Robotic Insertion Error Recovery Using Force-Position Anomaly Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robotic assembly systems face challenges in accurately detecting anomalies during insertion processes due to uncertainties and process noise, leading to inefficiencies in error recovery and increased computational complexity in data-driven models.
Innovation Solution
A machine learning-based anomaly detection system using probabilistic functional models that reduce dimensionality by establishing a stable relationship between force and position measurements, employing Gaussian process regression and local Gaussian process models to enable real-time, accurate anomaly detection with reduced computational requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If data-driven models are used for anomaly detection during robotic insertion, then detection accuracy improves, but computational complexity increases
Solution Approach 1:
The patent extracts and focuses only on the critical relationship between force and position measurements during insertion, separating this key anomaly detection mechanism from other potentially complex model components. By concentrating on this specific force-position relationship rather than attempting to model all insertion parameters simultaneously, the system achieves accurate anomaly detection with reduced computational burden.
Solution Approach 2:
The patent transforms the anomaly detection approach by changing from complex multi-parameter models to a simplified force-position relationship model. This parameter transformation maintains detection accuracy for critical anomalies while significantly reducing computational complexity by focusing on the most informative measurement pairings during the insertion process.
2Reliability
If model-based methods are used for fault detection, then robustness to uncertainty improves, but complexity of the underlying technique increases
Solution Approach 1:
The patent employs simple force and position measurements that are readily available from standard sensors during robotic insertion. These basic measurements serve as the foundation for anomaly detection without requiring complex or expensive specialized sensors, achieving robust fault detection through straightforward data collection and analysis.
Solution Approach 2:
The patent replaces complex physics-based mechanical models with a data-driven approach using readily measurable force and position parameters. This substitution eliminates the need for complex theoretical modeling while maintaining robustness through direct measurement of actual insertion conditions, thereby reducing technique complexity while preserving reliability.
Data Source
AI summary
A system for controlling a robotic arm performing insertion of a component along an insertion line accepts measurements of force experienced by the wrist of robotic arm at current position along insertion line and determines probability of value of the force conditioned on the current value of the position according to a probabilistic relationship for the force experienced by the wrist of the robotic arm along the insertion line as a probabilistic function of the positions of the wrist of the robotic arm along the line of insertion. The probabilistic function is learned from measurements of the operation repeatedly performed by one or multiple robotic arms having the configuration of the robotic arm under the control. The system determines a result of anomaly detection based on the probability of the current value of the force and controls the robotic arm based on the result of anomaly detection.