Self-Supervised Learning of Camera-based Drivable Surface Friction
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F21%3A00353565" target="_blank" >RIV/68407700:21230/21:00353565 - isvavai.cz</a>
Result on the web
<a href="http://dx.doi.org/10.1109/ITSC48978.2021.9564894" target="_blank" >http://dx.doi.org/10.1109/ITSC48978.2021.9564894</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1109/ITSC48978.2021.9564894" target="_blank" >10.1109/ITSC48978.2021.9564894</a>
Alternative languages
Result language
angličtina
Original language name
Self-Supervised Learning of Camera-based Drivable Surface Friction
Original language description
The visual predictor of a drivable surface friction ahead of the vehicle is presented. The image recognition neural network is trained in self-supervised fashion, as an alternative to tedious, error-prone, and subjective human annotation. The training images are labelled automatically by surface friction estimates from vehicle response during ordinary driving. The Unscented Kalman Filter algorithm is used to estimate tire-to-road interface friction parameters, taking into account the highly nonlinear nature of tire dynamics. Finally, the overall toolchain was validated using an experimental subscale platform and real-world driving scenarios. The resulting visual predictor was trained using about 3 000 images and validated on an unseen set of 800 test images, achieving 0.98 crosscorrelation between the visually predicted and the estimated value of surface friction.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
20204 - Robotics and automatic control
Result continuities
Project
—
Continuities
S - Specificky vyzkum na vysokych skolach
Others
Publication year
2021
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
ITSC 2021: IEEE Conference on Intelligent Transportation Systems
ISBN
978-1-7281-9142-3
ISSN
—
e-ISSN
—
Number of pages
8
Pages from-to
2773-2780
Publisher name
IEEE
Place of publication
Piscataway
Event location
Indianapolis
Event date
Sep 19, 2021
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—