Predicting aquaplaning performance from tyre profile images with machine learning

Weyde, T., Slabaugh, G.G., Fontaine, G. & Bederna, C. (2013). Predicting aquaplaning performance from tyre profile images with machine learning. Lecture Notes in Computer Science, 7950 L, pp. 133-142. doi: 10.1007/978-3-642-39094-4_16

[img]
Preview
PDF
Download (1MB) | Preview

Abstract

The tread of a tyre consists of a profile (pattern of grooves, sipes, and blocks) mainly designed to improve wet performance and inhibit aquaplaning by providing a conduit for water to be expelled underneath the tyre as it makes contact with the road surface. Testing different tread profile designs is time consuming, as it requires fabrication and physical measurement of tyres. We propose a supervised machine learning method to predict tyres’ aquaplaning performance based on the tread profile described in geometry and rubber stiffness. Our method provides a regressor from the space of profile geometry, reduced to images, to aquaplaning performance. Experimental results demonstrate that image analysis and machine learning combined with other methods can yield improved prediction of aquaplaning performance, even using non-normalised data. Therefore this method has can potentially save substantial cost and time in tyre development. This investigation is based on data provided by Continental Reifen Deutschland GmbH.

Item Type: Article
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Q Science > QC Physics
Divisions: School of Informatics > Department of Computing
URI: http://openaccess.city.ac.uk/id/eprint/2753

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics