Laser Scanner Point Clustering via Angular Sequence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for clustering scan points from laser scanners require calculating a large number of distances, which is computationally intensive and resource-heavy, especially in embedded systems used in automotive applications, making them inefficient for object detection in lidar sensor systems.
Innovation Solution
The method exploits the natural sequence of scan points defined by their angles of incidence to efficiently cluster scan points, reducing the number of distances that need to be calculated by considering the sequence and proximity within the scan point neighborhood, rather than calculating distances between all points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional clustering algorithms (e.g., k-nearest neighbor) are used to cluster scan points, then object detection accuracy is improved, but computational complexity and storage requirements increase significantly
Solution Approach 1:
The patent segments the scan points into sequential groups based on their angular order in the point cloud. Instead of treating all points as an unordered set requiring pairwise distance calculations, the method divides them into sequential segments that can be processed independently, reducing computational complexity from O(n²) to O(n).
Solution Approach 2:
The patent performs preliminary sorting of scan points by their angle of incidence before clustering. This preliminary action organizes the data in a sequence that enables efficient local clustering without requiring exhaustive pairwise comparisons, thus reducing computational burden while maintaining detection accuracy.
2Reliability
If traditional clustering algorithms are used to ensure accurate object detection, then detection reliability is improved, but processing time increases due to O(n²) distance calculations
Solution Approach 1:
The method segments the point cloud processing into sequential steps based on angular order, allowing the system to process points in manageable segments rather than requiring all points to be processed simultaneously. This segmentation reduces processing time while maintaining reliable object detection through sequential clustering.
Solution Approach 2:
The patent skips unnecessary distance calculations by leveraging the sequential angular ordering of points. Instead of calculating distances between all possible point pairs, the method only compares adjacent or nearby points in the sequence, rushing through the processing by eliminating redundant calculations.
3Device complexity
If embedded systems with limited resources are used in automotive applications, then system compactness and cost-effectiveness are improved, but computational and storage capabilities are constrained
Solution Approach 1:
The patent changes the processing parameter from unordered point sets to sequentially ordered points based on angle of incidence. This parameter change transforms the computational problem into one that can be solved with linear complexity, enabling embedded automotive systems with limited resources to perform efficient object detection.
Solution Approach 2:
The patent replaces the computationally intensive mechanical approach of calculating all pairwise distances with a more efficient algorithmic approach that uses sequential processing. This substitution reduces the computational burden on embedded systems while maintaining detection accuracy.
Data Source
AI summary
A method for recognizing an object in a surround of a laser scanner by clustering scan points of the laser scanner is disclosed. The method includes using the laser scanner to create a multiplicity of successive scan points and using at least one computing unit to determine, in a manner dependent on the sequence, at least one cluster of scan points containing some of the multiplicity of successive scan points. Each scan point is characterized by an angle of incidence. A sequence of the multiplicity of successive scan points is defined by the angles of incidence.