i-WASP – ein Forschungsupdate zu intelligenten und drahtlosen Technologien für automatisch erstellte Schichtprotokolle im konventionellen Tunnelbau
Enterprise AI Analysis for i-WASP – ein Forschungsupdate zu intelligenten und drahtlosen Technologien für automatisch erstellte Schichtprotokolle im konventionellen Tunnelbau
Das Forschungsprojekt i-WASP, ein Zusammenschluss spezialisierter deutscher und österrei- chischer Unternehmen und Forschungseinrichtungen, be- schäftigt sich mit der Frage, wie die komplexen Abläufe auf einer NATM-Tunnelbaustelle mithilfe innovativer Metho- den der Datenakquisition, -übertragung und -verarbeitung automatisiert erfasst, analysiert und optimiert werden kön- nen. Eine trainierte künstliche Intelligenz wird in der Lage sein, auf Basis der direkt an den Baumaschinen erfassten Beschleunigungs- und Positionsdaten automatisch Zyklus- diagramme zu erstellen.
Executive Impact & Key Findings
The i-WASP project revolutionizes tunneling operations through automated data capture and AI-driven analysis, yielding significant improvements.
0
Recognition Accuracy
0
Personnel Hours Saved
0
Meter Range
Deep Analysis & Enterprise Applications
Select a topic to dive deeper, then explore the specific findings from the research, rebuilt as interactive, enterprise-focused modules.
Technologies
The project integrates innovative sensor technology, wireless data transmission, and AI-powered analysis for automated data collection. Key components include vibration-based data acquisition, underground positioning systems, and robust wireless communication solutions. These technologies work in concert to capture, transmit, and process complex tunneling data automatically.
Data Acquisition
Vibration-based data acquisition units capture machine activities, paired with underground positioning systems for precise localization. Accelerometer data provides insights into machine operations, while ranging-based positioning (showing high stability and accuracy in tunnels) ensures precise location tracking. This combined data forms the basis for automated process classification.
AI & Processing
A trained artificial intelligence processes accelerometer and positional data to automatically generate minute diagrams, enhancing data quality and relieving personnel. The two-stage classification uses Fast Fourier Transformation (FFT) for feature extraction and a Convolutional Neural Network (CNN) for process classification. Initial field trials show over 94% recognition accuracy.
94.6%
Process Recognition Accuracy Achieved in Field Trials
Enterprise Process Flow
Vibration-based Data Acquisition
→
Underground Positioning
→
Wireless Data Transmission
→
AI-Powered Process Classification
→
Automated Shift Protocol Generation
| Feature | Manual Logging (Current) | Automated (i-WASP) |
|---|---|---|
| Data Resolution |
|
|
| Accuracy & Completeness |
|
|
| Personnel Effort |
|
|
| Data Analysis Potential |
|
|
Field Trials & System Validation in Angath (Tirol)
On a tunnel construction site in Tyrol, up to six different types of machines were instrumented simultaneously. Parallel to the automated data acquisition, manual logging was performed to generate training and validation datasets for the AI models. The results demonstrate the mechanical robustness and data quality of the systems under real construction conditions, achieving a recognition accuracy of over 94%.
Addressing Challenges in Sontra (Holstein)
Further field tests were conducted in the Sontra tunnel to expand the data basis with different geological and tunneling conditions. Insights were gained on sensor housing resilience, connector vulnerabilities, and optimal sensor positioning. The trials confirmed the system's suitability for harsh environments, highlighting specific challenges in real-time data processing and ergonomic UI design. The developed sensor technology proved highly capable of capturing vibration data effectively.
Calculate Your Potential ROI
Estimate the time and cost savings your enterprise could achieve with AI-driven process automation.
Estimated Annual Savings
$0
Annual Hours Reclaimed
0
Your AI Implementation Roadmap
A phased approach ensures smooth integration and maximum benefit realization for your enterprise.
Phase 1: System Integration (Ongoing)
Full integration of all hardware and software components, including vibration sensors, underground positioning systems, and robust wireless communication infrastructure. This phase focuses on seamless data flow from acquisition to the central server application.
Phase 2: UI/UX & Data Visualization
Development of intuitive user interfaces for displaying and analyzing automatically generated shift diagrams. Emphasis on clear visualization of cycle times, process durations, and machine utilization for enhanced operational overview.
Phase 3: Real-World Demonstration & Validation
Conducting extensive real-world demonstration phases on active tunneling sites to validate overall system performance under various conditions. This includes fine-tuning AI models and collecting user feedback for final refinements.
Ready to Transform Your Operations?
Book a free consultation with our AI specialists to discuss your specific needs and how i-WASP can drive efficiency and accuracy in your projects.
Ready to Get Started?
Book Your Free Consultation.
Let's Discuss Your AI Strategy!
Lets Discuss Your Needs
Select a Date
Sun
Mon
Tue
Wed
Thu
Fri
Sat