Robotic Anomaly Detection With Self-Updating Analytics Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing anomaly detection systems in industrial robotic systems face challenges such as the need for manual model generation for specific parameters, inability to handle new parameters, high data storage costs, and frequent model updates, which can be costly and require third-party vendor involvement.
Innovation Solution
A method and system that utilize a monitoring module to detect associations between configuration and process parameters, an analysis module to analyze optimal parameters using unsupervised machine learning, and a decision module to recommend actions and update analytics models autonomously, thereby improving anomaly detection accuracy and reducing operational costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual model generation is performed for specific parameters, then anomaly detection accuracy is improved, but device complexity and time consumption increase
Solution Approach 1:
The system performs automatic model generation using unsupervised machine learning algorithms that autonomously analyze process parameters and configuration data without requiring manual intervention. The analytics model automatically identifies patterns and anomalies in the data, eliminating the need for operators to manually create and maintain detection models for each parameter.
Solution Approach 2:
The patent replaces manual model generation processes with automated computational algorithms. Instead of operators manually analyzing parameters and creating models, the system uses unsupervised machine learning algorithms to automatically generate analytics models that detect anomalies based on learned patterns in the data.
2Reliability
If all raw data is uploaded for analysis, then anomaly detection completeness is improved, but data storage costs and transmission time increase
Solution Approach 1:
The system extracts and transmits only the essential features and aggregated statistics derived from process parameters rather than uploading complete raw datasets. The analytics model processes data locally at the robotic system, extracting only the relevant anomaly indicators for transmission to the server, thereby reducing data volume while maintaining detection effectiveness.
3Adaptability or versatility
If analytics models are updated frequently to adapt to parameter changes, then adaptability is improved, but operational costs and system complexity increase
Solution Approach 1:
The analytics model is designed to dynamically adapt to changing parameters through continuous unsupervised learning. The model automatically adjusts its detection thresholds and patterns based on evolving process data without requiring manual reconfiguration or vendor intervention, enabling the system to adapt to parameter changes while maintaining operational simplicity.
4Measurement precision
If third-party vendors are involved for model updates, then model accuracy is improved, but operational costs and update time increase
Solution Approach 1:
The system performs autonomous model updates using unsupervised machine learning algorithms that automatically adapt to new parameters and process data without requiring third-party vendor involvement. The analytics model continuously learns from incoming data streams, maintaining accuracy while eliminating external dependencies for model maintenance and updates.
Data Source
Figure 1
Figure 2A~2B
Figure 3
AI summary
The present invention relates to a method and a system for detecting anomalies in a robotic system in an industrial plant. The robotic system is associated with a computing system configured to detect an anomaly in the robotic system. The computer system monitors configuration parameters of the robotic system and process parameters associated with the robotic system. Further, the computing system detects an association between at least one configuration parameter and at least one process parameter for obtaining optimal configuration parameters and optimal process parameters. The optimal configuration parameters and optimal process parameters are analyzed for detecting an anomaly. At least one parameter among the configuration parameters and the process parameters is identified causing the anomaly. Thereafter, the detected anomaly is validated, valid setpoint is estimated and the estimated valid setpoint is updated in the analytics model. The updated analytics model is subsequently used to detect anomaly accurately.