Standard methods of comparing profile measurements usually place tolerances on agreement between individual elevation values over a broad waveband. A weakness of this approach is the emphasis on long wavelength content. This often prevents the detection of important measurement problems. This paper presents a procedure for rating the agreement between profile measurements using cross correlation. The method is capable of rating agreement between profiles after the content is isolated to a given waveband of interest, such as that of the International Roughness Index. The conventional cross correlation method is customised to assess agreement between profile shape and overall roughness level. Thus, a high rating depends on proper measurement of the overall roughness and its spatial distribution. The method is demonstrated for the purpose of rating agreement between profile measurements and deducing their longitudinal offset. Threshold correlation values are developed for qualification as research class, project class, and network class.
[1]
Steven M. Karamihas,et al.
Guidelines for longitudinal pavement profile measurement
,
1999
.
[2]
Michael W. Sayers,et al.
INTERPRETATION OF ROAD ROUGHNESS PROFILE DATA
,
1996
.
[3]
James Delton,et al.
Approaches to Evaluation of Profiler Accuracy
,
2003
.
[4]
J. Bendat,et al.
Random Data: Analysis and Measurement Procedures
,
1987
.
[5]
J. D. Robson.
ROAD SURFACE DESCRIPTION AND VEHICLE RESPONSE
,
1979
.
[6]
Clifford Goodman,et al.
American Society of Mechanical Engineers
,
1988
.