The main objective of this study was to assess the profiler gain validation technique as an alternative method for validating laser profilers that measure road roughness. Quantification was also sought for the influence that roughness and texture may have on the validation technique. The study also sought to identify a suitable reference profiler against which other profilers are assessed. Traditionally profilers are validated by comparing the International Roughness Index, IRI (m/km), against that measured by a reference device, such as a Walking Profiler (WP). The profiler gain validation technique is based on the theory that as long as the profiler exhibits unity gain (ratio of output amplitude to input amplitude) between the relevant waveband numbers (0.066 to 0.47 cycles/m) there should no effect on the measured IRI. A series of field tests were undertaken to address the main objective of this study, where the WP was used as the reference device as a comparison to laser profilers A and B in measuring IRI roughness over five test sites with varying roughness and texture. These field tests did not conclusively confirm the profiler gain validation technique. Both laser profilers failed to meet the specified gain limits at the majority of the five test sites. The specified gain limits were met at test site 3 (low roughness and high texture) for profiler A and at site 3 and site 5 (high roughness and low texture) for profiler B. This result does not appear to be due to any particular surface characteristic of the test sites as the acceptable gain results occurred on test sites with both high and low roughness and high and low texture. The field tests suggest that the profiler gain validation technique should be conducted under a standard test speed (60 to 80 km/h) for the profiler which can be maintained and allows repeatable tracking of the profiler to minimise variation in the auto-spectral density plots. At this stage a suitable reference profiler against which other profilers are assessed has not been identified because of the inconclusive outcome of the field tests. On the other hand, when applying the traditional method to validating laser profilers from the field tests, a strong correlation between the WP and profilers A and B, was found. However, this strong correlation does not mean that either laser profiler is not amplifying or attenuating a critical IRI waveband, but it is highly unlikely that this amplification or attenuation is occurring for the profilers under examination. There are two potential options regarding further work for this study: 1. Discontinue the study due to the inconclusive outcomes achieved so far and consider that even if the gain theory is sound, the practical nature of the validation technique, as indicated by this study, is currently uncertain. 2. Undertake a field test using different reference devices such as a rod and level survey. The testing can be confined to one test site. At least two different laser profilers, other than those used on the current study, should be tested against the reference devices with a follow-up profile gain analysis and report. The cost of the above field test and analysis is a significantly lower cost than the current field tests due to the reduction in test sites and should provide the confirmation, or otherwise, of the profiler gain validation technique and a suitable reference device. (Author/publisher)
Samenvatting