LiDAR Point Cloud Filtering for Ground and Atmospheric Noise
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing LiDAR systems struggle with accurately differentiating between atmospheric conditions and objects due to issues like high-density points from rain clouds leading to false object predictions and the inability to identify and filter out the ground portion in point clouds, resulting in inaccurate object identification.
Innovation Solution
A method and system that utilize a LiDAR system to identify and isolate ground points, filter out atmospheric conditions, and classify high-intensity and low-intensity points, using secondary scans like cameras and RADAR to enhance accuracy, generating a final processed point cloud with obstacle points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If density-based filtering is used to remove low-density points, then sparsity of point cloud is improved, but false objects are predicted due to high-density points from rain clouds
Solution Approach 1:
The patent changes the filtering parameters from density-based to intensity-based thresholds. By setting a minimum intensity threshold and removing points below this threshold, the system distinguishes between valid object reflections (higher intensity) and atmospheric interference (lower intensity), resolving the contradiction between sparsity and detection accuracy.
Solution Approach 2:
The patent replaces the density-based filtering mechanism with an intensity-based filtering mechanism. Instead of using spatial density calculations, the system uses optical intensity measurements to differentiate between objects and atmospheric conditions, leading to more reliable object detection.
2Reliability
If intensity analysis is applied to all points in point cloud, then object identification is improved, but false predictions occur due to mixed high-intensity and low-intensity points from objects
Solution Approach 1:
The patent segments the point cloud into distinct categories based on intensity thresholds: high-intensity points (potential objects), low-intensity points (atmospheric conditions), and ground points. This segmentation allows each point type to be processed differently, improving both object identification accuracy and classification precision.
Solution Approach 2:
The patent applies different filtering criteria to different regions of the point cloud. Ground points are filtered using intensity and density thresholds, while airborne points are filtered using intensity thresholds. This local differentiation allows the system to handle the mixed intensity characteristics of objects versus atmospheric conditions appropriately.
3Adaptability or versatility
If ground points are not filtered out, then complete point cloud analysis is maintained, but difficulty increases in identifying objects and atmospheric conditions at ground borders
Solution Approach 1:
The patent extracts and removes ground points from the point cloud analysis. By identifying ground points through intensity and density thresholds and filtering them out, the system eliminates the interference ground points cause in detecting objects and atmospheric conditions, particularly at ground borders.
Solution Approach 2:
The patent performs preliminary filtering of ground points before conducting object and atmospheric condition detection. This preliminary action removes the interfering ground points early in the processing pipeline, simplifying subsequent detection tasks and improving accuracy at ground borders.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances the reliability of LiDAR systems by accurately distinguishing between atmospheric conditions and objects, improving object detection and navigation safety in autonomous vehicles.
Implementation Method 1
A LiDAR sensor is configured to emit light, which strikes material (e.g., objects) within the vicinity of the LiDAR sensor. Once the light comes into contact with the material, the light is deflected. Some of the deflected light bounces back to the LiDAR sensor.
Implementation Method 2
LiDAR's reliability, particularly pertaining to its use with autonomous vehicles, can be affected by various atmospheric conditions such as, e.g., rain and fog, because these conditions can absorb and reflect light
Implementation Method 3
rain and fog, because these conditions can absorb and reflect light
Data Source
AI summary
Systems and methods are provided for filtering atmospheric conditions from LiDAR point clouds. The method includes generating, using a LiDAR system, at least one point cloud, wherein the LiDAR system includes a processor. The method further includes, using the processor, identifying and isolating one or more ground points within a point cloud of the at least one point cloud, wherein the one or more ground points indicate a ground portion within an environment of the point cloud, filtering out the ground portion from the point cloud, generating an initial processed point cloud, identifying and isolating one or more atmospheric condition points within the initial processed point cloud, wherein the one or more atmospheric condition points indicate one or more atmospheric conditions within an environment of the processed point cloud, and filtering out the atmospheric condition points from the initial processed point cloud, generating a final processed point cloud.