SEWER-SEER: Robotic Computer Vision System for the Diagnosis of Sewer Pipes using Artificial Intelligence for the Department of Drinking Water and Sanitation of the Intercultural GAD of the Cañar Canton

The Sewer-Seer project addresses the critical challenge of efficient sewer management in the Intercultural GAD of Cañar Canton, where traditional visual inspection methods by technical staff are prone to human error, expose workers to hazardous environments, and result in inefficient costs and time. The main objective is to develop a robotic computer vision system assisted by artificial intelligence for the precise diagnosis of sewer pipes up to 1 meter in diameter and 100 meters in length. The proposed solution involves creating a local knowledge base, training a diagnostic model based on neural networks (Res-Net) using real and simulated images, designing and implementing specialized robotic hardware (including vision and extension mechanisms), and developing real-time diagnostic software using the SCRUM methodology. This system is expected to significantly improve the accuracy in detecting blockages and structural damage, reduce occupational hazards by minimizing personnel exposure to hostile environments, and optimize operational costs and intervention time. The second stage focuses on technology transfer to the GAD of Cañar through training, detailed documentation, and technical support, aiming to reach a Technology Readiness Level (TRL) of 8, and promoting the scientific dissemination of the results.<br/><br/><b>Goal</b>: <br/>To develop a computer vision system for the diagnosis of sewer pipes using artificial intelligence, integrated onto a mobile inspection robot, for its implementation and validation within the Water and Sanitation Department of the Intercultural GAD of the Cañar Canton.<br/><br/><b>Research lines</b>: <br/>Condition Monitoring