Robot Learning Control System Vibration Convergence Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current learning control systems for robots often perform a predetermined number of learning controls, which may not be optimal, leading to redundant operations and inefficiencies in reducing vibrations at the robot's tip end portion, resulting in wasted time and suboptimal vibration reduction.
Innovation Solution
A control system that includes a servomotor-driven device, a control device with an operation control unit, a learning control processing unit, and a sensor to detect position-related values, which estimates time-series vibration data, calculates vibration correction amounts, and determines the optimal number of learning controls based on convergence determination values to minimize vibrations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If learning control is performed a predetermined number of times, then the control process is simplified, but the vibration may not be minimized optimally and redundant learning controls waste time
Solution Approach 1:
The system uses vibration detection feedback to determine when learning control should stop. The vibration detection unit continuously monitors vibration during learning control, and when vibration falls below a threshold, the system automatically terminates learning control, preventing redundant operations and optimizing both productivity and time efficiency
Solution Approach 2:
The predetermined number of learning controls is replaced with a dynamic termination condition based on actual vibration levels. The system adapts the number of learning control iterations based on real-time vibration feedback, making the process flexible and optimized for each specific situation rather than following a fixed predetermined count
2Loss of time
If learning control is performed fewer times, then time is saved, but vibration may not be sufficiently reduced
Solution Approach 1:
The vibration detection unit provides continuous feedback during learning control to ensure vibration is sufficiently reduced. The system monitors vibration levels in real-time and only terminates learning control when vibration falls below a predetermined threshold, guaranteeing adequate vibration reduction while avoiding unnecessary additional learning control cycles
3Device complexity
If a fixed predetermined number of learning controls is used, then the control system is simpler, but it cannot adapt to different vibration convergence characteristics
Solution Approach 1:
The system incorporates vibration detection feedback to automatically adapt to different vibration convergence characteristics. By monitoring actual vibration levels and comparing them against thresholds, the system dynamically determines the appropriate number of learning control iterations for each specific situation, eliminating the need for complex predetermined schedules while maintaining simplicity
Solution Approach 2:
The control system performs self-adjustment by using its own vibration detection capability to determine when learning control should terminate. The system serves itself by automatically recognizing when vibration has converged to an acceptable level, eliminating the need for external intervention or complex predetermined control parameters
Data Source
AI summary
A robot control system includes an operation control unit, a learning control processing unit and a storage unit. Whenever the operation control unit performs a single learning control, the learning control processing unit stores the number of learning controls, which indicates how many learning controls have been performed, and obtained time-series vibration data in correspondence with each other in the storage unit. The learning control processing unit calculates a convergence determination value to determine whether or not a vibration of a certain portion of a robot converges based on the time-series vibration data at each number of learning controls stored in the storage unit, and determines the number of learning controls having a minimum convergence determination value, out of the calculated convergence determination values, as the optimal number of learning controls.