Machine Safeguarding via Sensor Detection Capability Checks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing safety technologies face challenges in reliably determining the detection capability of sensors in dynamic environments, particularly when dealing with poor image quality or interference, which can lead to hazardous misjudgments.
Innovation Solution
A method and safeguarding system that utilize machine learning to evaluate sensor data and determine the quality of the detection capability, allowing for reliable estimation of hazardous situations even in dynamic or poorly lit environments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional sensors (light grids, laser scanners) are used for safety monitoring, then the system satisfies high safety demands (EN13849, IEC61496), but the system cannot reliably determine detection capability in dynamic environments with poor image quality or interference
Solution Approach 1:
The patent transitions from static safety monitoring to dynamic environmental assessment by continuously evaluating sensor data quality. The system adapts to changing lighting conditions, interference levels, and image quality by computing a time-varying detection capability value that reflects current environmental suitability for safe operation.
Solution Approach 2:
The system changes the parameter being monitored from binary safety detection to continuous detection capability assessment. By computing a detection capability value based on multiple sensor parameters (image quality, interference levels, lighting conditions), the system can dynamically adjust operational parameters to maintain safety margins in varying environmental conditions.
2Adaptability or versatility
If neural networks are used to evaluate sensor data in safety engineering, then the system can handle complex dynamic environments, but it becomes difficult to demonstrate detection capability and reveal defects
Solution Approach 1:
The system implements feedback by continuously monitoring sensor data quality and computing a detection capability value that reflects the current reliability of the neural network's assessments. This feedback mechanism allows verification of detection capability by comparing actual sensor data quality against required thresholds, enabling defect revelation when quality deteriorates.
Solution Approach 2:
The patent introduces an intermediary detection capability assessment layer between the neural network and safety decisions. This intermediary computes objective quality metrics (image quality, interference levels) that mediate between the complex neural network processing and the verifiable safety requirements, enabling demonstration of detection capability without sacrificing adaptability.
3Adaptability or versatility
If the sensor data quality is insufficient (poor illumination, interference), then the neural network may fail to recognize hazardous objects, but the system cannot identify the reason for failure
Solution Approach 1:
The system performs preliminary assessment of sensor data quality before neural network processing by evaluating illumination conditions, interference levels, and image quality metrics. This preliminary action identifies potential failure reasons in advance, allowing the system to prepare appropriate responses or alert operators before hazardous objects are missed.
Data Source
AI summary
A method of safeguarding a machine is provided in which a sensor monitors the machine and generates data thereon that are evaluated so that a hazardous situation is recognized and the machine is safeguarded in the event of a hazardous situation, wherein a check is made in a detection capability check whether an estimation of a hazardous situation is possible and the machine is otherwise safeguarded. In this respect, the sensor data are evaluated in a process of machine learning having at least one figure of quality in the detection capability check and an estimation of a hazardous situation is only considered possible with a sufficient figure of quality.