Interactive 3D Point Cloud Matching for HD Map Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing approaches to building high definition maps lack efficient tools for users to visually explore and interactively edit LiDAR scan data, leading to inaccuracies and inefficiencies in the mapping process.
Innovation Solution
The development of interactive user interfaces that enable users to view, edit, and align 3D point cloud and pose graph data, allowing for the improvement of high definition map quality by visually identifying and correcting inconsistencies and misalignments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated processing is used to generate HD maps from LiDAR data, then productivity is improved, but manufacturing precision deteriorates due to inaccuracies and misalignments
Solution Approach 1:
The system provides visual feedback by rendering 3D point cloud data and 2D map projections that users can interactively examine. Users can identify misalignments and inaccuracies in the automated processing results, allowing for iterative refinement and correction of HD map data to achieve higher precision while maintaining productivity
Solution Approach 2:
The patent introduces an intermediary interface between automated processing and final map generation. This interface includes visualization tools and adjustment mechanisms that allow users to intervene in the process, correct errors, and refine alignments before final map production, thus resolving the precision-productivity tradeoff
2Manufacturing precision
If interactive editing tools are added to allow manual adjustment of point clouds, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The interface is segmented into distinct functional modules: 3D point cloud visualization, 2D map projection, pose graph display, and adjustment tools. Each module handles specific tasks independently, making the complex system more manageable and easier to use while maintaining high precision editing capabilities
Solution Approach 2:
The system transitions between different dimensional representations of the same data - from 3D point cloud views to 2D map projections and pose graphs. This dimensional transformation allows users to perceive and edit spatial relationships in multiple ways, reducing the perceived complexity while maintaining editing precision
3Reliability
If users can visually explore and edit LiDAR scan data, then reliability is improved, but loss of time increases due to manual inspection requirements
Solution Approach 1:
The system performs preliminary automated processing to generate initial HD maps and pose graphs before user intervention. This pre-processing handles the bulk of the work, and users only need to intervene for specific corrections, thereby maintaining high reliability while minimizing time loss
Solution Approach 2:
Instead of requiring users to review entire datasets, the system enables local-quality editing by allowing users to selectively examine and adjust specific regions, points, or pose graph elements that require attention, thus maintaining reliability without unnecessary time consumption
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
These interfaces enhance the accuracy and efficiency of high definition map generation by enabling users to manually adjust and align point clouds, resulting in more precise and reliable maps for autonomous vehicles.
Implementation Method 1
point cloud data generated from a plurality of light detection and ranging (LIDAR) scans
Data Source
AI summary
Systems and methods are disclosed related to generating an interactive user interface that enables a user to move, rotate or otherwise edit 3D point cloud data in virtual 3D space to align or match point clouds captured from LiDAR scans prior to generation of a high definition map. A system may obtain point cloud data for two or more point clouds, render the point clouds for display in a user interface, then receive a user selection of one of the point clouds and commands from the user to move and/or rotate the selected point cloud. The system may adjust the displayed position of the selected point cloud relative to the other simultaneously displayed point cloud(s) in real time in response to the user commands, and store the adjusted point cloud position data for use in generating a new high definition map.