Road Geometry Correction Model Using Movement Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing map data systems face challenges in accurately and efficiently updating road geometries due to manual review delays and limitations of automated mapping vehicles.
Innovation Solution
A machine learning system that combines stored road geometry with historical movement data to automatically identify changes in road geometry, using a convolutional encoder-decoder neural network to generate road geometry corrections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual review of flagged errors in road geometries is performed by cartographers, then accuracy of map data is improved, but time delay increases
Solution Approach 1:
The patent introduces an automated error detection system that acts as an intermediary between raw movement data and manual cartographer review. The system processes movement data to generate candidate corrections, which are then presented to cartographers for verification. This intermediary layer filters and prioritizes errors, reducing the time cartographers need to spend while maintaining accuracy through their expert review of the most critical cases.
Solution Approach 2:
The system performs preliminary automated analysis of movement data to identify potential road geometry errors before they reach human cartographers. By pre-processing the data and generating candidate corrections in advance, the system reduces the workload and time required for manual review, allowing cartographers to focus their expertise on verifying and refining the automated findings.
2Productivity
If automated mapping vehicles are used to traverse the road network, then productivity of map data updating is improved, but time delay increases
Solution Approach 1:
The patent replaces the mechanical system of physical mapping vehicles with an electronic data processing system that analyzes movement data from various sources. Instead of deploying vehicles to physically traverse roads, the system processes digital movement data to identify road geometry errors, significantly reducing the time required for error identification while maintaining high productivity through automated analysis.
Solution Approach 2:
The system creates a digital representation (copy) of the road network by analyzing movement data, which can be processed and compared against existing map data without requiring physical traversal. This digital copying approach allows for rapid, repeated analysis of the same area without the time constraints of physical vehicle deployment.
3Measurement precision
If extensive effort is expended in updating existing mapping data, then accuracy of road geometries is improved, but loss of time increases
Solution Approach 1:
The patent implements a feedback mechanism where movement data from navigation systems continuously provides information about actual road conditions. This feedback loop allows the system to automatically detect discrepancies between stored map data and actual road geometries, enabling ongoing updates without requiring extensive periodic manual efforts. The feedback-driven approach maintains accuracy through continuous, automated monitoring rather than time-consuming periodic updates.
Data Source
AI summary
A method is provided for identifying changes in road geometry, comprising obtaining an image of an initial road geometry for a geographical area, obtaining an image of movement data for the geographical area, forming a composite image from at least the image of initial road geometry and the image of movement data and generating an image of road geometry corrections by applying a trained road geometry correction model to the composite image, wherein the image of road geometry corrections identifies one or more differences between the actual road geometry of the geographical area and the initial road geometry. A method of training a suitable road geometry correction model is also described.