Partial 3D Workpiece Data for Faster Robot Vision Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for generating three-dimensional data for workpiece detection in robot vision face challenges such as requiring trial and error in online teaching due to glossiness and illumination issues, and in offline teaching, full three-dimensional data slows down the matching process and is not applicable to all applications like bulk picking.
Innovation Solution
A three-dimensional data generation device that arranges a virtual workpiece and virtual vision sensor in a virtual space to generate partial three-dimensional data within the imaging range, reducing the number of steps and enabling high-speed matching processing by calculating the arrangement relationship between the workpiece and vision sensor.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full three-dimensional data representing the entire shape of the workpiece is used, then comprehensive detection capability is improved, but matching processing speed deteriorates
Solution Approach 1:
The patent divides the full three-dimensional workpiece data into multiple partial three-dimensional data sets, each representing a specific portion of the workpiece that corresponds to the imaging range of the vision sensor. This segmentation enables faster matching processing while maintaining detection accuracy for the visible portions.
Solution Approach 2:
The patent extracts only the relevant partial three-dimensional data that corresponds to the imaging range of the vision sensor, discarding unnecessary portions. This extraction process reduces the data volume for matching while preserving the essential detection information needed for the specific application.
2Measurement precision
If online teaching is performed with actual workpieces, then data accuracy is improved, but time consumption and complexity increase due to trial and error
Solution Approach 1:
The patent performs preliminary generation of partial three-dimensional data in advance, before actual robot operation. By pre-processing the workpiece data and organizing it into partial data sets corresponding to vision sensor ranges, the system eliminates the need for time-consuming trial and error during online teaching while maintaining data accuracy.
3Productivity
If partial three-dimensional data is used for high-speed matching, then processing speed is improved, but detection reliability may deteriorate if the data does not represent the actual workpiece shape
Solution Approach 1:
The patent ensures that each partial three-dimensional data set accurately represents the specific local portion of the workpiece that corresponds to the vision sensor's imaging range. By matching the spatial characteristics and geometric features of each partial data set to its corresponding physical region, the system maintains detection reliability while achieving high-speed processing.
Data Source
AI summary
A three-dimensional data generation device includes a virtual space arrangement section for arranging a virtual workpiece and a virtual vision sensor in a virtual space so that a part of the virtual workpiece fits in an imaging range of the virtual vision sensor, an arrangement relationship calculation section for calculating an arrangement relation between the arranged virtual workpiece and the virtual vision sensor, and a partial three-dimensional data generation section for generating partial three-dimensional data representing a partial shape of a workpiece which fits in the imaging range based on the calculated arrangement relationship.