Details
| Originalsprache | Englisch |
|---|---|
| Aufsatznummer | 7004904 |
| Fachzeitschrift | IEEE Sensors Letters |
| Jahrgang | 7 |
| Ausgabenummer | 10 |
| Publikationsstatus | Veröffentlicht - 21 Aug. 2023 |
Abstract
Inertial measurement units (IMUs) are used for inertial motion tracking (IMT) in a growing number of applications as sensor fusion methods are being advanced in three directions: magnetometer-free IMT methods that eliminate the effect of magnetic disturbances; sparse IMT approaches that lead to reduced setup complexity; and automatic self-calibration of sensor-to-segment positions or orientations. In this letter, we propose an approach that combines all three achievements and, for the first time, enables plug-and-play, magnetometer-free, and sparse IMT. This is accomplished by training a recurrent neural-network-based observer (RNNo) on just-in-time generated simulated motion data of kinematic chains. We demonstrate that domain-specific training data augmentations lead to a trained RNNo which zero shot generalizes to previously unseen experimental data and, thus, overcomes the sim-to-real gap. The trained RNNo achieves a tracking error of < 4 degrees when estimating the relative pose of a three-segment kinematic chain with two hinge joints. The proposed method offers a novel simulation-data-driven approach for solving complex sparse sensing problems while assuring robust and plug-and-play generalizability to experimental data.
ASJC Scopus Sachgebiete
- Physik und Astronomie (insg.)
- Instrumentierung
- Ingenieurwesen (insg.)
- Elektrotechnik und Elektronik
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
in: IEEE Sensors Letters, Jahrgang 7, Nr. 10, 7004904, 21.08.2023.
Publikation: Beitrag in Fachzeitschrift › Artikel › Forschung › Peer-Review
}
TY - JOUR
T1 - Plug-and-Play Sparse Inertial Motion Tracking With Sim-to-Real Transfer
AU - Bachhuber, Simon
AU - Lehmann, Dustin
AU - Dorschky, Eva
AU - Koelewijn, Anne D.
AU - Seel, Thomas
AU - Weygers, Ive
PY - 2023/8/21
Y1 - 2023/8/21
N2 - Inertial measurement units (IMUs) are used for inertial motion tracking (IMT) in a growing number of applications as sensor fusion methods are being advanced in three directions: magnetometer-free IMT methods that eliminate the effect of magnetic disturbances; sparse IMT approaches that lead to reduced setup complexity; and automatic self-calibration of sensor-to-segment positions or orientations. In this letter, we propose an approach that combines all three achievements and, for the first time, enables plug-and-play, magnetometer-free, and sparse IMT. This is accomplished by training a recurrent neural-network-based observer (RNNo) on just-in-time generated simulated motion data of kinematic chains. We demonstrate that domain-specific training data augmentations lead to a trained RNNo which zero shot generalizes to previously unseen experimental data and, thus, overcomes the sim-to-real gap. The trained RNNo achieves a tracking error of < 4 degrees when estimating the relative pose of a three-segment kinematic chain with two hinge joints. The proposed method offers a novel simulation-data-driven approach for solving complex sparse sensing problems while assuring robust and plug-and-play generalizability to experimental data.
AB - Inertial measurement units (IMUs) are used for inertial motion tracking (IMT) in a growing number of applications as sensor fusion methods are being advanced in three directions: magnetometer-free IMT methods that eliminate the effect of magnetic disturbances; sparse IMT approaches that lead to reduced setup complexity; and automatic self-calibration of sensor-to-segment positions or orientations. In this letter, we propose an approach that combines all three achievements and, for the first time, enables plug-and-play, magnetometer-free, and sparse IMT. This is accomplished by training a recurrent neural-network-based observer (RNNo) on just-in-time generated simulated motion data of kinematic chains. We demonstrate that domain-specific training data augmentations lead to a trained RNNo which zero shot generalizes to previously unseen experimental data and, thus, overcomes the sim-to-real gap. The trained RNNo achieves a tracking error of < 4 degrees when estimating the relative pose of a three-segment kinematic chain with two hinge joints. The proposed method offers a novel simulation-data-driven approach for solving complex sparse sensing problems while assuring robust and plug-and-play generalizability to experimental data.
KW - inertial measurement units (IMUs)
KW - magnetometer-free
KW - recurrent neural networks
KW - sensor fusion
KW - Sensor signal processing
KW - sparse sensing
UR - http://www.scopus.com/inward/record.url?scp=85168741697&partnerID=8YFLogxK
U2 - 10.1109/LSENS.2023.3307122
DO - 10.1109/LSENS.2023.3307122
M3 - Article
AN - SCOPUS:85168741697
VL - 7
JO - IEEE Sensors Letters
JF - IEEE Sensors Letters
IS - 10
M1 - 7004904
ER -