Deep Learning Methods for Speed Estimation of Bipedal Motion from Wearable IMU Sensors
Identifikátory výsledku
Kód výsledku v IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23220%2F22%3A43965368" target="_blank" >RIV/49777513:23220/22:43965368 - isvavai.cz</a>
Výsledek na webu
<a href="https://www.mdpi.com/1424-8220/22/10/3865/htm" target="_blank" >https://www.mdpi.com/1424-8220/22/10/3865/htm</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.3390/s22103865" target="_blank" >10.3390/s22103865</a>
Alternativní jazyky
Jazyk výsledku
angličtina
Název v původním jazyce
Deep Learning Methods for Speed Estimation of Bipedal Motion from Wearable IMU Sensors
Popis výsledku v původním jazyce
The estimation of the speed of human motion from wearable IMU sensors is required in applications such as pedestrian dead reckoning. In this paper, we test deep learning methods for the prediction of the motion speed from raw readings of a low-cost IMU sensor. Each subject was observed using three sensors at the shoe, shin, and thigh. We show that existing general-purpose architectures outperform classical feature-based approaches and propose a novel architecture tailored for this task. The proposed architecture is based on a semi-supervised variational auto-encoder structure with innovated decoder in the form of a dense layer with a sinusoidal activation function. The proposed architecture achieved the lowest average error on the test data. Analysis of sensor placement reveals that the best location for the sensor is the shoe. Significant accuracy gain was observed when all three sensors were available. All data acquired in this experiment and the code of the estimation methods are available for download.
Název v anglickém jazyce
Deep Learning Methods for Speed Estimation of Bipedal Motion from Wearable IMU Sensors
Popis výsledku anglicky
The estimation of the speed of human motion from wearable IMU sensors is required in applications such as pedestrian dead reckoning. In this paper, we test deep learning methods for the prediction of the motion speed from raw readings of a low-cost IMU sensor. Each subject was observed using three sensors at the shoe, shin, and thigh. We show that existing general-purpose architectures outperform classical feature-based approaches and propose a novel architecture tailored for this task. The proposed architecture is based on a semi-supervised variational auto-encoder structure with innovated decoder in the form of a dense layer with a sinusoidal activation function. The proposed architecture achieved the lowest average error on the test data. Analysis of sensor placement reveals that the best location for the sensor is the shoe. Significant accuracy gain was observed when all three sensors were available. All data acquired in this experiment and the code of the estimation methods are available for download.
Klasifikace
Druh
J<sub>imp</sub> - Článek v periodiku v databázi Web of Science
CEP obor
—
OECD FORD obor
20201 - Electrical and electronic engineering
Návaznosti výsledku
Projekt
<a href="/cs/project/FW01010189" target="_blank" >FW01010189: Nástroje virtuální reality pro interaktivní simulátory s pohyblivou plošinou</a><br>
Návaznosti
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)
Ostatní
Rok uplatnění
2022
Kód důvěrnosti údajů
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Údaje specifické pro druh výsledku
Název periodika
SENSORS
ISSN
1424-8220
e-ISSN
1424-8220
Svazek periodika
22
Číslo periodika v rámci svazku
10
Stát vydavatele periodika
CH - Švýcarská konfederace
Počet stran výsledku
16
Strana od-do
1-16
Kód UT WoS článku
000802493900001
EID výsledku v databázi Scopus
2-s2.0-85130456586