Since motion patterns extracted from IMU signals are different between breeds and gaits, we trained the models based on data from 40 Icelandic and Franches-Montagnes horses during walk, trot, t{\"o}lt, pace, and canter. In addition, we studied the estimation accuracy between IMU locations on the body (sacrum, withers, head, and limbs). The models were evaluated per gait and were compared between ML algorithms and IMU location. The model yielded the highest estimation accuracy of speed (RMSE = 0.25 m/s) within equine and most of human speed estimation literature. In conclusion, highly accurate horse speed estimation models, independent of IMU(s) location on-body and gait, were developed using ML.",

keywords = "Breed, Feature extraction, Gait, Inertial measurement unit, Machine learning",

author = "Hamed Darbandi and \{Serra Braganca\}, F.M. and \{van der Zwaag\}, \{Berend Jan\} and voskamp, \{john P.\} and \{Imogen Gmel\}, Annik and \{Halla Haraldsd{\'o}ttir\}, Eyrun and Paul Havinga",

note = "Funding Information: This study was partly funded by EFRO OP-Oost (project ?Paardensprong?), and the Swiss Federal Office for Agriculture (contract number 627001325). The authors wish to thank Sigridur Bjornsdottir and V?kingur Gunnarsson from H?lar University College, Elin Hernlund and Marie Rhodin from Swedish University of Agricultural Sciences, Michael Weishaupt and Marie-Theres Dittmann from University of Zurich, and all the animal caretakers that helped with the data collection. Funding Information: Funding: This study was partly funded by EFRO OP-Oost (project “Paardensprong”), and the Swiss Federal Office for Agriculture (contract number 627001325). Publisher Copyright: {\textcopyright} 2021 by the authors. Licensee MDPI, Basel, Switzerland. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.",

year = "2021",

month = feb,

day = "1",

doi = "10.3390/s21030798",

language = "English",

volume = "21",

pages = "1--12",

journal = "ACS Sensors",

issn = "1424-3210",

publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",

number = "3",

}

TY - JOUR

T1 - Using Different Combinations of Body-Mounted IMU Sensors to Estimate Speed of Horses-A Machine Learning Approach

AU - Darbandi, Hamed

AU - Serra Braganca, F.M.

AU - van der Zwaag, Berend Jan

AU - voskamp, john P.

AU - Imogen Gmel, Annik

AU - Halla Haraldsdóttir, Eyrun

AU - Havinga, Paul

N1 - Funding Information: This study was partly funded by EFRO OP-Oost (project ?Paardensprong?), and the Swiss Federal Office for Agriculture (contract number 627001325). The authors wish to thank Sigridur Bjornsdottir and V?kingur Gunnarsson from H?lar University College, Elin Hernlund and Marie Rhodin from Swedish University of Agricultural Sciences, Michael Weishaupt and Marie-Theres Dittmann from University of Zurich, and all the animal caretakers that helped with the data collection. Funding Information: Funding: This study was partly funded by EFRO OP-Oost (project “Paardensprong”), and the Swiss Federal Office for Agriculture (contract number 627001325). Publisher Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.

PY - 2021/2/1

Y1 - 2021/2/1

N2 - Speed is an essential parameter in biomechanical analysis and general locomotion research. It is possible to estimate the speed using global positioning systems (GPS) or inertial measurement units (IMUs). However, GPS requires a consistent signal connection to satellites, and errors accumulate during IMU signals integration. In an attempt to overcome these issues, we have investigated the possibility of estimating the horse speed by developing machine learning (ML) models using the signals from seven body-mounted IMUs. Since motion patterns extracted from IMU signals are different between breeds and gaits, we trained the models based on data from 40 Icelandic and Franches-Montagnes horses during walk, trot, tölt, pace, and canter. In addition, we studied the estimation accuracy between IMU locations on the body (sacrum, withers, head, and limbs). The models were evaluated per gait and were compared between ML algorithms and IMU location. The model yielded the highest estimation accuracy of speed (RMSE = 0.25 m/s) within equine and most of human speed estimation literature. In conclusion, highly accurate horse speed estimation models, independent of IMU(s) location on-body and gait, were developed using ML.

AB - Speed is an essential parameter in biomechanical analysis and general locomotion research. It is possible to estimate the speed using global positioning systems (GPS) or inertial measurement units (IMUs). However, GPS requires a consistent signal connection to satellites, and errors accumulate during IMU signals integration. In an attempt to overcome these issues, we have investigated the possibility of estimating the horse speed by developing machine learning (ML) models using the signals from seven body-mounted IMUs. Since motion patterns extracted from IMU signals are different between breeds and gaits, we trained the models based on data from 40 Icelandic and Franches-Montagnes horses during walk, trot, tölt, pace, and canter. In addition, we studied the estimation accuracy between IMU locations on the body (sacrum, withers, head, and limbs). The models were evaluated per gait and were compared between ML algorithms and IMU location. The model yielded the highest estimation accuracy of speed (RMSE = 0.25 m/s) within equine and most of human speed estimation literature. In conclusion, highly accurate horse speed estimation models, independent of IMU(s) location on-body and gait, were developed using ML.

KW - Breed

KW - Feature extraction

KW - Gait

KW - Inertial measurement unit

KW - Machine learning

UR - https://www.scopus.com/pages/publications/85099875721

U2 - 10.3390/s21030798

DO - 10.3390/s21030798

M3 - Article

C2 - 33530288

SN - 1424-3210

VL - 21

SP - 1

EP - 12

JO - ACS Sensors

JF - ACS Sensors

IS - 3

M1 - 798

ER -

Using Different Combinations of Body-Mounted IMU Sensors to Estimate Speed of Horses-A Machine Learning Approach

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this