Road Boundary Detection Using Point Cloud Clustering and Line Fitting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing road boundary detection techniques face challenges in accurately detecting road boundaries when there are shielding objects like parked vehicles and in scenarios with fine undulations on the road surface, leading to erroneous detections and decreased accuracy.
Innovation Solution
A road boundary detection device and method that processes point cloud data to remove erroneous line detections caused by shielding objects and analyzes height differences robustly, even in the presence of fine undulations, by clustering candidate points, reducing adjacent clusters, fitting lines, and connecting them to output accurate road boundary information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If point cloud data is processed using conventional road boundary detection techniques, then road boundary lines can be detected, but erroneous detections occur when shielding objects like parked vehicles are present
Solution Approach 1:
The patent segments the point cloud data processing into multiple distinct stages: candidate point detection, clustering, adjacent cluster reduction, line fitting, and line connection. This segmentation allows each stage to address specific aspects of the detection problem, with the adjacent cluster reduction specifically targeting erroneous detections from shielding objects by removing clusters that don't conform to expected road boundary patterns
Solution Approach 2:
The patent performs preliminary clustering of candidate points before line fitting, and performs adjacent cluster reduction to remove erroneous clusters before the final line fitting stage. This preliminary processing prepares the data by eliminating obvious errors from shielding objects before the main detection algorithm operates, improving overall accuracy
2Productivity
If point cloud data from rough road surfaces with fine undulations is processed, then detection can proceed, but points from raised portions are erroneously recognized as steps
Solution Approach 1:
The patent applies local quality analysis by examining the distribution characteristics of points within each cluster and comparing adjacent clusters. The adjacent cluster reduction method evaluates local point density and height difference patterns to distinguish between legitimate road boundaries and false positives from road undulations, applying different processing criteria to different local regions
Solution Approach 2:
The patent performs preliminary clustering and adjacent cluster reduction before final line fitting, which allows erroneous points from road undulations to be identified and removed in advance. This preliminary filtering prevents these points from interfering with the final road boundary extraction
3Ease of manufacture
If conventional line fitting is applied to clustered points, then road boundary lines can be generated, but erroneous lines from shielding objects are not removed
Solution Approach 1:
The patent segments the line detection process into distinct phases: candidate point detection, clustering, adjacent cluster reduction, line fitting, and line connection. The adjacent cluster reduction phase specifically targets and removes clusters that would generate erroneous lines from shielding objects, while preserving valid road boundary clusters for line fitting
Solution Approach 2:
The patent performs adjacent cluster reduction as a preliminary step before line fitting, which removes erroneous clusters from shielding objects before the line fitting algorithm operates. This ensures that only valid road boundary clusters are processed into lines, improving accuracy without significantly complicating the overall process
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
The solution enables the acquisition of road boundary lines with high accuracy, effectively addressing the issues of shielding objects and fine road undulations, and improving the overall precision of road boundary detection.
Implementation Method 1
Point cloud data can be acquired by using a LiDAR (light detection and ranging, laser imaging detection and ranging) as a measuring device
Data Source
AI summary
A road boundary detection device is a road boundary detection device that acquires a set of lines corresponding to a road boundary from point cloud data as road boundary information. The road boundary detection device includes: a candidate point detection unit that detects each point of road boundary candidates corresponding to candidates of a road boundary from the point cloud data; a candidate point clustering unit that clusters each point of the road boundary candidates; an adjacent cluster reduction unit that reduces a cluster from a distribution of points in clusters in an adjacency relationship by using a predetermined cluster reduction method; a line fitting unit that fits one or more straight lines or curved lines to one or more of the clusters and output fitted lines as road boundary candidates; a line connecting unit that connects some of the fitted lines by using a predetermined analysis method; and an information output unit that outputs a calculated line as the road boundary information.