3D Towered Checkerboard for LiDAR Camera Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional two-dimensional checkerboards are inadequate for accurate three-dimensional calibration of LiDAR and camera systems due to planar features causing misalignment and requiring multiple frames of data, leading to inaccuracies and complex manual operations in joint calibration processes.
Innovation Solution
A three-dimensional towered checkerboard is designed by stacking polyhedrons with identical shapes and orientations, covered with two-dimensional checkerboards, allowing for the creation of a multilayer structure that generates a three-dimensional point cloud for precise calibration, enabling accurate extrinsic parameter determination between LiDAR and camera systems through a joint calibration method.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a two-dimensional checkerboard calibration plate is used for joint calibration of LiDAR and camera, then the calibration process can be performed, but the calibration accuracy deteriorates due to misalignment between LiDAR perceived plane and camera perceived corners
Solution Approach 1:
The patent transitions from a two-dimensional checkerboard to a three-dimensional towered checkerboard structure. This dimensional change enables the LiDAR to capture three-dimensional point cloud data with vertical features, allowing direct correspondence between LiDAR points and camera corners. The three-dimensional structure resolves the misalignment issue by providing depth information that the planar checkerboard lacked, thereby improving calibration accuracy without increasing operational difficulty.
Solution Approach 2:
The calibration target combines multiple materials and structures: a three-dimensional framework made of vertical rods or towers, checkerboard patterns attached to these structures, and reflective or high-contrast surfaces. This composite structure simultaneously provides three-dimensional geometric features for LiDAR detection and two-dimensional pattern recognition for camera detection, enabling accurate joint calibration of both sensors.
2Measurement precision
If a two-dimensional checkerboard calibration plate is used for joint calibration, then the calibration can be performed, but the calibration accuracy deteriorates when LiDAR beams are relatively sparse
Solution Approach 1:
The three-dimensional towered structure provides vertical extent that increases the number of detectable points along each LiDAR beam path. Instead of a flat surface that may be missed by sparse beams, the vertical towers ensure that beams intersecting the calibration target at various heights will detect features, thereby maintaining calibration accuracy even with lower beam density.
Solution Approach 2:
The calibration target is divided into multiple vertical towers or rods distributed across the structure. This segmentation ensures that sparse LiDAR beams are more likely to intersect at least one tower, providing sufficient point cloud data for accurate calibration. Each tower acts as an independent feature detector, increasing the probability of successful feature detection with fewer beams.
3Measurement precision
If multiple frames of checkerboard calibration plate data are required for joint calibration, then more data can be collected, but the calibration process becomes more complex and time-consuming
Solution Approach 1:
The three-dimensional structure enables accurate calibration to be achieved in a single frame by providing sufficient geometric constraints in three-dimensional space. The vertical towers create strong perspective cues and depth information that allow the calibration algorithm to compute extrinsic parameters from one static view, eliminating the need for multiple frames or temporal data collection sequences.
Solution Approach 2:
The calibration target is pre-configured with a specific three-dimensional geometry and checkerboard pattern placement that is optimized for single-frame calibration. This preliminary design ensures that all necessary calibration information is encoded in the spatial arrangement and visual patterns, allowing the system to extract calibration parameters from a single captured frame without requiring additional frames or complex temporal processing.
4Productivity
If a two-dimensional checkerboard is used for joint calibration, then the process can be completed, but manual operation becomes more troublesome due to the requirement for multiple frames of data
Solution Approach 1:
The three-dimensional towered checkerboard structure enables the system to complete calibration from a single static capture, eliminating the need for operators to collect and process multiple frames. The added vertical dimension provides sufficient geometric constraints that simplify the calibration workflow, reducing manual intervention and automating the process to require only a single capture action.
Data Source
AI summary
The disclosure is a three-dimensional towered checkerboard for multi-sensor calibration, and a LiDAR and camera joint calibration method based on the checkerboard. The joint calibration method includes: establishing a modeling coordinate system taking the three-dimensional towered checkerboard as a basis, and generating a point cloud of the three-dimensional towered checkerboard; denoising a three-dimensional point cloud obtained by LiDAR, and obtaining an actual point cloud of the three-dimensional towered checkerboard under a LiDAR coordinate system; determining a transformation relationship between the LiDAR coordinate system and the modeling coordinate system; generating a corner point set of two-dimensional checkerboards under the modeling coordinate system in sequence according to actual positions of corners of the two-dimensional checkerboards, and transforming into the LiDAR coordinate system; obtaining a corner point set of the two-dimensional checkerboards on a photo; and calculating a transformation relationship between the camera coordinate system and the LiDAR coordinate system.