Adjacent Lidar Data Alignment with Buffered Overlap Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Lidar data sets generated at different times or by different scanners often exhibit positional offsets, particularly vertical offsets, leading to errors in high-definition maps, such as erroneous ridges or drop-offs at the interface between data sets.
Innovation Solution
The method involves defining polygons for each lidar data set, forming buffers around these polygons, identifying an overlap region, determining nearest neighbor lidar point pairs within this region, and calculating positional offsets to apply correction factors and reduce or eliminate these errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If lidar scans are performed at different times or by multiple scanners to cover large environments, then the coverage area increases, but positional offsets occur between adjacent lidar data sets
Solution Approach 1:
The patent introduces an overlap region as an intermediary zone between adjacent lidar data sets. This overlap region contains corresponding lidar points from both data sets that can be used to calculate transformation parameters. By using this intermediary region, the system can resolve positional offsets and achieve precise alignment while maintaining large coverage area through multiple scans.
Solution Approach 2:
The patent implements a feedback mechanism where transformation parameters are calculated by comparing lidar points in the overlap region, then these parameters are applied to correct the positional offsets in the adjacent data sets. This iterative correction process ensures that the combined point cloud maintains high positional accuracy across the entire coverage area.
2Measurement precision
If positional offsets are corrected by processing all lidar points, then alignment accuracy improves, but computational complexity increases
Solution Approach 1:
The patent divides the lidar data into distinct segments with a designated overlap region. Instead of processing all lidar points across the entire coverage area, the system focuses computational efforts only on the overlap region to calculate transformation parameters. This segmentation approach maintains high alignment accuracy while significantly reducing computational complexity.
Solution Approach 2:
The patent applies local quality by concentrating detailed processing resources on the overlap region where transformation calculations are needed, while the rest of the data sets can be combined using the calculated parameters without intensive processing. This approach achieves precise alignment with optimized computational resource allocation.
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 enhances the accuracy of high-definition maps by correcting positional offsets, ensuring a more precise representation of the environment, which is crucial for autonomous vehicle navigation.
Implementation Method 1
lidar scanners emit light (e.g., infrared light) into an environment and have one or more detectors that receive and detect light emitted by the scanner that has been reflected by objects within the environment
Implementation Method 2
The distance of an object from the lidar scanner, and thus the location of an object within the environment, is determined by the amount of time it takes for the reflected light to be received by the one or more detectors
Data Source
AI summary
In one embodiment, a method for resolving positional offsets between adjacent lidar data sets includes defining a first polygon for a first lidar data set and defining a second polygon for a second lidar data set, forming a first buffer around the first polygon and a second buffer around the second polygon, defining an overlap region between the first buffer and the second buffer, for each lidar point of the first lidar data set or the second lidar data set within the overlap region, determining a lidar point pair by finding a nearest neighbor of the other of the first lidar data set or the second lidar data set, determining a positional offset for each lidar point pair, and determining a polygon offset based at least in part on the positional offsets for the lidar point pairs.