Frequency-Domain Pose Positioning for Noisy LiDAR Map Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous driving systems face challenges in accurately determining the location and orientation of a subject, such as an autonomous vehicle, due to low texture and high noise in local maps generated by sensors like LiDAR, making precise matching with pre-built maps difficult.
Innovation Solution
A system that processes estimated pose data to generate local maps and correlate them with reference maps in the frequency domain using phase correlation techniques, determining target pose data by analyzing cross power spectra and geometric transformations to improve matching accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If local map matching is performed using traditional spatial domain methods, then the process is straightforward, but the matching accuracy is low due to low texture and high noise in the local map
Solution Approach 1:
The patent transforms the matching problem from the spatial domain to the frequency domain by applying Fourier transform. This parameter change in the domain of analysis allows the system to overcome the limitations of low texture and high noise in the spatial domain, enabling accurate matching even when traditional methods fail. The frequency domain representation reveals patterns and correlations that are not visible in the noisy spatial data.
2Measurement precision
If frequency domain correlation is used to improve matching accuracy, then the positioning precision is enhanced, but the computational complexity increases
Solution Approach 1:
The patent replaces traditional spatial domain matching algorithms with frequency domain correlation methods. By substituting the mechanical/spatial approach with a frequency-based mathematical transformation (Fourier transform), the system achieves higher precision positioning. The frequency domain correlation efficiently handles the computational aspects through mathematical properties that simplify the matching process despite the initial transformation requirement.
Data Source
AI summary
The present disclosure is related to systems and methods for positioning. The method includes obtaining estimated pose data of a subject. The method also includes generating a local map associated with the estimated pose data. The method also includes obtaining, based on the estimated pose data, a reference map. The method also includes correlating the local map and the reference map in a frequency domain. The method further includes determining, based on the estimated pose data and the correlation between the local map and the reference map in the frequency domain, target pose data of the subject.