KIT | KIT-Bibliothek | Impressum | Datenschutz

Unsupervised deep learning based ego motion estimation with a downward facing camera

Gilles, Maximilian; Ibrahimpasic, Sascha


Knowing the robot's pose is a crucial prerequisite for mobile robot tasks such as collision avoidance or autonomous navigation. Using powerful predictive models to estimate transformations for visual odometry via downward facing cameras is an understudied area of research. This work proposes a novel approach based on deep learning for estimating ego motion with a downward looking camera. The network can be trained completely unsupervised and is not restricted to a specific motion model. We propose two neural network architectures based on the Early Fusion and Slow Fusion design principle: “EarlyBird” and “SlowBird”. Both networks share a Spatial Transformer layer for image warping and are trained with a modified structural similarity index (SSIM) loss function. Experiments carried out in simulation and for a real world differential drive robot show similar and partially better results of our proposed deep learning based approaches compared to a state-of-the-art method based on fast Fourier transformation.

Verlagsausgabe §
DOI: 10.5445/IR/1000140923
Veröffentlicht am 08.12.2021
DOI: 10.1007/s00371-021-02345-6
Web of Science
Zitationen: 3
Zitationen: 3
Cover der Publikation
Zugehörige Institution(en) am KIT Institut für Fördertechnik und Logistiksysteme (IFL)
Publikationstyp Zeitschriftenaufsatz
Publikationsjahr 2023
Sprache Englisch
Identifikator ISSN: 0178-2789, 1432-2315
KITopen-ID: 1000140923
Erschienen in Visual Computer
Verlag Springer
Band 39
Seiten 785–798
Vorab online veröffentlicht am 27.11.2021
Nachgewiesen in Web of Science
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page