Integrating AI and UAVs for Real-Time Analysis and Maintenance of Stormwater Infrastructure

Abstract Submission: Artificial Intelligence (AI) systems are powerful tools that utilize machine learning and advanced algorithms to process vast amounts of data, recognize patterns, and make informed predictions. These systems continuously improve through training and can be applied to various fields, including environmental engineering. The objective is to harness AI for evaluating stormwater Best Management Practices (BMPs) using image processing techniques on data captured by Unmanned Aerial Vehicles (UAVs). The AI system trained to identify land use and estimate stormwater runoff, enabling the assessment of current BMP installations. If existing BMPs are found unsuitable for the evaluated site, the AI will suggest preferred sizes and types for optimal performance. Data collected using a DJI Mavic 2 Pro drone, processed through OpenCV, WebODM, and ArcGIS Pro. The aerial images captured by the drone undergo image processing to assess land use, BMP status, and environmental conditions. The AI system also generates predictions for future maintenance needs to ensure long-term BMP effectiveness.

Learning Objectives/Expected Outcome (Optional) : Participants will learn how Artificial Intelligence (AI) systems, combined with UAV-captured data and image processing techniques, can be applied to evaluate stormwater Best Management Practices (BMPs). They will gain insights into AI's role in identifying land use, estimating stormwater runoff, assessing BMP effectiveness, and suggesting optimized BMP designs. Additionally, participants will understand how AI can predict future maintenance needs, contributing to sustainable stormwater management and long-term environmental benefits.