Robot Motion Path Learning for Welding Cycle Time Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automatic path generation for robots in welding tasks is inefficient as it does not consider factors like robot posture and load distribution, leading to increased cycle time and burden on workers, as the generated paths are not ideal and require manual teaching.
Innovation Solution
An automatic path generation device that uses a machine learning device to learn the correlation between automatically generated temporary motion paths and manually created actual motion paths, allowing it to estimate and derive an efficient motion path by identifying differences and improving the temporary paths.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If motion path is automatically generated using motion planning algorithm, then productivity is improved by reducing manual teaching time, but manufacturing precision deteriorates because robot posture, load distribution, and efficient movement are not considered
Solution Approach 1:
The system performs preliminary action by pre-learning the correlation between temporary motion paths and actual motion paths through machine learning during the preprocessing stage. This learned model is then applied to automatically generate high-quality motion paths without requiring manual teaching, thus improving both productivity and motion path quality simultaneously.
2Manufacturing precision
If manual teaching is performed to create ideal motion paths, then manufacturing precision is improved by considering robot posture and load, but productivity deteriorates due to increased time consumption and worker burden
Solution Approach 1:
The system uses copying by creating a learned model that replicates the knowledge embedded in actual motion paths created by skilled workers. This model copies the implicit expertise regarding robot posture, load distribution, and efficient movement, allowing automatic generation of high-quality paths without manual teaching, thus improving productivity while maintaining motion path quality.
3Device complexity
If temporary motion path is automatically generated, then device complexity is reduced by eliminating manual teaching processes, but ease of operation deteriorates because the generated path is not ideal and requires correction
Solution Approach 1:
The system implements feedback by using the learned model to automatically refine and correct temporary motion paths based on the correlation with actual motion paths. This feedback mechanism ensures that the automatically generated paths are of high quality and directly usable, eliminating the need for manual correction while keeping the process simple.
Data Source
AI summary
An automatic path generation device includes a preprocessing unit creating teacher data based on a temporary motion path which is a motion path between a plurality of motion points where a robot moves and which is automatically generated with a motion planning algorithm and an actual motion path which is a motion path between the motion points and which is created by a skilled worker and a motion path learning unit generating a learned model which has learned a difference between the temporary motion path and the actual motion path with teacher data created by the preprocessing unit.