Drones & Laser Scanning

Modern mine surveying uses drones (UAVs), photogrammetry, LiDAR and mobile laser scanning. Learn ground sample distance (GSD), image footprint and overlap, flight planning, ground control points (GCPs) and RTK/PPK, point clouds and accuracy checks, and applications in pits, stockpiles and underground.

A drone can survey an entire pit in an hour — work that once took a survey crew days, often in hazardous areas near highwalls. Laser scanners capture millions of points per second, producing detailed 3D models for geotechnical mapping, reconciliation and digital twins.

An analogy

Photogrammetry is like your two eyes: seeing a scene from slightly different positions lets the brain (software) reconstruct depth in 3D.

Photogrammetry

Overlapping images (typically 75 % front, 65–70 % side) are processed with structure-from-motion software (Pix4D, DJI Terra, Agisoft) into point clouds, orthomosaics and DTMs. GSD sets the detail — ≈ 2–3 cm is typical for mining.

Accuracy

Ground control points (surveyed targets) or RTK/PPK GNSS on the drone georeference the model. Check points not used in processing measure accuracy (RMSE ≈ 1–3 × GSD). Vegetation and dust reduce quality.

LiDAR and scanning

Laser scanning measures distance directly: terrestrial scanners for highwalls, drone LiDAR through vegetation, mobile/SLAM scanners for underground drives and stopes, and fixed scanners monitoring stockpiles and slopes continuously.

Advanced uses

Automated geotechnical mapping (joint orientations from point clouds), change detection for slope monitoring, digital twins updated daily, and autonomous drones flying underground (BVLOS in stopes) for CMS-style inspections.

Formulas

Worked example: GSD

Given: Sensor 13.2 mm, f = 8.8 mm, 5472 px; H = 100 m

Find: GSD and footprint.

Answer: 2.7 cm GSD, 150 × 100 m footprint

Worked example: Photo spacing

Given: Footprint 150 × 100 m; Front overlap 75 %, side 70 %

Find: Spacing.

Answer: Photo every 25 m, flight lines 45 m apart

Common mistakes

Applications

Key takeaways

Going further