Command-Level Process Error Identification in Robotic Time Series
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for monitoring and optimizing robotic processes are limited by their reliance on conventional time series analysis, which requires expert knowledge, is complex to program, and cannot detect complex events, leading to time-consuming trial-and-error optimizations.
Innovation Solution
A method that involves detecting data time series representing process parameters, assigning these series to parts of the process program, and using algorithms to determine results and intermediate results, allowing for the evaluation, monitoring, and optimization of robotic processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If conventional time series analysis with fixed limits is used, then error detection speed is improved, but the ability to detect complex events deteriorates
Solution Approach 1:
The patent segments the time series data into individual command-level segments, each associated with specific process commands. This segmentation allows the system to apply different analysis methods to different segments, enabling both fast detection for simple events and sophisticated analysis for complex events without compromising overall performance.
Solution Approach 2:
The system dynamically adapts the analysis method based on the complexity of the event being detected. For simple threshold violations, fast conventional methods are used. For complex patterns, machine learning models are applied. This dynamic approach optimizes the balance between detection speed and capability.
2Adaptability or versatility
If machine learning is applied to entire time series, then complex event detection is improved, but the ability to draw conclusions about specific program commands deteriorates
Solution Approach 1:
The patent divides the entire time series into smaller command-level segments, each linked to specific process commands. Machine learning models are applied to these segmented data, allowing the system to detect complex events while maintaining the ability to trace findings back to specific commands, thus preserving diagnostic information.
Solution Approach 2:
The system introduces an intermediary layer that maps machine learning predictions back to the original command structure. This intermediary maintains the connection between the abstract model outputs and the concrete process commands, enabling both complex detection and specific diagnostics.
3Reliability
If trial-and-error optimizations are used, then process optimization is possible, but time consumption increases
Solution Approach 1:
The system performs preliminary analysis by continuously monitoring and storing command-level data during normal operation. When optimization is needed, this pre-collected data is immediately available for analysis, eliminating the need for time-consuming trial-and-error experiments and enabling faster process optimization.
Solution Approach 2:
The system implements continuous feedback loops where process data is constantly monitored, analyzed, and used to automatically adjust process parameters. This automated feedback mechanism replaces manual trial-and-error optimization, significantly reducing the time required while maintaining or improving optimization quality.
4Measurement precision
If expert knowledge is required for time series analysis, then analysis accuracy is improved, but system complexity and programming difficulty increase
Solution Approach 1:
The system employs machine learning models that automatically learn optimal analysis parameters and thresholds from historical data without requiring manual expert configuration. The models self-adjust to process characteristics, eliminating the need for expert knowledge in programming while maintaining high analysis accuracy.
Solution Approach 2:
The system dynamically adjusts analysis parameters based on learned patterns from data rather than using fixed expert-defined parameters. This adaptive parameter adjustment maintains accuracy while simplifying the system, as the parameters are automatically optimized rather than manually configured.
Data Source
AI summary
A method for evaluating and/or monitoring a process, in particular a robotic process, includes detecting at least one data time series that describes at least one parameter of the process, and wherein the data time series is created by the process, which executes a process program with process commands, and wherein the at least one data time series is assigned to a part of the process program, in particular a process command or a part of the process commands of the process program. The method further includes determining a result using a first algorithm or at least a part of an algorithm based on the at least one data time series, wherein the result describes a state of the process, and wherein the result can be assigned, in particular is assigned, to the part of the process program, in particular the process command or the part of the process commands of the process program.