Records of the locations and properties of vast underground utility networks are incomplete, inaccurate, and many times unavailable. This lack of information of underground pipes poses a critical challenge to maintaining and upgrading underground infrastructure, a grand engineering challenge outlined by the National Academy of Engineering (NAE). It is also a main reason for the over six million of utility interruptions every year that cause injuries, fatalities, property damages, and environmental pollutions, amounting to billions of dollars in loss. Although sensing methods such as ground penetrating radar (GPR) has been exploited and proven its promise, it usage in urban environment is limited because of the complexity of the underworld and its reflection on the GPR signals and image signatures. We have created new methods to more accurately determine the location and dimension of underground pipes by fusing data from utility specifications and regulations and contextual cues, termed as “virtual sensors”, and GPR scans. Algorithms have been devised to extract image signatures from GPR radargrams to guide field robots in real-time, extract spatial rules from text documents, and integrating multiple sources of heterogeneous data. Research findings have been implemented as an autonomous robotic system with the capacity of adapting to complex spatial configurations of underground networks to generate accurate, geospatial 3D pipe models. Research findings are expected to shift the paradigm of the current practice in mapping underground pipes.
Dr. Hubo Cai is an Associate Professor of Civil and Construction Engineering at Purdue University. He is the founding director of the Laboratory of Computer-Integrated Infrastructure Informatics (LCIII) and the codirector of the Discrete Event Simulation (DES) Laboratory.