Subterranean Safety Monitoring With Edge AI Risk Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Subterranean environments face challenges with inconsistent network connectivity, leading to unreliable data collection and manual, delayed operational analysis, hindering real-time response to safety issues and operational optimization.
Innovation Solution
A context-based safety system utilizing sensors and a processor to collect and analyze operating data, applying vision-language models to identify abnormal and risky activities, generating recommendations for improving safety and productivity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If network infrastructure is limited to main travel passages due to cost and time constraints, then infrastructure cost and deployment time are reduced, but data collection reliability and coverage are worsened
Solution Approach 1:
The patent introduces local edge computing devices as intermediaries between sensors in network-unavailable areas and the central system. These edge devices collect and pre-process data locally, then transmit aggregated data when network connectivity is available, acting as a mediator that bridges the gap between network-constrained environments and data collection requirements
Solution Approach 2:
The system performs preliminary data collection and processing at the edge before network transmission is needed. Sensors continuously collect data and store it locally in buffer memory, performing preliminary actions (data collection, buffering) in advance so that data is ready for transmission when network connectivity becomes available
2Device complexity
If manual data reporting is used in areas with inconsistent network access, then infrastructure complexity is reduced, but operational analysis timeliness and accuracy are worsened
Solution Approach 1:
The patent replaces manual reporting mechanisms with automated sensor-based data collection systems. Sensors automatically capture operational data and edge computing devices automatically process and transmit this data, substituting the mechanical manual reporting process with an automated electronic system that eliminates human delay and subjectivity
Solution Approach 2:
The system implements continuous feedback loops where sensors monitor operational conditions, edge devices analyze data in real-time, and the system automatically adjusts operations based on detected anomalies. This feedback mechanism enables real-time operational analysis without manual intervention, reducing time loss while maintaining manageable system complexity
3Adaptability or versatility
If equipment operates in network-unavailable areas for extended periods, then operational flexibility and mobility are improved, but real-time monitoring and safety response are worsened
Solution Approach 1:
The patent segments the monitoring system into distributed sensor nodes and edge computing devices that can operate independently in network-unavailable areas. Each segment (sensor + edge device) functions autonomously, collecting and buffering data locally, which allows equipment mobility while maintaining monitoring capability through modular, distributed architecture
Solution Approach 2:
Edge computing devices perform preliminary data processing and anomaly detection before network transmission is needed. The system pre-processes data locally, identifies potential safety issues in advance, and prepares alerts for transmission, ensuring that safety monitoring reliability is maintained even when equipment operates in network-unavailable areas
Data Source
AI summary
There is provided context-based safety systems, and methods for operating thereof, for a subterranean environment. An example context-based safety system for a subterranean environment includes: a plurality of sensors configured to collect a set of operating data; and a processor configured to: continuously receive the operating data from sensors, the set of operating data including at least one visual data; receive a user prompt from a user defining a risk assessment in respect of the operating data; apply the user prompt to a vision-language model to: identify one or more abnormal activities observed from the set of operating data, each abnormal activity being unexpected within a safety context associated with the subterranean environment and an activity type of that abnormal activity; and identify one or more risky activities from the abnormal activities for the risk assessment; and generate one or more recommendations in response to the risky activities.