Ground-Penetrating Radar Mapping with Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for locating and mapping subsurface infrastructure, such as ground-penetrating radar, require expertise and are inefficient due to the need for historical maps of dubious accuracy and the risk of unplanned encounters during construction, which can lead to injuries, property damage, and delays.
Innovation Solution
A method combining ground-penetrating radar with machine learning to create three-dimensional maps of subsurface environments by storing and annotating radar data, developing a trained classifier to associate features with types of structures, and generating navigation plans to avoid infrastructure during excavations or construction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If ground-penetrating radar is used to locate subsurface infrastructure, then the ability to detect and map subsurface structures is improved, but the requirement for expert interpretation and the complexity of data analysis increases
Solution Approach 1:
The system enables self-service by using machine learning algorithms to automatically interpret radar data and generate three-dimensional maps without requiring expert human intervention. The trained classifier autonomously processes radar returns, identifies subsurface infrastructure, and produces actionable maps, allowing non-experts to utilize the technology effectively.
Solution Approach 2:
A machine learning classifier acts as an intermediary between the raw radar data and the end user. This intermediary automatically processes and interprets the complex radar returns, translating them into intuitive three-dimensional maps that are easy to understand and use, thereby eliminating the need for expert interpretation while maintaining high detection accuracy.
2Ease of operation
If historical maps are used to locate subsurface infrastructure, then the initial planning process is simplified, but the accuracy and reliability of infrastructure location information deteriorates
Solution Approach 1:
The system performs preliminary action by conducting comprehensive radar surveys and creating accurate three-dimensional maps of subsurface infrastructure before construction or maintenance projects begin. This advance mapping provides reliable location information that eliminates the need to rely on inaccurate historical maps, while the automated nature of the process maintains simplicity in the planning workflow.
3Measurement precision
If test holes are dug to confirm suspected locations of infrastructure, then the accuracy of infrastructure location is improved, but the time and cost of construction projects increases
Solution Approach 1:
The system replaces the mechanical process of digging test holes with a non-invasive ground-penetrating radar approach combined with machine learning interpretation. The radar system penetrates the ground and detects subsurface infrastructure electronically, providing accurate location verification without the need for physical excavation, thereby eliminating time loss and construction delays associated with test holes.
4Measurement precision
If manual interpretation of radar data by experts is performed, then the quality of subsurface mapping is improved, but the productivity and scalability of the mapping process deteriorates
Solution Approach 1:
The system achieves self-service by employing machine learning algorithms that automatically interpret radar data with expert-level quality. The trained classifier processes radar returns autonomously, eliminating the bottleneck of manual expert interpretation and enabling high-volume, scalable mapping operations without sacrificing mapping quality.
Solution Approach 2:
The system changes the parameter of interpretation speed from manual to automated processing. By training machine learning models on extensive radar data, the system achieves expert-quality interpretation speeds that are orders of magnitude faster than manual analysis, thereby dramatically increasing productivity and scalability while maintaining high mapping quality.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach reduces the risk associated with planning, design, maintenance, and construction by providing accurate, scalable three-dimensional maps that help identify and navigate around subsurface infrastructure, thereby minimizing the risk of encounters and delays.
Implementation Method 1
Both the training data and the target data result from illumination by ground-penetrating radar
Data Source
AI summary
A method and system for generating a map that shows subsurface structures includes the use of machine learning to develop a trained classifier that associates features in data with types of subsurface structures.