Programmable Robot Fault Diagnosis Using Motion-Specific Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Collaborative robots in smart factories face challenges in health evaluation due to complex data patterns from diverse task programs, making it difficult to apply existing anomaly detection methods effectively.
Innovation Solution
A method and apparatus that define standard patterns for each program and motion-specific operation of a programmable robot, allowing for fault diagnosis by comparing execution patterns with predefined standards, identifying faults based on threshold differences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If fixed threshold-based anomaly detection methods are used, then the detection process is simple, but the method is difficult to apply to collaborative robots with diverse task programs and complex data patterns
Solution Approach 1:
The patent segments the robot's operation data by program ID and motion type, creating separate standard patterns for each combination. This segmentation allows the system to handle diverse task programs effectively by treating each program-motion combination as an independent evaluation unit, resolving the contradiction between simple implementation and adaptability to diverse programs.
2Measurement precision
If standard patterns are defined for each program and motion, then fault diagnosis accuracy is improved, but the complexity of the diagnostic system increases
Solution Approach 1:
The patent performs preliminary actions by pre-defining standard patterns for each program-motion combination before actual fault diagnosis. During operation, the system simply compares real-time data against these pre-established standards, which improves diagnostic accuracy while maintaining relatively simple real-time processing. The complexity is front-loaded during the pattern definition phase rather than during operation.
Solution Approach 2:
The system changes parameters by creating multiple standard patterns with different program IDs and motion types as classification parameters. This parameter-based organization allows the system to handle complexity through structured data management rather than complex algorithms, improving diagnostic precision across diverse programs while keeping the system architecture manageable.
3Ease of manufacture
If threshold-based anomaly detection is applied to wide range of sensing data, then the detection method is simple, but it becomes difficult to accurately diagnose faults in collaborative robots
Solution Approach 1:
The patent applies local quality by creating specialized standard patterns tailored to each specific program-motion combination rather than using a single universal threshold. Each local pattern is optimized for its specific context, enabling reliable fault diagnosis across diverse sensing data while maintaining the simplicity of threshold-based comparison within each localized context.
Data Source
AI summary
The present invention relates to a method and apparatus for fault diagnosis of a programmable robot, and includes the steps of collecting sensing data corresponding to an operation of a programmable robot, identifying a program and motion related to the sensing data, generating an execution pattern based on the sensing data, extracting a standard pattern corresponding to the identified program and motion, and diagnosing a fault of the programmable robot by comparing the standard pattern with the execution pattern, and the present invention may be applicable as another embodiment.