Roger Roberts

Bio: Roger Roberts is an academic researcher from Geophysical Survey. The author has contributed to research in topics: Ballast & Track (rail transport). The author has an hindex of 12, co-authored 18 publications receiving 379 citations. Previous affiliations of Roger Roberts include University of Illinois at Urbana–Champaign.

Scattering analysis of ground-penetrating radar data to quantify railroad ballast contamination

Abstract: This paper will evaluate ground-penetrating radar (GPR) as a non-destructive method to rapidly, effectively, and continually assess the conditions of railroad ballast. Compared to uniformly graded, clean ballast, fouled ballast has a finer, well-graded particle size with fewer air voids. Ballast under different conditions generates various GPR electromagnetic scattering patterns. A field GPR survey with multiple sets of 1 and 2 GHz air-horn antennae was conducted in summer 2005 at the Transportation Technology Center, Inc. (TTCI) in Pueblo, Colorado. The 2 GHz antenna was found to be more sensitive to the change in scattering pattern. Appropriate data processing was used to remove the effects of the rails to obtain clear GPR images of the subsurface layers. From the image analysis, ballast thickness, ballast fouling condition, and trapped water can be assessed.

Data Analysis Techniques for GPR Used for Assessing Railroad Ballast in High Radio-Frequency Environment

Abstract: Railroad ballast supports heavy rail loading, prevents track deformation, and provides drainage of water from the track structure. However, over time, ballast is fouled by the breakdown of ballast aggregate and/or the infiltration of fines, which undermine the ballast functions and affect the railroad track structural capacity. Ground penetrating radar (GPR) provides a rapid, effective, and continuous way to assess railroad track substructure condition; especially ballast. However, the GPR system faces some challenges during field surveys including high radio-frequency interference from railroad communication and automation, and strong reflections from rails. In this study, appropriate techniques were used to remove the interference and reduce the strong clutter from rails to obtain clear GPR data of railroad substructure. A time-frequency method, short-time Fourier transform, was then applied to extract ballast fouling condition over depth. A field survey using multiple sets of 2-GHz air-horn antennae was conducted during summer 2007 at the Transportation Technology Center, Inc. in Pueblo, Colo. Compared to ground-truth excavation and ballast gradation analysis results, GPR was found to be an effective technique to assess railroad track ballast substructure condition.

Railroad Ballast Fouling Detection Using Ground Penetrating Radar - A New Approach Based on Scattering from Voids

Abstract: Railroad ballast is the uniformly-graded aggregate between and underneath railroad ties. The purpose of ballast is to provide support for the heavy loading applied by trains. New ballast contains significant void space between the aggregate. As ballast ages it is progressively fouled by fine-grained material that fills the void space. The structural integrity of seriously fouled ballast can be compromised leading to track instability and ultimately, train derailments. For this reason it is very important to be able to detect fouled ballast. Ground penetrating radar (GPR) has been used over the past 25 years for ballast evaluation and has yielded mixed results. Traditionally, GPR data have been interpreted with strategies focusing on reflections from layer interfaces. This can be a severe handicap when obtaining data on a ballast structure which contains gradational fouling. GSSI developed a 2 GHz horn antenna in 2003 which was initially tested on ballast at the Transportation Technology Center, Inc. in Pueblo, CO in 2005 and subsequently on Amtrak Rails near Boston, MA in 2006. It was noted that the antenna is much more focused compared to the commonly used 1 GHz horn antenna and provides data that contain significant scattering energy from the void space in clean ballast. The data from fouled ballast contained minimal scattering from the remaining void space. Comparison of the data obtained using the 2 GHz antenna with available ground truth supports this interpretation. This GPR methodology shows potential for routine inspection of working railroads and is especially suited to assess the ballast fouling condition near the bottom of railroad ties.

Optimization of antenna configuration in multiple-frequency ground penetrating radar system for railroad substructure assessment

Abstract: This paper discusses the use of ground penetrating radar (GPR) to rapidly, effectively, and continuously assess railroad track substructure conditions, especially ballast. To overcome the limited electromagnetic waves penetration for high-frequency antennae and the low resolution of low-frequency antennae, this study uses a multiple-frequency GPR system to assess railroad substructure conditions. High-frequency antennae were used to detect the scattering pattern, which is related to air void volume in railroad ballast, and low-frequency antennae are used to assess deeper substructure conditions. Considering the scattering energy attenuation is highly frequency and material dependent, a time–frequency method based on tracking the frequency spectrum and energy change over depth can be used to extract ballast fouling conditions. From GPR field collected data, ground-truth observation, and ballast gradation analysis, the multiple-frequency GPR system demonstrates a promising capability to assess railroad track substructure condition.

NDE methods for quality assurance of new pavement thickness

Abstract: Accurate measurement of pavement thickness is an essential aspect of the quality assurance of new pavement construction. Current coring methods are time consuming and provide a very limited representation of the overall pavement structure. The objective of the work described in this paper has been to demonstrate the use of non-destructive evaluation (NDE) methods for rapidly determining the average pavement thickness on a newly constructed section to within 2.5 mm of the true value, without extensive reliance on cores. The effort has considered ground penetrating radar (GPR) and impact echo methods applied to both asphalt and concrete pavement, and has included laboratory and field-testing, with field correlations based on 172 cores. The results show that the 2.5 mm accuracy objective can be met for asphalt pavement, but that accuracy on concrete is limited to 4 mm. The paper describes the techniques that were evaluated, the testing that was conducted, and the results of correlation with core data.

Full-resolution 3D GPR imaging

Abstract: Noninvasive 3D ground-penetrating radar (GPR) imaging with submeter resolution in all directions delineates the internal architecture and processes of the shallow subsurface. Full-resolution imaging requires unaliased recording of reflections and diffractions coupled with 3D migration processing. The GPR practitioner can easily determine necessary acquisition trace spacing on a frequency-wavenumber (f-k) plot of a representative 2D GPR test profile. Quarter-wavelength spatial sampling is a minimum requirement for full-resolution GPR recording. An intensely fractured limestone quarry serves as a test site for a 100-MHz 3D GPR survey with 0.1 m × 0.2 m trace spacing. This example clearly defines the geometry of fractures in four different orientations, including vertical dips to a depth of 20 m. Decimation to commonly used half-wavelength spatial sampling or only 2D migration processing makes most fractures invisible. The extra data-acquisition effort results in image volumes with submeter resolution, both ...

Method and apparatus for detecting, mapping and locating underground utilities

Abstract: The method and device for locating underground utilities within an area includes traversing the area with a plurality of underground utility sensors and obtaining area location data to locate the area traversed. The sensor data and area location data are used to map the location of one or more utilities within the area traversed.

Porous Asphalt Pavement Temperature Effects for Urban Heat Island Analysis

Abstract: Increased nighttime temperatures caused by retained heat in urban areas is a phenomenon known as the urban heat island (UHI) effect. Urbanization requires an increase in pavement surface area, which contributes to UHI as a result of unfavorable heat retention properties. In recent years, alternative pavement designs have become more common in an attempt to mitigate the environmental impacts of urbanization. Specifically, porous pavements are gaining popularity in the paving industry because of their attractive storm water mitigation and friction properties. However, little information regarding the thermal behavior of these materials is available. This paper explores the extent to which porous asphalt pavement influences pavement temperatures and investigates the impact on UHI by considering the diurnal temperature cycle. A one-dimensional pavement temperature model developed at Arizona State University was used to model surface temperatures of porous asphalt, traditional dense-graded asphalt, and portlan...

Development and validation for in situ asphalt mixture density prediction models

Abstract: In situ asphalt mixture density is an important quality property of flexible pavements. A previous study introduced the potential of ground penetrating radar (GPR) to estimate in situ asphalt mixture density continuously, rapidly, and nondestructively. Three density prediction models were developed based on the relationship between the asphaltic mixture volumetric characteristics and the components' dielectric constants. In this study, a full-scale test site was carefully designed and constructed for the model validation. Five different mixes were placed in the test site, and each was compacted at four density levels. Both GPR data and cores were collected from the test site to validate the performance of the density models developed in the previous study. The validation results indicated that all three models provided reasonably accurate predictions with errors in the range of 2.2–2.8%, and the modified Bottcher model (Al-Qadi, Lahouar and Leng (ALL) model) performed the best. In addition, the authors provided the appropriate algorithm for predicting in situ asphalt mixture density through a GPR survey.