Emerging Geospatial Technologies and Applications
Emerging Geospatial Technologies and Applications
This domain tests newer acquisition and delivery technologies: UAV mapping, mobile and SLAM mapping, precise GNSS, cloud GIS, machine learning, and digital twins.
Core rules and concepts
- UAV photogrammetry and LiDAR require ground control and quality checks; high overlap improves matching but does not replace control.
- PPP and PPP-RTK provide precise positioning without a local base station, but convergence time, correction service, and atmospheric conditions affect reliability.
- SLAM supports mapping in GNSS-denied environments such as tunnels and indoor spaces; drift must be controlled by loop closure or control points.
- Machine-learning classification requires representative training data, validation, and attention to class imbalance and spatial autocorrelation.
- Digital twins integrate BIM, GIS, and sensor data; they are only as reliable as their update and calibration processes.
Worked example
A UAV uses a camera with 5 µm pixels and a 50 mm focal length at 100 m above ground. GSD = pixel size × flying height / focal length = (0.005 mm / 50 mm) × 100 m = 0.0001 × 100 m = 0.01 m, or 1 cm. A 1 cm GSD does not by itself guarantee 1 cm positional accuracy; ground control, camera calibration, and terrain also affect results.
Traps
- Equating GSD with map accuracy.
- Using UAV or SLAM data without ground control or loop closure.
- Treating machine-learning output as authoritative without validation.
- Assuming PPP-RTK is instantaneous and immune to multipath or canopy.