Double-Digitized Road Geometry from Probe Data and Directional Grouping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mapping technologies face challenges in creating accurate road geometry representations that exclude special areas and provide three-dimensional data efficiently, while maintaining scalability and fidelity, especially for autonomous vehicle navigation.
Innovation Solution
A system that processes probe data using a cloud-based distributed computing environment, applies geo-spatial partitioning, and identifies seed points to create a road geometry representation, excluding special areas and enhancing 3D data, using sensor data to distinguish actual travel paths and connectivity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional centralized map creation methods are used, then processing accuracy can be maintained, but processing efficiency and scalability deteriorate
Solution Approach 1:
The patent divides the map creation process into discrete computational tasks that can be executed in parallel across multiple processing units. Probe data processing is segmented into independent operations including seed point identification, road geometry generation, and special area detection, allowing distributed computing implementation that improves processing efficiency while maintaining accuracy
Solution Approach 2:
The patent performs preliminary processing of probe data to identify seed points and filter significant vehicle activities before main map creation operations. This preliminary action includes detecting common directions of travel, identifying constrained areas, and pre-processing elevation data, which reduces the computational burden during subsequent processing stages and improves overall productivity
2Reliability
If all probe data is processed to ensure complete road coverage, then mapping completeness improves, but processing time and computational resources increase
Solution Approach 1:
The patent applies different processing strategies to different regions based on their characteristics. High-traffic areas with significant vehicle activities receive more intensive processing to ensure accurate road geometry, while low-traffic areas use simplified processing. This local quality approach maintains reliability for critical regions while reducing overall processing time
Solution Approach 2:
The patent dynamically adjusts processing parameters such as seed point density thresholds, direction grouping criteria, and special area detection sensitivity based on probe data characteristics. By changing these parameters adaptively, the system maintains high road geometry accuracy where needed while reducing computational effort in less critical areas, thereby decreasing processing time
3Adaptability or versatility
If conventional 2D road geometry is used, then data simplicity is maintained, but autonomous vehicle navigation capabilities deteriorate
Solution Approach 1:
The patent extends traditional 2D road geometry to three dimensions by integrating elevation data from multiple sources including probe data, digital elevation models, and LiDAR. This dimensionality change enables autonomous vehicle navigation on complex infrastructure like bridges and overpasses while maintaining data organization through structured storage of 3D coordinates and vertical curve parameters
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
An approach is provided for creating a road geometry representation (e.g., a base map with double digitized roads representing actual travel paths and connectivity). The approach involves, for example, retrieving probe data collected from sensors of probe devices traveling within a geographic area including at least one geographic partition. The approach also involves dividing each of the at least one geographic partition into grid cells. The approach further involves detecting directional groupings of the probe data for each grid cell. The directional groupings comprise probes of the probe data that share a respective common travel direction. The approach further involves determining a density maxima location of the probes sharing the respective travel direction, for said each grid cell and each directional grouping. The approach further involves creating a continuous road path based on the density maxima locations for the grid cells. The approach further involves including the continuous road path in an output representing a base map of the geographic area.