Industrial Robot Torque Monitoring via Feed Forward Feedback Difference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial robots' performance monitoring is challenging due to subjective diagnosis methods and difficulty in detecting subtle changes in control errors caused by wear in drive components, leading to time-consuming fault identification and costly stand-still periods.
Innovation Solution
An automated method that calculates feed forward and feedback torques for actuators based on reference values and mathematical models, monitoring the difference between them to detect non-normal changes and predict mechanical issues, such as wear in motors, gears, and bearings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If automated monitoring using control error is implemented, then monitoring coverage is improved, but detection precision deteriorates due to large natural variations in control error masking wear-induced changes
Solution Approach 1:
The invention extracts the feedback torque component from the total control torque by calculating the difference between reference torque and feed-forward torque. This isolation technique separates the wear-related feedback torque variations from the dominant feed-forward torque variations, enabling precise detection of component wear while maintaining automated monitoring.
2Adaptability or versatility
If operator diagnosis based on personal experience is used, then diagnostic flexibility is improved, but diagnostic consistency deteriorates due to subjective measures varying between operators
Solution Approach 1:
The invention implements automated feedback monitoring that continuously compares actual torque with reference torque and automatically identifies wear patterns. This systematic feedback mechanism replaces subjective operator judgment with objective, consistent automated analysis, ensuring uniform diagnostic standards across all operators while maintaining the ability to handle diverse robot configurations.
Solution Approach 2:
The invention substitutes the mechanical diagnostic process (operator inspection and analysis) with an automated electronic monitoring system that calculates torque differences and detects wear. This replacement eliminates human subjectivity and variability while maintaining diagnostic capability across different operators and situations.
3Measurement precision
If comprehensive monitoring of all robot components is implemented, then detection capability is improved, but system complexity deteriorates due to multiple measurement points and data sources
Solution Approach 1:
The invention creates a universal monitoring method that uses the existing torque control infrastructure for multiple purposes: both for robot control and for wear detection. By leveraging the feed-forward and feedback torque calculations already present in the control system, the invention achieves comprehensive component monitoring without adding separate measurement systems or increasing overall system complexity.
Data Source
AI summary
A method for monitoring the condition of an industrial robot having a plurality of links movable relative to each other, and a plurality of actuators controlling the movements of the links. A feed forward torque is calculated for at least one of the actuators based on reference values for the position of the actuator and a mathematical model of the robot calculating a feedback torque for the actuators based on measured values from the actuators and reference values for the position of the actuator. A torque is calculated for the actuator at least based on the feedback torque. A difference is monitored between the calculated torque for the actuator and the feed forward torque. It is determined whether the difference is normal or non-normal, and based thereon monitoring the condition of the robot.