Autonomous Tunnel Boring Machines: Artificial Intelligence Progress of 2025

Tunnel Boring Machine

Tunnel Boring Machines have always represented the cutting edge of engineering, capable of carving through kilometers of soil and rock with remarkable precision. For decades, however, their operation has depended heavily on skilled human crews making continuous adjustments: steering the cutterhead, regulating thrust and torque, monitoring pressures, and interpreting ground conditions. Even with advanced instrumentation, most decisions were reactive, based on what could be observed and measured in the moment.

That paradigm is now beginning to change. In 2025, artificial intelligence and real-time data integration are reshaping tunneling from a process guided mainly by human expertise into one increasingly driven by intelligent systems. The TBM is no longer just a powerful mechanical tool—it is evolving into a learning machine

Technologies of 2025 Tunneling

Recent projects illustrate how far tunneling technology has advanced:

• Variable Density TBMs capable of switching between Earth Pressure Balance and slurry mode to handle mixed ground conditions.

• Semi-continuous tunneling systems that install segments without fully stopping excavation, boosting advance rates.

• Smart operator-assist systems providing adaptive steering, automatic lubrication, and real-time navigation support.

• Automated segment handling with robotic unloading, inspection, and ring building for safer, faster assembly.

• Robotic fit-out systems installing tunnel services autonomously, reducing manual work in confined spaces.

• IoT-based geotechnical monitoring using wireless sensors and fiber optics to deliver continuous feedback on ground and lining behavior.

• Adaptive feedback controls that allow TBMs to automatically adjust excavation parameters based on sensor input.

• Operational digital twins—live 3D models that integrate TBM and ground data for predictive simulation and asset management.

• AI-enhanced analytics to forecast advance rates, cutter wear, and ground anomalies.

• Integrated data environments centralizing TBM, monitoring, and BIM information for faster, data-driven decisions.

Artificial Intelligence in 2025: Changing the World

Artificial intelligence has moved far beyond research labs and prototypes—it has become a mainstream driver of transformation across society. In 2025, AI powers autonomous vehicles, assists doctors in surgery, predicts market shifts in finance, and personalizes education for millions of students. It is embedded in smartphones, logistics networks, and even the energy grids that balance electricity demand in real time.

The progress is visible everywhere. AI models now translate languages with near-human accuracy, detect diseases earlier than traditional screenings, and enable factories to run with minimal human supervision. In law and policy, AI systems are being used to analyze vast legal databases, accelerating decisions that once took weeks. In climate science, they help model extreme weather patterns and optimize renewable energy production.

What makes today’s AI remarkable is its shift from simply automating repetitive tasks to making sense of complexity—finding hidden patterns, predicting future outcomes, and recommending optimal actions. Its ability to learn and improve continuously has turned it into a collaborator, not just a tool. This global leap in capability is the same force now entering industries like tunneling, where massive amounts of data demand intelligent interpretation and adaptive decision-making.

DAARWIN: One Step Forward to Autonomous TBMs

The leap toward autonomous tunneling would not have been possible without pioneers willing to rethink how data could be used in geotechnics. This is where SAALG Geomechanics took the initiative and reshaped the industry with the creation of DAARWIN.

Traditionally, TBM and monitoring data were scattered across reports, spreadsheets, and disconnected systems. Valuable insights were often delayed or even lost once a project ended. SAALG broke that cycle by introducing a platform that does more than store information—it uses artificial intelligence to transform raw excavation data into actionable intelligence in real time.

What DAARWIN Can Do for TBM Projects

• Analyze all excavation and monitoring data in real time.

  
  

• Identify the most influential parameters affecting TBM performance.

  
  

• Predict TBM advance rates with higher accuracy for better planning.

  
  

• Detect geotechnical anomalies in advance, turning monitoring into prevention.

  
  

• Assist TBM pilots with AI-driven recommendations for optimizing excavation phases.

  
  

• Centralize and structure all TBM and ground information in one platform.

  
  

• Create a digital memory so knowledge from past projects strengthens future ones.

With DAARWIN, thousands of data points from the TBM and the ground are ingested, cleaned, and analyzed as the tunnel advances. AI models then highlight the parameters that truly govern performance, forecast advance rates before they happen, and flag anomalies early, giving engineers time to respond proactively. For TBM pilots, DAARWIN acts as an intelligent co-pilot, suggesting how to optimize excavation phases rather than relying solely on manual judgment.

But the real industry-changing step is DAARWIN’s ability to centralize and preserve knowledge across projects. Every tunnel excavated with DAARWIN becomes part of a growing base of digital intelligence that future teams can build upon. What was once fragmented experience is now structured learning, shared across the industry.

The Road Ahead

Autonomous tunneling is no longer a distant dream—it is unfolding today. The combination of advanced TBMs, real-time monitoring, and artificial intelligence has already begun to redefine how we excavate underground.

DAARWIN stands at the center of this transformation. By turning raw data into foresight, by learning from every tunnel, and by supporting both operators and engineers in real time, it has changed the way the industry thinks about progress.

The TBM of the future will not only bore through the ground—it will understand it, adapt to it, and remember it. With platforms like DAARWIN, each project becomes more than just an engineering achievement; it becomes a step toward a truly autonomous future.