Robot Joint Motor Anomaly Detection From Time-Series Behavior Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robot maintenance support systems, such as PTL 1, are inadequate in accurately detecting impending robot failures, relying on simple diagnostic items like I2 monitor, duty, and peak current, which may not effectively grasp the symptom of failure.
Innovation Solution
A robot failure symptom detection apparatus and method that includes a behavior time-sequence data acquisition unit, evaluation value calculation unit, representative evaluation value generation unit, sequence processing unit, and determination unit, which acquire and process data from a robot's motor operations to generate representative evaluation values and create a determination model for atypicality detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If simple diagnostic items (I2 monitor, duty, peak current) are used for robot maintenance monitoring, then the monitoring system is simple and easy to implement, but the detection accuracy of failure symptoms is insufficient
Solution Approach 1:
The patent segments the monitoring system into multiple functional units: behavior time-sequence data acquisition unit, evaluation value calculation unit, representative evaluation value generation unit, sequence processing unit, and determination unit. Each unit processes specific aspects of motor behavior data, allowing complex analysis to be divided into manageable stages that improve detection accuracy without overwhelming system complexity
Solution Approach 2:
The patent transitions from monitoring simple scalar diagnostic items to analyzing multi-dimensional behavior time-sequence data including current values, speed, acceleration, and positional information across multiple joints. This dimensional expansion enables comprehensive failure symptom detection by capturing complex motor behavior patterns that simple metrics cannot detect
2Reliability
If frequent maintenance is performed to prevent robot breakdown, then robot reliability is improved, but maintenance costs increase
Solution Approach 1:
The patent implements preliminary detection of failure symptoms by continuously monitoring motor behavior and comparing it against learned normal patterns. By detecting abnormalities before they lead to breakdown, the system enables maintenance to be performed at the optimal moment - early enough to prevent failure but not so early that maintenance is performed unnecessarily, thus balancing reliability with cost efficiency
Solution Approach 2:
The system establishes a feedback loop where motor behavior data is continuously acquired, analyzed, and used to update the determination model. This feedback mechanism allows the system to learn from normal operation patterns and progressively improve its ability to detect genuine anomalies, reducing false alarms and unnecessary maintenance while maintaining high reliability
3Measurement precision
If complex diagnostic analysis is implemented to accurately detect failure symptoms, then detection precision is improved, but the complexity of the monitoring system increases
Solution Approach 1:
The determination model performs self-learning by automatically analyzing behavior time-sequence data during normal robot operation and building its own recognition capabilities. This self-service approach eliminates the need for complex manual configuration or expert intervention, allowing high detection precision to be achieved through automated learning while keeping the system architecture relatively simple
Data Source
AI summary
In a robot failure symptom detection apparatus, a behavior time-sequence data acquisition unit performs processing of acquiring behavior time-sequence data relating to a motor of a joint of a robot from a robot operation, for each data collection unit period. An evaluation value calculation unit calculates an evaluation value for the behavior time-sequence data. A representative evaluation value generation unit generates a representative evaluation value representing the evaluation values for each data collection unit period. A sequence processing unit generates a sequence including the representative evaluation values. A determination unit creates a determination model, based on an initial sequence during an initial operation of the robot. After the initial operation, the determination unit inputs determination data including data based on an robot operation after the initial operation into the determination model, and acquires atypicality of the determination data.