New funding from NCDOT: Spatially Explicit Deep Learning-based Underground Pipe Prediction for Urban Stormwater Management (Deeppipe)

Stormwater management is an essential urban infrastructure as it helps protects people and property from flooding, improves water quality, and reduces the risk of infrastructure damage. However, stormwater pipeline networks that create these services are complex underground systems that require regular inspection to guide maintenance and maintain integrity.

Because the pipeline systems are underground, accurately locating aging pipe locations has been historically challenging (almost as challenging as mineral prospecting) and expensive. To help find these pipelines in advance of digging, and remotely detect anomalies that could indicate damage, CAGIS researchers, sponsored by NC Department of Transportation, are developing a new tool called DeepPipe that uses network analyses and artificial intelligence (AI) to predict where these pipes are located.

Led by Executive Director Wenwu Tang, and CAIGIS Faculty Associates Craig Allan and Shen-en Cheng, DeepPipe will focus on the prediction of pipe location, features, and service life by leveraging the fact that pipe networks are fundamentally graphs, and that missing or hidden features are topologically related to known networks. Robust, spatially explicit deep learning algorithms and other machine learning techniques will be developed as a core component of DeepPipe to resolve the challenge facing the auto-recognition, extraction/migration and transfer of pipe network data. Web- and mobile app-based implementations will be provided to facilitate the use of the DeepPipe system within in-situ environments.

In addition to delivering significant cost benefits, this research will assist NCDOT in managing urban flooding, as well as establishing policies and decision-making pertaining to extreme climate adaptation strategies.

2023-2026, NCDOT, DeepPipe: Spatially explicit deep learning-based underground pipe prediction for urban stormwater management, ($404,403), [PIs: Wenwu Tang, Craig Allan, Shenen Chen]