Robot Learning Control for Machining Quality Under Device Delay
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robot control technologies are limited in improving machining quality as they only focus on robot vibration, neglecting workpiece vibration and machining device performance, leading to suboptimal results in applications like sealing, welding, and laser machining.
Innovation Solution
A robot system with a learning control unit that calculates machining device performance and operation speed corrections based on machining results, ensuring preset acceptable conditions are met, and determining learning completion to enhance overall system performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If robot vibration is eliminated by controlling only the robot tip, then robot vibration is reduced, but machining quality cannot be sufficiently improved due to workpiece vibration and machining device performance limitations
Solution Approach 1:
The system introduces feedback by detecting actual machining results (sealing bead width, thickness, volume) and using this information to calculate machining device performance and generate operation speed correction commands. This closed-loop feedback enables the system to adapt to workpiece vibration and machining device variations, resolving the limitation of open-loop robot tip control.
Solution Approach 2:
The system performs self-diagnosis and self-correction by automatically calculating machining device performance from machining results and generating correction commands without external intervention. The learning control unit continuously improves operation speed based on detected machining quality, enabling the system to self-optimize and overcome limitations of manual tuning.
2Manufacturing precision
If sealant flow rate is changed according to robot operation speed to maintain constant bead width and thickness, then sealing quality is maintained under steady conditions, but rapid changes in robot speed cannot be responded to quickly enough due to machining device response delay
Solution Approach 1:
The system calculates operation speed correction commands in advance based on detected machining results and predicted performance requirements. By proactively adjusting operation speed before machining operations, the system compensates for the inherent response delay in sealant flow rate changes, ensuring sealing quality is maintained during rapid speed transitions.
Solution Approach 2:
The system dynamically adjusts robot operation speed based on real-time machining results and calculated machining device performance. Rather than using fixed flow rate-speed relationships, the system continuously adapts operation speed to match actual machining conditions and device capabilities, enabling rapid response to speed changes while maintaining sealing quality.
3Manufacturing precision
If learning control is performed by trial and error with manual modification by teachers, then machining quality can be improved through skill-based adjustments, but significant time and effort are required
Solution Approach 1:
The system automatically performs learning control by detecting machining results, calculating machining device performance, and generating operation speed correction commands without manual intervention. This self-learning capability eliminates the need for skilled operators to perform time-consuming trial-and-error adjustments, automatically achieving optimal machining quality.
Solution Approach 2:
The system replaces manual skill-based adjustment with automated computational algorithms. The learning control unit uses detection data and performance calculations to automatically determine optimal operation speeds, substituting human expertise and trial-and-error processes with systematic computational methods that are faster and more consistent.
Data Source
AI summary
A learning control unit includes a machining device performance calculation section for calculating performance of a machining device during machining or after machining based on a motion command issued to a robot by a controller, a machining command issued to the machining device by the controller, and machining results measured by a sensor, an operation speed correction information calculation section for calculating operation speed correction information of the robot based on the performance of the machining device so as to satisfy a preset acceptable condition of the machining error, and under an allowable load condition of the robot, and a learning completion determination section for determining whether or not learning has completed based on previous correction information and current correction information.