Robot Processing Curve Construction from Point Cloud Boundary
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for extracting and reconstructing point cloud boundary characteristics are inefficient and prone to inaccuracies, particularly in processing curves for thin-walled parts with small curvature, leading to poor measurement consistency and high costs due to reliance on third-party software and complex algorithms.
Innovation Solution
A method involving multiple clustering of projection points within a plane to fit straight lines and corners, followed by back-projection to construct an actual processing curve in three-dimensional space, eliminating the need for third-party software and enhancing precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If curved surface reconstruction method is used to extract boundary characteristics, then boundary characteristic line can be obtained, but calculation amount is large and algorithm efficiency is low
Solution Approach 1:
The patent segments the point cloud boundary into multiple chains of adjacent boundary points, processing each chain independently through projection and clustering operations. This divides the complex boundary extraction problem into manageable segments, reducing overall calculation complexity while maintaining extraction accuracy
Solution Approach 2:
The patent extracts only the essential boundary characteristic points and their projection relationships, discarding redundant information. By focusing solely on boundary point chains and their projected clustering patterns, the method achieves efficient extraction without unnecessary computational overhead
2Productivity
If point cloud boundary is used directly for processing curve guidance, then actual processing curve can be obtained, but fitting error at rounded and sharp corners cannot be controlled
Solution Approach 1:
The patent performs preliminary projection of boundary points onto a standard plane before clustering and curve fitting. This preliminary geometric transformation establishes a standardized coordinate system that facilitates accurate corner detection and fitting, preventing subsequent fitting errors at critical features like rounded and sharp corners
Solution Approach 2:
The patent transforms boundary point coordinates from three-dimensional space to a projected two-dimensional plane, changing the spatial parameters to enable more accurate clustering and corner detection. This parameter transformation allows for precise control of fitting operations at corner regions
3Measurement precision
If third-party software is used to plan actual processing curve, then processing curve can be obtained, but cost is high and procedure is complicated
Solution Approach 1:
The patent implements a self-contained algorithmic system that performs boundary extraction, projection, clustering, and curve fitting operations entirely in-house without requiring external third-party software. The method uses intrinsic geometric relationships and mathematical operations to generate processing curves, eliminating dependency on complex external systems while maintaining high accuracy
Data Source
AI summary
The disclosure discloses a method for constructing actual processing curve of part with small curvature based on a point cloud boundary. The method includes: (a) encrypting a three-dimensional ordered boundary curve of a part to be processed; (b) fitting encrypted boundary points into a plane, and projecting each boundary point into the plane to obtain a projection point; (c) performing Euclidean cluster within the plane to obtain point sets, and fitting the obtained point sets into a straight line; (d) performing the Euclidean cluster on projection points that are not fitted into the straight line to obtain corner point sets, and fitting a sharp corner or a rounded corner of the corner point sets to obtain a fitted boundary curve within the plane; (e) mapping the fitted boundary curve to a curved surface of the three-dimensional ordered boundary curve to obtain an actual processing curve of the part to be processed.