SURVEYING A QUARRY WITH A SLAM 100 SCANNER: A PRACTICAL EXPERIENCE

Preparation step

Prior to the start of field works, a set of preliminary measures was carried out to ensure maximum efficiency of the surveying process:

1. Equipment setup:
   • Fully charged SLAM 100 battery packs.
   • Mobile device (smartphone) with an installed SLAM GO software for real-time scanning process monitoring.
2. Route planning:
   • Development of optimal operator movement trajectories
   • Visualisation of complete coverage areas including:
     • Working benches
     • Slope edges
     • Pit footings
   • Minimisation of potential scanning blind spots
3. Creation of a geodetic reference network:
   • Marking and plan-altitude referencing of landmarks to control the accuracy of the final data.

Figure 3. GCP visualization: RGB point cloud with density metrics.

Figure 4. Field reconnaissance.

Scanning step

Figure 5. Backpack-mounted SLAM 100 system with RTK module.

As noted earlier, scanning was performed using an RTK module based on the UNICORE UM-982 chip. This innovative solution enabled complete elimination of ground control points (GCPs), direct point cloud generation in the local coordinate system, and high-precision positioning via corrections from the CORS reference station.

Pre-survey accuracy validation was conducted on two client-installed permanent benchmarks. The 13 km baseline to the CORS reference station maintained RTK corrections without data degradation.

After achieving fixed RTK positioning, the entire 100-hectare quarry was divided into logical survey sectors. An operator then performed sequential scanning using a backpack-mounted SLAM 100 system with RTK module, completing approximately 28 km of traverse lines.

GNSS signal stability was maintained throughout the traverse except for a single 250 meter long. To control the quality of the obtained data in this area, we additionally increased the % of data overlap, so that we could ensure their suitability. Concurrently, real-time data visualization via the SLAM GO app enabled rapid coverage monitoring and immediate gap correction during scanning operations.

Day two encountered severe weather: intermittent drizzle transitioning to intense rainfall. Operational pauses preserved data quality, with all survey objectives met within planned timelines.

Figure 6. Backpack-mounted SLAM 100 with RTK module

Video 1. Scanning stage

Figure 7. Overview map with all traverse paths

Data post-processing step

1. Point cloud generation and processing. The primary data processing included:

• Automated generation of a dense point cloud using the standard SLAM GO Post software;
• Optimization of point cloud spatial distribution;
• Precise georeferencing in the target coordinate system using RTK trajectory data.

Figure 8. Point cloud colorized by elevation.

Figure 9. Point cloud colorized by elevation.

2. Data cleaning and classification. Comprehensive filtering was performed:

• Dynamic objects removed (construction vehicles);
• Vegetation and scan artifacts filtered out;
• Ground points accurately classified.

3. Digital model generation. Processed data was used to generate:

• High-precision Digital Elevation Model (DEM);
• Contour lines:
  • Base: 0.5 m spacing
  • Index: 2.5 m intervals
• Deployment of a picket network with a density corresponding to a scale of 1:500.

Figure 10. Digital elevation model.

Figure 11. Contour lines.

Figure 12. Contour lines and survey pickets

4. Quality control and analysis. A multi-stage accuracy verification was performed:

• Visual data verification;
• Topological surface continuity analysis;
• Cross-validation against control points.

№	X	Y	Z	H	dz
GCP-1	2263861.146	291800.019	134.211	134.2333	0.0223
GCP-2	2263877.373	291487.738	139.176	139.2267	0.0507
GCP-3	2263619.989	291219.029	142.290	142.3233	0.0333
GCP-4	2263384.567	291959.923	132.575	132.6167	0.0417
GCP-5	2263491.007	291884.847	123.809	123.84	0.031
GCP-6	2263606.684	291780.832	123.205	123.22	0.015
GCP-7	2263320.353	291827.357	133.391	133.4133	0.0223
GCP-8	2263369.471	291569.831	120.513	120.55	0.037
GCP-9	2263250.812	291511.787	123.85	123.9	0.05
GCP-10	2262938.912	291588.296	118.246	118.2667	0.0207

Standard deviation: 0.022
RMSE (Root Mean Square Error): 0.034
Mean error: 0.032
Minimum error: 0.015
Maximum error: 0.0507

Table 1. Accuracy assessment

Upon completion, the customer received a final deliverables package containing:

• A high-density point cloud;
• Digital elevation models (DEMs);
• Contour lines and survey pickets.

To evaluate the performance of mobile scanning, the customer carried out a detailed comparison between the SLAM-derived data and the baseline surface previously generated using aerial survey data.

This comparative analysis made it possible to:

• Visualise the dynamics of terrain changes over the inter-survey period
• Quantitatively assess the accuracy of different remote sensing methods
• Evaluate the timeliness of results obtained using various technologies
• Identify the advantages and limitations of each method under specific operational conditions

Key advantages of the presented solution:

• Ability to integrate with previously acquired data
• Compatibility with various GIS platforms
• Data ready for direct use in design and planning
• Full metric and semantic accuracy of the information

This approach demonstrates the effectiveness of combining modern surveying technologies with traditional control methods, ensuring comprehensive monitoring of changes within the mining allotment.