Roadside Point Cloud Alignment for Real-Time SLAM Deviation Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional loop-closure detection methods in SLAM environments for autonomous vehicles are inefficient in correcting accumulated errors in a timely manner, especially in outdoor environments where returning to the same position can take a long time, leading to low frequency of deviation correction and inaccurate frame matching due to lack of accurate world coordinate system registration.
Innovation Solution
A method for roadside-assisted vehicle sensor deviation correction in SLAM environments, where point cloud data from roadside units is received and transformed to match the vehicle's point cloud data, allowing for real-time correction of the vehicle's position and deviation within the world coordinate system without relying on loop-closure detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If loop-closure detection is used to correct accumulated errors in SLAM, then position accuracy is improved, but the correction frequency is too low because the vehicle must return to the same position
Solution Approach 1:
The patent introduces a global coordinate system as an intermediary reference framework. Instead of relying on the vehicle returning to the same position for loop-closure detection, the system establishes a fixed global coordinate system and uses it to continuously correct vehicle position and sensor deviations. This intermediary reference enables frequent real-time corrections without requiring the vehicle to complete full loops, thereby resolving the contradiction between position accuracy and correction frequency.
2Reliability
If traditional loop-closure detection is used, then accumulated errors are corrected, but frame matching accuracy deteriorates due to lack of accurate world coordinate system registration
Solution Approach 1:
The patent performs preliminary registration of the global coordinate system before conducting frame matching operations. By establishing an accurate world coordinate system in advance and using it to transform and align point cloud data from different time points, the system ensures that frame matching is performed with accurate spatial reference. This preliminary coordinate system setup eliminates the accumulation of coordinate transformation errors and improves frame matching accuracy while maintaining error correction capability.
3Measurement precision
If the vehicle returns to the same position for loop-closure detection, then position accuracy is improved, but time consumption increases significantly in outdoor environments
Solution Approach 1:
The global coordinate system serves as a time-efficient intermediary that eliminates the need for the vehicle to physically return to the same position. By continuously referencing the fixed global coordinate system and performing real-time transformations, the system achieves position accuracy corrections at any location and any time, rather than waiting for loop-closure conditions to be met. This resolves the contradiction by providing accurate position correction without the time penalty of requiring full loop completion.
Data Source
AI summary
A method for position and/or deviation correction of a moving object in a simultaneous localization and mapping (SLAM) environment. The method comprises receiving at the moving object point cloud data for an area associated with the moving object from a sensor located at a fixed location. A processor of the moving object performs the steps of: transforming the received point cloud data to a coordinate system of the moving object; matching the transformed point cloud data with point cloud data generated by a sensor at the moving object for the area associated with the moving object; and using the matched point cloud data to determine a corrected position and/or deviation of the moving object within a selected coordinate system.