AI for mine topography starts with the ground surface and ends with the rock beneath it. This small but foundational category covers the survey work that produces 3D terrain models. It also covers the geological model that terrain and subsurface data are fused into. Blast design, haul-road planning and pit design all build on top of these two layers.
What AI Does in Mine Topography
Survey and modelling work is data-heavy and repetitive, so automation pays off quickly. AI for mine topography does three jobs. It navigates survey drones and LiDAR units through pits and tunnels where GPS fails, using SLAM. It reconstructs 3D surfaces from overlapping images with computer vision. Finally, it interpolates drillhole and survey data into continuous geological surfaces, so models update when new data arrives.
Where to Start
Each page below answers one “how do I do this with AI?” question.
- Capture the surface. A drone survey in mining turns photogrammetry and Hovermap SLAM point clouds into 3D terrain models.
- Model what lies beneath. A mine geological model is built with implicit modelling in Leapfrog, Micromine or Datamine, and trained ML can support parts of that workflow.
From there, the natural next step is implicit geological modelling of lithology, structure and ore zones. Surveys also feed AI-optimized blast design, which needs accurate bench surfaces.
What’s Mature and What’s Emerging
Drone photogrammetry and LiDAR capture are mature, commercially deployed technologies. Implicit modelling is now the industry standard for geological surfaces too. However, it is ML-adjacent interpolation rather than a trained model. By contrast, ML tools that interpret drillhole data, and research methods that learn geological patterns, are still emerging. Keep a geologist or surveyor reviewing every output.