Point Cloud Relation Graphs for Consistent HD Map Splicing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current high-definition map creation methods face challenges in optimizing consistency among blocks of point cloud data, which are then spliced to create the map, leading to potential deformations and reduced map quality.
Innovation Solution
A method involving the association of K frames of point clouds based on initial poses to create a point cloud relation graph, followed by point cloud registration to determine target relative poses, and finally splicing the point clouds to generate a high-definition map.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If point cloud data is divided into blocks and spliced to create high-definition map, then the map coverage area is improved, but block deformation occurs and map quality deteriorates
Solution Approach 1:
The patent divides the point cloud data into multiple blocks for parallel processing and management, allowing the system to handle large-scale map creation tasks efficiently. Each block is processed independently with associated frames grouped together, enabling scalable map coverage while maintaining quality through proper block management
Solution Approach 2:
The patent introduces a point cloud relation graph as an intermediary structure that establishes spatial relationships between adjacent blocks. This relation graph serves as a mediator that coordinates the splicing process, ensuring geometric consistency and preventing deformations when blocks are combined to form the complete high-definition map
2Productivity
If point cloud registration is performed on divided blocks, then processing efficiency is improved, but consistency among blocks deteriorates
Solution Approach 1:
The patent segments the point cloud registration process into independent block-level operations, allowing parallel processing of multiple blocks simultaneously. This segmentation maintains processing efficiency while the point cloud relation graph ensures that adjacent blocks maintain geometric consistency through shared spatial relationships
Solution Approach 2:
The patent applies equipotentiality by establishing a common reference frame and consistent coordinate system across all blocks through the point cloud relation graph. This ensures that all blocks are registered to the same spatial potential, preventing inconsistencies and deformations when blocks are spliced together
3Quantity of substance
If blocks are spliced to create complete map, then map completeness is improved, but deformations occur and quality deteriorates
Solution Approach 1:
The patent merges multiple point cloud blocks into a complete high-definition map by utilizing the point cloud relation graph to guide the splicing process. The relation graph provides spatial constraints and geometric relationships that ensure blocks are combined accurately, maintaining map quality while achieving complete coverage
Solution Approach 2:
The point cloud relation graph acts as an intermediary that mediates the splicing process between adjacent blocks. It provides the necessary spatial relationship information to ensure that when blocks are merged to create the complete map, geometric consistency is maintained and deformations are prevented
Data Source
AI summary
A high-definition map creation method includes: obtaining point cloud data collected with respect to a target region, the point cloud data including K frames of point clouds and an initial pose of each frame of point cloud, K being an integer greater than 1; associating the K frames of point clouds with each other in accordance with the initial pose to obtain a first point cloud relation graph of the K frames of point clouds; performing point cloud registration on the K frames of point clouds in accordance with the first point cloud relation graph and the initial pose to obtain a target relative pose of each frame of point cloud in the K frames of point clouds; and splicing the K frames of point clouds in accordance with the target relative pose to obtain a point cloud map of the target region.