Road Boundary Mapping Using Trajectory-Filtered Candidate Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating boundary location information for roads, such as demarcation lines and road shoulder edges, are prone to inaccuracies due to disturbance factors like smudges, weeds, and vehicles, leading to deviations from the actual shapes of these features in automated driving systems.
Innovation Solution
An information processing device that acquires feature measurement information and trajectory data, generates candidate location information, selects relevant candidate elements based on the trajectory, and generates boundary location information to exclude disturbance factors, ensuring accurate representation of road boundaries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If candidate points are extracted using reflection luminance level information from three-dimensional point cloud data, then demarcation lines can be plotted automatically, but disturbance factors like road smudges cause inaccuracies in the generated boundary location information
Solution Approach 1:
The patent segments the point cloud data by dividing the road area into multiple regions (e.g., lane regions, shoulder regions) based on trajectory information. By processing each region separately and selecting candidate points within specific regions, the system maintains automation while reducing the impact of disturbance factors like smudges that appear in specific locations.
Solution Approach 2:
The patent introduces trajectory information as an intermediary element that mediates between the raw point cloud data and the final boundary location information. This trajectory data serves as a reference to filter and select candidate points, enabling automatic plotting while improving accuracy by excluding points that do not align with the expected road trajectory.
2Measurement precision
If manual plotting process is used to generate map information in wide range, then accurate boundary location information can be obtained, but enormous work costs are incurred due to the enormous amount of point cloud information
Solution Approach 1:
Instead of processing all candidate points uniformly, the patent applies partial action by selectively processing only those candidate points that fall within specific regions defined by trajectory information. This reduces the computational burden compared to manual processing of all points while maintaining accuracy for the relevant boundary features.
Solution Approach 2:
The system enables self-service automation where the plotting process uses the trajectory information and point cloud data to automatically identify and plot boundary features without human intervention. This eliminates the enormous work costs of manual plotting while maintaining acceptable accuracy through the region-based filtering approach.
3Productivity
If candidate points are connected simply to generate demarcation lines, then the plotting process is fast, but the generated lines deviate from the real demarcation line shapes due to disturbance factors
Solution Approach 1:
The patent applies local quality by treating different spatial regions differently. Candidate points are selected and connected based on their local region characteristics defined by trajectory information. This ensures that points are connected in a way that respects the local geometry and trajectory, improving shape accuracy while maintaining the automated fast processing.
Data Source
AI summary
An information processing device includes an information acquisition unit to acquire feature measurement information indicating locations of features including a road having a boundary and trajectory information indicating a trajectory along the road; a candidate location information generation unit to generate candidate location information indicating locations of candidate elements for the boundary, based on the feature measurement information acquired by the information acquisition unit; a selection unit to select candidate elements from among the candidate elements indicated by the candidate location information generated by the candidate location information generation unit, based on the trajectory information acquired by the information acquisition unit; and a boundary location information generation unit to generate boundary location information indicating determined locations of the boundary, using the candidate elements selected by the selection unit.