Lane Geometry Clustering for Faster Boundary and Centreline Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining lane boundaries and lane centrelines in digital maps are computationally expensive and prone to inaccuracies due to noisy and incomplete sensor data, requiring significant processing effort to mitigate anomalies.
Innovation Solution
A method utilizing a spatial indexing system to cluster data sets representing lane boundaries or centrelines, involving resampling and distance threshold comparisons to identify and determine geometries and positions efficiently, reducing unnecessary processing by filtering out remote data points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If large quantities of sensor data are collected and processed to determine lane boundary geometries and lane centreline positions, then measurement precision and reliability are improved, but computational cost and processing time increase significantly
Solution Approach 1:
The patent divides the geographical area into a spatial indexing system with multiple tiles, and further segments the data processing by creating candidate groups based on spatial proximity. This segmentation allows the system to process data in manageable chunks rather than handling all data at once, reducing computational complexity while maintaining accuracy.
Solution Approach 2:
The patent extracts and removes anomalies and noise from the sensor data through identification and mitigation processes. By taking out problematic data points, the system reduces the computational burden of processing invalid data while maintaining the precision of the final lane geometry determination.
2Reliability
If data processing includes identification and mitigation of anomalies and noise, then reliability of lane boundary determination is improved, but processing time and computational expense increase
Solution Approach 1:
The patent performs preliminary actions by first organizing data into spatial tiles and identifying candidate groups before conducting full anomaly detection and mitigation. This preliminary organization allows for more efficient subsequent processing and reduces the time required to identify and mitigate anomalies in the final analysis.
3Measurement precision
If comprehensive data processing is performed on all observed lane boundaries, then measurement precision is improved, but device complexity and processing requirements increase
Solution Approach 1:
The patent applies segmentation by dividing the data into spatial tiles and creating candidate groups based on spatial proximity. This segmentation simplifies the processing system by allowing it to handle data in organized, manageable portions rather than processing all data comprehensively, reducing device complexity while maintaining precision.
Data Source
AI summary
Disclosed is a method in which plural sets of data representing a plurality of separate observations of lane boundaries within the road section are obtained. An initial candidate group of sets of data is identified based on corresponding data points within the sets of data fitting into the same tile or in n-level neighbouring tiles of a spatial indexing system. A cluster of sets of data that relate to the same, first lane boundary is determined from the identified initial candidate group of sets of data by calculating respective distances between corresponding data points for different sets of data and comparing the calculated distances to a distance threshold. A geometry of the first lane boundary is determined using the cluster of sets of data that have been determined to relate to the same, first lane boundary.