LiDAR Point Cloud Stitching Using Visibility Grids
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing LiDAR point cloud stitching methods in autonomous vehicles face challenges such as motion artifacts and increased computational load, which can lead to reduced precision in object detection and delayed route planning decisions.
Innovation Solution
The implementation of a visibility grid system to intelligently select LiDAR points for inclusion in the point cloud, supplementing occluded regions with data from secondary LiDAR devices to enhance coverage while minimizing motion artifacts and computational burden.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If LiDAR point cloud stitching is performed using multiple LiDAR devices, then occlusion coverage is improved, but motion artifacts and computational load increase
Solution Approach 1:
The patent segments the point cloud processing by creating a visibility grid that divides the field of view into discrete cells. Each cell is independently evaluated for occlusion status, allowing selective processing of only those regions that require supplementation from secondary LiDAR devices, rather than processing the entire point cloud uniformly.
Solution Approach 2:
The patent applies local quality by treating different regions of the point cloud differently based on their occlusion characteristics. The visibility grid identifies specific occluded cells and applies supplementation only to those regions, while leaving non-occluded regions unchanged, thereby maintaining high precision where needed while reducing computational burden elsewhere.
2Loss of information
If LiDAR point cloud stitching is performed using multiple LiDAR devices, then occlusion coverage is improved, but computational load increases
Solution Approach 1:
The visibility grid segments the processing task into discrete cells, allowing the system to evaluate and process only occluded regions rather than the entire point cloud. This segmentation dramatically reduces computational load while maintaining comprehensive occlusion coverage.
Solution Approach 2:
The patent implements partial action by supplementing only the occluded portions of the point cloud identified through the visibility grid, rather than processing the complete point cloud. This selective approach reduces computational burden to the minimum necessary level while still achieving the goal of comprehensive occlusion coverage.
3Measurement precision
If LiDAR points are selectively selected using visibility grid, then motion artifacts are reduced, but processing complexity increases
Solution Approach 1:
The visibility grid acts as an intermediary structure that simplifies the selection process. Instead of implementing complex algorithms to directly identify and select appropriate LiDAR points, the system uses the visibility grid as a mediating layer that pre-organizes spatial information and identifies occluded regions, making the subsequent point selection process more straightforward and 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 improves the precision of object detection by reducing occlusions and motion artifacts, while also reducing latency and computational load, thereby enhancing the efficiency of autonomous vehicle navigation and route planning.
Implementation Method 1
a light detection and ranging (LiDAR) sensor can be used to determine ranges (variable distance) of one or more targets by directing a laser to a surface of an entity (e.g., a person, an object, a structure, an animal, etc.) and measuring the time for light reflected from the surface to return to the LiDAR
Data Source
AI summary
Systems and techniques are provided for implementing LiDAR point cloud stitching. An example method includes determining a visibility grid that is based on a plurality of measurements from a first LiDAR device, wherein the visibility grid includes a plurality of range values corresponding to a field of view of the first LiDAR device; identifying, based on the visibility grid, at least one occluded region in the field of view; and modifying a point cloud that is based on the plurality of measurements to include additional points corresponding to one or more measurements obtained from a second LiDAR device, wherein the one or more measurements correspond to the at least one occluded region.