Details
| Originalsprache | Englisch |
|---|---|
| Titel des Sammelwerks | 2022 25th International Conference on Information Fusion, FUSION 2022 |
| Herausgeber (Verlag) | Institute of Electrical and Electronics Engineers Inc. |
| ISBN (elektronisch) | 9781737749721 |
| Publikationsstatus | Veröffentlicht - 2022 |
| Extern publiziert | Ja |
| Veranstaltung | 25th International Conference on Information Fusion, FUSION 2022 - Linkoping, Schweden Dauer: 4 Juli 2022 → 7 Juli 2022 |
Publikationsreihe
| Name | 2022 25th International Conference on Information Fusion, FUSION 2022 |
|---|
Abstract
Inertial measurement units are widely used for motion tracking of kinematic chains in numerous applications. While magnetometer-free sensor fusion enables reliably high accuracy in indoor environments and near magnetic disturbances, the use of sparse sensor setups would yield additional advantages in cost, effort, and usability. However, it is unclear which sparse sensor setups can be used to track which motions of which kinematic chains, since observability of the underlying nonlinear dynamics is barely understood to date. We propose a method that utilizes recurrent neural networks (RNNs) and automatically generated training data to assess the observability of the relative pose of kinematic chains in sparse inertial motion tracking (IMT) systems. We apply this method to a range of double-hinge-joint systems that perform fully-exciting random motion. Results show how the degree of observability depends on the kinematic structure and that RNN-based observers can achieve small tracking errors in a large range of sparse and magnetometer-free setups. The proposed methods enable systematic assessment of observability properties in complex nonlinear dynamics and represent a key step toward enabling reliably accurate and non-restrictive IMT solutions.
ASJC Scopus Sachgebiete
- Informatik (insg.)
- Maschinelles Sehen und Mustererkennung
- Informatik (insg.)
- Information systems
- Informatik (insg.)
- Signalverarbeitung
- Entscheidungswissenschaften (insg.)
- Informationssysteme und -management
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
2022 25th International Conference on Information Fusion, FUSION 2022. Institute of Electrical and Electronics Engineers Inc., 2022. (2022 25th International Conference on Information Fusion, FUSION 2022).
Publikation: Beitrag in Buch/Bericht/Sammelwerk/Konferenzband › Aufsatz in Konferenzband › Forschung › Peer-Review
}
TY - GEN
T1 - RNN-based Observability Analysis for Magnetometer-Free Sparse Inertial Motion Tracking
AU - Bachhuber, Simon
AU - Weber, Daniel
AU - Weygers, Ive
AU - Seel, Thomas
N1 - Funding Information: The authors gratefully acknowledge the compute resources and support provided by NHRattheFAU.
PY - 2022
Y1 - 2022
N2 - Inertial measurement units are widely used for motion tracking of kinematic chains in numerous applications. While magnetometer-free sensor fusion enables reliably high accuracy in indoor environments and near magnetic disturbances, the use of sparse sensor setups would yield additional advantages in cost, effort, and usability. However, it is unclear which sparse sensor setups can be used to track which motions of which kinematic chains, since observability of the underlying nonlinear dynamics is barely understood to date. We propose a method that utilizes recurrent neural networks (RNNs) and automatically generated training data to assess the observability of the relative pose of kinematic chains in sparse inertial motion tracking (IMT) systems. We apply this method to a range of double-hinge-joint systems that perform fully-exciting random motion. Results show how the degree of observability depends on the kinematic structure and that RNN-based observers can achieve small tracking errors in a large range of sparse and magnetometer-free setups. The proposed methods enable systematic assessment of observability properties in complex nonlinear dynamics and represent a key step toward enabling reliably accurate and non-restrictive IMT solutions.
AB - Inertial measurement units are widely used for motion tracking of kinematic chains in numerous applications. While magnetometer-free sensor fusion enables reliably high accuracy in indoor environments and near magnetic disturbances, the use of sparse sensor setups would yield additional advantages in cost, effort, and usability. However, it is unclear which sparse sensor setups can be used to track which motions of which kinematic chains, since observability of the underlying nonlinear dynamics is barely understood to date. We propose a method that utilizes recurrent neural networks (RNNs) and automatically generated training data to assess the observability of the relative pose of kinematic chains in sparse inertial motion tracking (IMT) systems. We apply this method to a range of double-hinge-joint systems that perform fully-exciting random motion. Results show how the degree of observability depends on the kinematic structure and that RNN-based observers can achieve small tracking errors in a large range of sparse and magnetometer-free setups. The proposed methods enable systematic assessment of observability properties in complex nonlinear dynamics and represent a key step toward enabling reliably accurate and non-restrictive IMT solutions.
UR - http://www.scopus.com/inward/record.url?scp=85136557940&partnerID=8YFLogxK
U2 - 10.23919/FUSION49751.2022.9841375
DO - 10.23919/FUSION49751.2022.9841375
M3 - Conference contribution
AN - SCOPUS:85136557940
T3 - 2022 25th International Conference on Information Fusion, FUSION 2022
BT - 2022 25th International Conference on Information Fusion, FUSION 2022
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 25th International Conference on Information Fusion, FUSION 2022
Y2 - 4 July 2022 through 7 July 2022
ER -