A GPR-GPS-GIS-integrated, information-rich and error-aware system for detecting, locating and characterizing underground utilities
Underground utilities have proliferated throughout the years. The location and dimension of many underground utilities have not always been properly collected and documented, leading to utility conflicts and utility strikes, and thus resulting in property damages, project delays, cost overruns, environment pollutions, injuries and deaths. The underlying reasons are twofold. First, the reliable data regarding the location and dimension of underground utility are missing or incomplete. Existing methods to collect data are not efficient and effective. Second, positional uncertainties are inherent in the measured utility locations. An effective means is not yet available to visualize and communicate the inherent positional uncertainties associated with utility location data to end-users (e.g., excavator operator). To address the aforementioned problems, this research integrate ground penetrating radar (GPR), global positioning system (GPS) and geographic information system (GIS) to form a total 3G system to collect, inventory and visualize underground utility data. Furthermore, a 3D probabilistic error band is created to model and visualize the inherent positional uncertainties in utility data. Three main challenges are addressed in this research. The first challenge is the interpretation of GPR and GPS raw data. A novel method is created in this research to simultaneously estimate the radius and buried depth of underground utilities using GPR scans and auxiliary GPS data. The proposed method was validated using GPR field scans obtained under various settings. It was found that this newly created method increases the accuracy of estimating the buried depth and radius of the buried utility under a general scanning condition. The second challenge is the geo-registration of detected utility locations. This challenge is addressed by integration of GPR, GPS and GIS. The newly created system takes advantages of GPR and GPS to detect and locate underground utilities in 3D and uses GIS for storing, updating, modeling, and visualizing collected utility data in a real world coordinate system. The third challenge is positional error/uncertainty assessment and modeling. The locational errors of GPR system are evaluated in different depth and soil conditions. Quantitative linkages between error magnitudes and its influencing factors (i.e., buried depths and soil conditions) are established. In order to handle the positional error of underground utilities, a prototype of 3D probabilistic error band is created and implemented in GIS environment. This makes the system error-aware and also paves the way to a more intelligent error-aware GIS. To sum up, the newly created system is able to detect, locate and characterize underground utilities in an information-rich and error-aware manner.
Cai, Purdue University.
Off-Campus Purdue Users:
To access this dissertation, please log in to our