Iterative Map Warping for Localized Lane Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing map generation techniques for autonomous vehicles and simulators face challenges in accurately updating maps to reflect changes in geographic locations, particularly due to inconsistencies in lane segment division and reliance on unreliable position assumptions, leading to difficulties in identifying corresponding positions between old and new maps.
Innovation Solution
A method involving iterative warping of lane segments around selected warp points to maximize overlapping areas between old and new maps, using inverse distance weighted interpolation and iterative adjustments to achieve a continuous function for spatial location calculation, independent of lane segmentation assumptions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Iterative Closest Point (ICP) techniques are used to determine transformation between point clouds, then the transformation can be calculated to minimize difference between maps, but the transformation applies uniformly to the entire point cloud and does not account for localized differences
Solution Approach 1:
The patent divides the point cloud into multiple overlapping regions or tiles, allowing different transformation parameters to be applied to different segments. This segmentation enables localized adjustments while maintaining overall map alignment, resolving the contradiction between uniform transformation calculation and localized difference handling.
Solution Approach 2:
The patent implements local quality by allowing different regions of the map to have different transformation characteristics. Each local region can have its own transformation parameters derived from ICP, enabling the system to account for localized differences in map features while maintaining global consistency through overlapping regions.
2Ease of manufacture
If lane segment division is used for position identification, then map data can be structured and processed, but the division is not reliable and converting locations between lane segments is unreliable
Solution Approach 1:
The patent extracts position information from geometric features of the point cloud itself rather than relying on lane segment divisions. By identifying positions based on direct geometric relationships and transformations of point cloud data, the system removes the unreliable lane segment intermediate structure while maintaining ease of map data processing through structured point cloud operations.
3Productivity
If uniform transformation is applied to the entire point cloud, then the transformation process is simple and computationally efficient, but the transformation does not account for localized differences in map features
Solution Approach 1:
The patent segments the point cloud into overlapping regional tiles that can be transformed independently. This segmentation allows the system to maintain computational efficiency by processing smaller regional subsets rather than the entire point cloud at once, while simultaneously improving localized position accuracy through region-specific transformation parameters.
Solution Approach 2:
The patent applies transformation partially to overlapping regional subsets of the point cloud rather than attempting to transform the entire cloud uniformly. This partial action approach maintains computational efficiency by limiting each transformation operation to a manageable subset while the overlapping regions ensure comprehensive coverage and localized accuracy.
Data Source
AI summary
Methods, systems, and products for generating an updated map for use with an autonomous vehicle driving operation or a simulation thereof may include obtaining first map data associated with a first map of a geographic location including a roadway, and the first map data may include at least one first lane segment. Second map data associated with a second map of the geographic location may be obtained, and the second map data may include at least one second lane segment. A plurality of non-overlapping areas may be determined based on the first lane segment(s) and the second lane segment(s). A first non-overlapping and/or a first warp point within the first non-overlapping area may be selected. The first lane segment(s) may be warped around the first warp point to increase a total overlapping area based on the based on the second lane segment(s) and the first lane segment(s) after warping.