Conduit Inspection Vehicle GPS Mapping for Underground Anomaly Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional conduit inspections for wastewater and sewage systems are labor-intensive and prone to human error, with manual review of video streams requiring significant time and effort to identify anomalies like fissures, cracks, and cross-bores, and lacking precise location data for underground anomalies.
Innovation Solution
An optical imaging device coupled with a machine learning protocol and GPS system automatically detects conduit anomalies by analyzing video streams, applying alerts, and recording precise GPS locations of anomalies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual review of video streams is performed by operators and technicians, then conduit anomalies can be identified, but labor time and inspection costs increase significantly
Solution Approach 1:
The patent replaces the manual mechanical review process with an automated machine learning system. The ML model analyzes video streams captured by inspection cameras, automatically detecting anomalies such as cracks, roots, and deformations without human intervention. This substitution dramatically reduces inspection time while maintaining or improving detection accuracy.
Solution Approach 2:
The system enables self-service anomaly detection through autonomous ML algorithms that process inspection data independently. The machine learning model continuously learns from training data and automatically identifies anomalies, allowing the system to perform inspections without requiring human operators to manually review each video frame.
2Reliability
If manual review of video streams is performed, then anomalies can be detected, but human errors occur and anomalies may be missed
Solution Approach 1:
The patent replaces human manual review with automated machine learning analysis to eliminate human errors. The ML system consistently applies detection criteria across all inspection data without fatigue, distraction, or variability in attention levels that cause human operators to miss anomalies.
Solution Approach 2:
The system incorporates feedback mechanisms where the ML model continuously learns from training data and validation results. The model is trained on labeled anomaly examples and receives feedback during inference to improve its detection accuracy, reducing false negatives and ensuring consistent anomaly identification.
3Measurement precision
If travel distance measurement from starting point is used to locate anomalies, then a single set of data points is obtained, but precise location identification becomes difficult when conduit is not visible
Solution Approach 1:
The patent enhances location identification by integrating GPS coordinates and spatial mapping data with the single-dimensional travel distance measurement. This creates a multi-dimensional location system that can precisely identify anomaly positions even when the conduit is not visually accessible, by cross-referencing multiple spatial parameters.
Solution Approach 2:
The system employs a universal location tracking mechanism that works regardless of conduit visibility conditions. By combining GPS data, travel distance, and spatial mapping, the system provides consistent location identification functionality whether the conduit is visible or concealed, making the location system adaptable to all inspection scenarios.
Data Source
AI summary
A method for locating at least one anomaly inside of a conduit in real-time computing. The method includes steps of: moving a controlled inspection vehicle of a system, by a controller, inside of the conduit at a starting point; viewing at least one anomaly inside of the conduit at a point of interest (POI) with the controlled inspection vehicle; emitting a detection signal, by the controlled inspection vehicle, at the POI; finding the detection signal by a locator that is remote of the conduit; and recording the detection signal of the POI with the locator.