Robot Abnormality Detection Using Integrated Torque Trend Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing abnormality detection systems for articulated robots struggle to accurately detect abnormalities due to varying ranges of disturbance torque, often overlooking small changes or incorrectly identifying normal operation as abnormal, due to fixed threshold settings.
Innovation Solution
An abnormality detection device that analyzes time-series data by calculating differences and integrating numerical values to detect continuous trends in disturbance torque, allowing for accurate detection regardless of the range of fluctuation, using defined numerical values to indicate increases or decreases and a moving average comparison with a threshold range.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a fixed threshold is set for disturbance torque comparison, then the detection method is simple, but abnormalities may be overlooked when the variation range is narrow or false positives occur when the threshold is set low
Solution Approach 1:
The patent transforms the static fixed threshold into a dynamic adaptive threshold that automatically adjusts based on the actual variation range of disturbance torque. The system calculates the standard deviation of disturbance torque values and uses this to dynamically set the threshold, allowing the detection criterion to adapt to different operating conditions and variation ranges, thereby improving detection accuracy without requiring complex manual calibration
Solution Approach 2:
The patent changes the parameter of the threshold from a fixed constant to a variable that depends on the statistical properties (standard deviation) of the disturbance torque. By modifying the threshold parameter to be adaptive rather than static, the system can accurately detect abnormalities across different variation ranges while maintaining a relatively simple detection framework
2Reliability
If the threshold is set to a low value to avoid overlooking abnormalities, then sensitivity increases, but normally operating robots are erroneously determined to be abnormal
Solution Approach 1:
The patent implements a feedback mechanism where the system continuously monitors the disturbance torque, calculates its statistical properties (mean and standard deviation), and uses this information to dynamically adjust the threshold. This feedback loop ensures that the threshold adapts to the actual operating conditions, maintaining high sensitivity while reducing false positives by baseing the threshold on real-time statistical data rather than a fixed low value
3Measurement precision
If the threshold is set to a high value to reduce false positives, then false positive rate decreases, but abnormalities are overlooked
Solution Approach 1:
The patent makes the threshold dynamic by tying it to the calculated standard deviation of disturbance torque. This allows the threshold to automatically scale with the variation range - when variation is small, the threshold is low to catch subtle abnormalities; when variation is large, the threshold is high to avoid false positives. This dynamic adjustment resolves the contradiction between sensitivity and false positive rate
Data Source
AI summary
An abnormality detection device detects an abnormality of a device based on time-series data acquired from a device having a movable part. The abnormality determination device determines whether time-series data at a specific time have increased or decreased with respect to time-series data from a certain time prior to the specific time as the specific time is shifted, indicates an increase or decrease of the time-series data by defined numerical values, and detects an abnormality of the device based on integrated values obtained by integrating the defined numerical values.