Editorial Type:
Article Category: Research Article
 | 
Online Publication Date: 01 Sept 2010

Evaluation of LuGre Tire Friction Model with Measured Data on Multiple Pavement Surfaces

,
, and
Page Range: 213 – 227
DOI: 10.2346/1.3481671
Save
Download PDF

Abstract

Accurate modeling of tire/pavement friction phenomena is of utmost importance in many applications such as vehicle braking control and frictional evaluation of pavements. LuGre tire friction model is a model which is used for this purpose by estimating its parameters using measured pavement friction data. In this investigation, LuGre model parameters were estimated using field data collected by a standard pavement friction measuring device (Locked Wheel Skid Trailer-ASTM E 274) at a group of pavements with different surface friction properties. Adequacy of the model to predict measured friction data from the device was statistically evaluated and the accuracy of estimated model parameters was determined. The results show the potential of this model to facilitate frictional evaluation of pavements using dynamic friction measuring equipment.

Keywords: LuGre; tire friction; skid number; locked wheel skid trailer
Copyright: The Tire Society
FIG. 1
FIG. 1

Locked wheel skid trailer.

FIG. 2
FIG. 2

Typical skid data graph provided by Winskid) software.

FIG. 3
FIG. 3

Tampa data at five different speeds (a) average, (b) standard deviation.

FIG. 4
FIG. 4

Calibration results with observed data from Tampa (a) calibration at 96 km/h (R2)=0.76, RMS Error=0.029), (b) calibration at 80 km/h (R2)=0.86, RMS Error=0.029), (c) calibration at 48 km/h (R2)=0.93, RMS Error=0.023), (d) calibration at 32 km/h (R2)=0.92, RMS Error=0.028).

FIG. 5
FIG. 5

Prediction results of Tampa data at 64 km/h with estimated model parameters using observed data at other 4 speeds (R2)=0.90, RMS Error=0.025).

FIG. 6
FIG. 6

Dispersion of measured data around model predicted results (a) view from a top angle, (b) view from a bottom angle.

  • Download PDF