3D Safety Sensing With Digital Twin Point Cloud Subtraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial safety systems face challenges in accurately detecting hazardous conditions due to the complexity of analyzing comprehensive 3D point cloud data in dynamic environments, often resulting in false safety trips and reduced accuracy.
Innovation Solution
A functional safety system that utilizes a digital twin of an industrial automation system to emulate its behavior, generating shadow point cloud data which is subtracted from measured data by TOF sensors, resulting in reduced point cloud data that focuses only on anomalous entities, thereby improving hazard detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If comprehensive 3D point cloud data is analyzed in dynamic industrial environments, then hazard detection coverage is improved, but false safety trips increase and detection accuracy decreases
Solution Approach 1:
The patent segments the point cloud data into two distinct sets: a first set representing expected objects (from the digital twin model) and a second set representing actual measured objects (from TOF sensors). By processing these segments separately and comparing them, the system isolates hazardous conditions without being overwhelmed by the complexity of comprehensive data analysis.
Solution Approach 2:
The patent extracts only the relevant information needed for safety analysis by generating a difference set between the first point cloud data (expected objects) and the second point cloud data (measured objects). This extraction approach removes unnecessary data about normal operational elements, focusing analysis only on anomalous or hazardous conditions.
2Reliability
If comprehensive 3D point cloud data is analyzed, then hazard detection coverage is improved, but false safety trips increase
Solution Approach 1:
The patent creates a digital twin model that generates a first point cloud data set serving as a virtual copy or representation of expected objects in the industrial environment. This copying approach allows the system to compare expected versus actual conditions without directly processing all raw sensor data, reducing false positives caused by normal operational variations.
Solution Approach 2:
The patent implements a feedback mechanism where the digital twin model continuously updates the expected point cloud data based on the industrial controller's operational state. This feedback loop ensures that the system adapts to legitimate changes in the environment, preventing false safety trips while maintaining accurate hazard detection.
3Measurement precision
If digital twin emulation is used to generate shadow point cloud data, then hazard detection accuracy is improved, but system complexity increases
Solution Approach 1:
The patent introduces a digital twin model as an intermediary component that mediates between the industrial controller and the TOF sensors. This intermediary generates the first point cloud data set, serving as a reference framework that simplifies the comparison process and improves measurement precision without requiring direct complex interactions between sensors and control systems.
Solution Approach 2:
The patent performs preliminary action by pre-generating the first point cloud data set from the digital twin model before conducting safety analysis. This preliminary preparation of expected object data allows the system to quickly compare against actual measurements, improving detection precision while organizing system complexity into manageable preprocessing and analysis stages.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances the accuracy of hazard detection by filtering out known elements, reducing false safety triggers and improving the reliability of safety actions in dynamic industrial environments.
Implementation Method 1
measured point cloud data generated by a TOF sensor that monitors an area comprising the industrial automation system
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A functional safety system performs safety analysis on three-dimensional point cloud data measured by a time-of-flight (TOF) sensor that monitors a hazardous industrial area that includes an automation system. To reduce the amount of point cloud data to be analyzed for hazardous conditions, the safety system executes a real-time emulation of the automation system using a digital twin and live controller data read from an industrial controller that monitors and controls the automation system. The safety system generates simulated, or shadow, point cloud data based on the emulation and subtracts this simulate point cloud data from the measured point cloud data received from the TOF sensor. This removes portions of the point cloud data corresponding to known or expected elements within the monitored area. Any remaining entities detected in the reduced point cloud data can be further analyzed for safety concerns.