LiDAR Scan Smoothing for Transient Object Filtering in Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurately performing localization due to transient elements like people, bicycles, and foliage, which change frequently, making it difficult to maintain consistent LiDAR scan data for accurate navigation.
Innovation Solution
A smoothness coefficient is calculated for each LiDAR data point based on neighboring points, and if it exceeds a threshold, the point is discarded to focus on less transient structures like buildings, improving the accuracy and repeatability of the localization process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If all LiDAR data points including transient elements are used for localization, then more data is available for processing, but localization accuracy decreases due to transient elements like people, bicycles, and foliage
Solution Approach 1:
The patent extracts and removes transient elements from the LiDAR scan data by calculating a smoothness coefficient for each data point and comparing it to a threshold. Data points with low smoothness coefficients (indicating transient elements) are discarded, while points with high coefficients (permanent structures) are retained for localization processing.
Solution Approach 2:
The patent applies different quality criteria to different data points based on their local characteristics. By calculating smoothness coefficients locally for each data point and its neighbors, the system identifies and retains only those points that exhibit permanent, stable structural characteristics, while discarding points with transient local variations.
2Reliability
If transient elements are removed from LiDAR scans, then localization consistency improves, but the complexity of data processing increases due to smoothness coefficient calculation
Solution Approach 1:
The patent segments the LiDAR scan data into individual data points and processes each point independently by calculating its smoothness coefficient based on neighboring points. This segmentation allows for efficient parallel processing and reduces the overall computational complexity compared to analyzing the entire scan as a single unit.
Solution Approach 2:
The patent transforms the raw LiDAR data into a new parameter space by calculating smoothness coefficients for each data point. This parameter transformation simplifies the identification of transient elements, as points with low smoothness coefficients can be easily filtered out, reducing processing complexity in subsequent localization steps.
3Object-affected harmful factors
If a strict threshold is used to filter LiDAR data points, then transient element removal is more effective, but more permanent structures may be incorrectly discarded
Solution Approach 1:
The patent employs a dynamic thresholding approach where the threshold for discarding data points is not fixed but adapts based on the distribution of smoothness coefficients in the current scan. This dynamic adjustment ensures that transient elements are effectively removed while preserving permanent structures, even under varying environmental conditions.
Solution Approach 2:
The system incorporates feedback mechanisms where the localization results are used to adjust the threshold parameters for transient element removal. By continuously monitoring localization consistency and adjusting the threshold accordingly, the system optimizes the balance between removing transient elements and retaining permanent structures.
Data Source
AI summary
Among other things, techniques are described for identifying, in a light detection and ranging (LiDAR) scan line, a first LiDAR data point and a plurality of LiDAR data points within a vicinity of the first LiDAR data point. The techniques may further include identifying, based on a comparison of the first LiDAR data point to at least one LiDAR data point of the plurality of LiDAR return points, a coefficient of the first LiDAR data point, wherein the coefficient is related to image smoothness. The techniques may further include identifying, based on a comparison of the coefficient to a threshold, whether to include the first LiDAR data point in an updated LiDAR scan line, and then identifying, based on the updated LiDAR scan line, a location of the autonomous vehicle.