Several ADAS (Advanced Driver Assistance Systems) need information about the maximum available friction force between tyre and road. This can be estimated from the output of vehicle dynamics sensors, but the problem is that the friction potentials vary considerably between different road types. One possibility is to use an enviromental sensor to estimate the current road type, and hence to allow the usage of vehicle dynamics based friction estimates on all driving situations. In this article we propose a Bayesian method for sequentially classifying the road surface measurements. The approach is based on casting the problem into a multiple target tracking framework, in which the tracked targets are the measurement regions of different road types. Simple heuristic dynamic models are proposed for the measurement regions, which allow to estimate the data association indicators together with the target states efficiently with a Rao-Blackwellized particle filter. The proposed framework allows also to model clutter measurements, and to define dynamic models for the data association variables. Preliminary classification results of real road surface measurements are presented for the proposed classification method. The results show that the classification results for overlapping road types can be improved by defining suitable dynamic models for data associations. We also illustrate with a simple simulated example how the surface classification results can be combined with vehicle dynamics based friction potential estimates.
Tyre-road friction potential estimation by data fusion. Bayesian approach to road type classification
im Spannungsfeld von Sicherheit und Umwelt, Internationale VDI-Tagung Reifen-Fahrwerk-Fahrbahn, 12 ; 271-288
VDI-Berichte ; 2086
2009
18 Seiten, 8 Bilder, 19 Quellen
Aufsatz (Konferenz)
Englisch
Tyre-road friction potential estimation by data fusion
British Library Conference Proceedings | 2009
|Estimation of tyre–road friction coefficient
Online Contents | 2012
|Estimation of tyre-road friction coefficient
Kraftfahrwesen | 2012
|Tyre-road friction estimation - a comparative study
Kraftfahrwesen | 2008
|