Welding Robot Program Generation from Marker-Based Point Clouds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing work program production systems for industrial robots require pre-defined layout patterns and elements, limiting flexibility in generating correct programs for different layouts or elements.
Innovation Solution
A system that includes a photographing terminal to capture images of objects, a coordinate system setting unit to establish a user coordinate system based on markers, and a program production unit to generate work programs allowing industrial robots to perform tasks while avoiding interference with plotted point group data, using a distance measurement sensor to detect specific positions and adjust operation trajectories.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If pre-defined layout patterns and elements are used for work program generation, then program generation efficiency is improved, but flexibility in handling various layout patterns and elements deteriorates
Solution Approach 1:
The system creates a virtual copy of the actual workspace layout by photographing the real layout and generating corresponding virtual layout data. This virtual copy can be freely adapted to match any actual layout without requiring pre-defined patterns, thus achieving both efficiency (through automated copying) and flexibility (through customizable virtual representations).
Solution Approach 2:
The patent replaces the mechanical system of manually selecting and configuring pre-defined layout patterns with an optical-based system that automatically captures and reproduces actual layouts through photographing and image processing. This substitution eliminates the constraint of pre-defined patterns while maintaining automated program generation.
2Ease of operation
If pre-defined layout elements are selected in advance, then program generation process is simplified, but ability to handle new or different layout elements deteriorates
Solution Approach 1:
The system enables self-service by automatically capturing layout information through photographing and autonomously generating the corresponding virtual layout data and work programs. This eliminates the need for manual selection and configuration of layout elements, allowing the system to handle any layout element that can be captured in the photograph without requiring advance preparation.
Solution Approach 2:
The patent changes the fundamental parameter of layout representation from fixed pre-defined categories to flexible photographed images that can represent any layout configuration. By transforming the input from structured pre-defined elements to unstructured photographic data, the system gains the ability to handle any layout element while maintaining automated processing through image analysis.
3Reliability
If virtual layout correction is performed automatically, then robot operation reliability is improved, but preparation time and system complexity increase
Solution Approach 1:
The system performs preliminary action by automatically detecting and correcting layout element positions and robot trajectories in the virtual space before actual robot execution. This preliminary virtual correction ensures that the robot operation will be reliable without requiring complex real-time adjustment mechanisms, as the corrections are pre-computed based on photographed layout data.
Solution Approach 2:
The patent introduces a virtual space as an intermediary between the physical layout and robot execution. This virtual intermediary layer handles all complexity of layout analysis, element detection, and trajectory correction, isolating the robot control system from complexity while ensuring reliable operation through pre-simulated corrections.
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
Enables flexible handling of various layout patterns and elements, allowing industrial robots to perform tasks accurately and avoid interference regardless of object disposition or obstacles, improving work accuracy and reliability.
Implementation Method 1
a photographing terminal that photographs an image including an object to be worked
Implementation Method 2
point group data acquired by a distance measurement sensor that measures a distance to the object to be worked
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The work program production system includes a photographing unit that photographs an image including an object to be welded, a coordinate system setting unit that sets a user coordinate system based on a marker included in the image photographed by the photographing unit, a point-group-data plotting unit that detects a specific position of the marker on the basis of the image, sets the detected specific position on point group data acquired by a distance measurement sensor that measures a distance to the object to be welded, and plots, in the user coordinate system, the point group data to which coordinates in the user coordinate system using the set specific position as an origin are given, and a program production unit that produces a welding program so as to allow a welding robot virtually placed in the user coordinate system to perform a welding operation on the basis of the point group data plotted in the user coordinate system, while avoiding interference with the point group data.