LIDAR Point Cloud Low-Height Obstacle Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous vehicle systems using LIDAR devices often fail to detect low-height obstacles due to the removal of corresponding low points, which can lead to collisions.
Innovation Solution
A method for determining low-height obstacles based on LIDAR device outputs involves analyzing the differences between neighboring points in the point cloud, specifically calculating quantities of points with significant differences in forward, backward, and up/down directions to identify potential obstacles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If LIDAR points below a threshold distance to the ground plane are removed to account for road surface height variation, then the ground plane estimation accuracy is improved, but low-height obstacles are missed in obstacle detection
Solution Approach 1:
The patent segments the point cloud processing into two distinct pathways: (1) removal of low points for ground plane estimation to achieve accurate road surface modeling, and (2) retention and analysis of low points through neighbor-based comparison for obstacle detection. This segmentation allows each processing stream to serve its specific function optimally without compromising the other.
Solution Approach 2:
The patent introduces an intermediary comparison mechanism that evaluates each low point against its neighboring points at the same height level. This intermediary step acts as a filter to distinguish between points representing road surface variation (which can be removed) and points representing low-height obstacles (which must be retained), thereby resolving the contradiction between ground plane accuracy and obstacle detection reliability.
2Reliability
If all LIDAR points are retained for obstacle detection, then low-height obstacles can be detected, but ground plane estimation accuracy deteriorates due to height variation of the roadway
Solution Approach 1:
The patent segments the point cloud processing into two distinct pathways: (1) removal of low points for ground plane estimation to achieve accurate road surface modeling, and (2) retention and analysis of low points through neighbor-based comparison for obstacle detection. This segmentation allows each processing stream to serve its specific function optimally without compromising the other.
Solution Approach 2:
The patent introduces an intermediary comparison mechanism that evaluates each low point against its neighboring points at the same height level. This intermediary step acts as a filter to distinguish between points representing road surface variation (which can be removed) and points representing low-height obstacles (which must be retained), thereby resolving the contradiction between ground plane accuracy and obstacle detection reliability.
3Device complexity
If low points are removed before obstacle detection to simplify processing, then processing complexity is reduced, but detection completeness of low-height obstacles is lost
Solution Approach 1:
The patent extracts only the critical information needed for obstacle detection from the low points - specifically, the relative height differences between neighboring points. By taking out only this essential comparative data rather than processing all low points equally, the system maintains detection completeness while avoiding the complexity of processing every single low point in detail.
Solution Approach 2:
The patent applies partial action by performing neighbor-based comparison only on a limited number of neighboring points for each low point, rather than analyzing all possible relationships. This partial approach provides sufficient information to distinguish obstacles from road surface variations while keeping processing complexity manageable.
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
This approach allows for the detection of low-height obstacles that were previously missed, enabling the autonomous vehicle to avoid collisions by accurately identifying and responding to these obstacles.
Implementation Method 1
Individual points in the point cloud can be determined by transmitting a laser pulse and detecting a returning pulse, if any, reflected from the object
Implementation Method 2
determining the distance to the object according to the time delay between the transmitted pulse and the reception of the reflected pulse
Data Source
AI summary
A method, apparatus, and system for determining a low-height obstacle based on outputs of a LIDAR device in an autonomous vehicle is disclosed. A point cloud comprising a plurality of points is generated based on outputs of a LIDAR device. For each point within a first number of lowest rings of points, a neighboring point in a same ring to a first direction is determined, and a first and a second coordinate values-related differences are determined. A first, a second, a third, and a fourth quantities are determined based on the first and second differences. In response to determining that the first, the second, the third, and the fourth quantities satisfy a predetermined condition, a low-height obstacle is determined based on the points within the first number of lowest rings of points. Operations of an autonomous vehicle are controlled based at least in part on the determined low-height obstacle.