Název:	Deep Learning Methods for Speed Estimation of Bipedal Motion from Wearable IMU Sensors
Autoři:	Justa, Josef Šmídl, Václav Hamáček, Aleš
Citace zdrojového dokumentu:	JUSTA, J. ŠMÍDL, V. HAMÁČEK, A. Deep Learning Methods for Speed Estimation of Bipedal Motion from Wearable IMU Sensors. SENSORS, 2022, roč. 22, č. 10, s. 1-16. ISSN: 1424-8220
Datum vydání:	2022
Nakladatel:	MDPI
Typ dokumentu:	článek article
URI:	2-s2.0-85130456586 http://hdl.handle.net/11025/49273
ISSN:	1424-8220
Klíčová slova v dalším jazyce:	motion speed estimation;inertial measurement unit;deep learning;walking speed;autoencoder architecture
Abstrakt v dalším 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.
Práva:	© authors
Vyskytuje se v kolekcích:	Články / Articles (RICE) Články / Articles (KET) OBD

Všechny záznamy v DSpace jsou chráněny autorskými právy, všechna práva vyhrazena.