Search

From Sensing to Steering: Multi-Source Data Fusion and Predictive Analytics Towards Intelligent and Autonomous TBM Tunneling

The future of tunnel boring machine (TBM) technology should not be limited to performance and speed, but evolve into a predictive, adaptive, and intelligently orchestrated autonomous system. While engineering advancements and automation have significantly improved TBM capabilities, the true key to achieving safe, efficient tunneling lies in a deep understanding of its core driver: data.

This special session aims to bridge this critical gap. We seek to bring together scholars and practitioners from civil engineering, geotechnical engineering, robotics, artificial intelligence, and advanced manufacturing. The goal is to explore cutting-edge methods for real-time data fusion from geology, machinery, and sensors, and develop AI approaches for predictive analytics and autonomous decision-making. We hope to collectively shift the paradigm of TBM tunneling from being experience-driven to being driven by predictive intelligence and autonomous control.

We invite abstract submissions on topics including, but not limited to:

(1)     Advanced Sensing & Data Collection:

  • Multi-modal data fusion (e.g., computer vision for trajectory analysis, sound recognition for rock mass identification). 
  • Capturing mechanics and health status signals (e.g., Using distributed optical fiber for structural health status; innovative cutterhead or mechanical health status sensors).
  • Emerging paradigms like VR/AR-based shield driving and IoT-enabled disturbance sensing.

(2)     Data Interpretation & Insight Generation:

  • Uncovering geology, excavation patterns and rock–machine interaction through data mining.
  • TBM excavation models and training methods with scenario or task generalization capabilities
  • TBM driving status and risk assessment for scenarios in special geological conditions.

(3)     Autonomous Tunneling & Control:

  • Multi-objective model-predictive control trading off energy use, wear cost and schedule deviation in real time.
  • Promoting interpretability, robustness, and adaptability in intelligent control of TBM excavation.
  • Enhancing operational efficiency and safety in TBM excavation through decision-making frameworks.