Drone Mapping

Drone mapping allows you to generate a high-resolution custom LIDAR and custom map image from raw drone imagery and use them directly in your Aurora designs. With drone mapping you can achieve more accurate roof geometry, elevation detail, and site context compared to standard imagery and LiDAR.

In this article, we’ll cover:

• How to capture high quality drone images
• Uploading and processing drone data
• Reviewing processed 3D surface models
• Using drone data in Design Mode
• Using pre-processed (third-party) drone LIDAR and map imagery
• FAQs

Before You Fly - quick checklist

A few things to confirm before each flight to avoid processing failures later:

Your drone has acquired GPS lock before takeoff (most drones show a green indicator or "Ready to Fly" message). Photos taken before GPS lock are missing location data and can't be processed.

You're flying at the same location as your Aurora project pin. If your drone's GPS is more than 1 km from the project location, processing will fail.

You plan to capture 10–100 images with ~60–80% overlap between consecutive shots.

Your drone is set to save images as JPG (not RAW/DNG).

How to capture high-quality images

To create a high resolution surface model, we recommend a flight plan with these details:

1. 40–90 photos (minimum 10 required). More images generally produce better results
2. ~30 meters altitude (100 ft)
3. Image overlap ~60-80%
4. Auto-flight recommended (nadir grid or oblique grid both work). Manual orbit flights are also supported.
5. Camera angle: A mix of straight-down (nadir) and slightly angled (oblique) shots produces the best roof slope detail. Pure top-down imagery works but loses some elevation precision.

Either of these patterns will produce good results:

Uploading and processing drone images

To process drone images, start from the project you want to update.

1. Open a project and find “Drone mapping section”
2. Click “Import files”
3. Upload the raw drone imagery files for the project
   1. Max file size 25MB per image
   2. JPGs only
   3. Don’t exit modal until all files have been uploaded
4. Click “Process files” once uploaded

Once processing begins, Aurora will analyze the imagery and generate a custom LIDAR source with a scaled top-down map image. Processing typically completes in 15–20 minutes; allow up to 40 minutes for larger uploads.

Note: Drone image processing runs asynchronously. You can leave the project while processing completes and return later.

Reviewing drone LIDAR and map imagery

After processing completes, the drone LIDAR and map imagery will be ready to review. Clicking “Review model” to verify your drone data was processed correctly. 

If the results do not meet expectations, you can click “Cancel” and re-upload drone data. 

Note: You can only have one processed drone upload per project.

Using drone LIDAR and map in Design Mode

Once clicking “Accept model”, your drone LIDAR and custom map image will be imported to your project and pre-selected for use in your designs.

Designers can:

• Run AI SmartRoof
• Request EDS 
• Or manually model 

Drone mapping provides the underlying geometry for site modeling, but does not lock or restrict standard Design Mode tools.

Note: Designers in Design Mode will have access to the map image (map dropdown > “Drone Image”) and surface model (LIDAR > Source > Custom) in any design of a project with processed drone imagery. See How to Upload Custom Imagery (Including Drone Imagery) for more information. 

Using pre-processed (third-party) Drone LIDAR and map imagery

Aurora also supports uploading your own pre-processed drone data from third-party software. You can manually upload:

1. A custom drone map “orthophoto” (GeoTIFF) as a custom image. The image should automatically scale based on GPS data in the file. (Max 200 MB file size)

A custom LIDAR ( “Digital surface model” - DSM). Inside LIDAR Inspector click “Upload DSM”. (Max 250 MB file size)

Known limitations

A few cases where drone mapping output may not meet expectations- these are physics limits of photogrammetry, not Aurora-specific:

• Bare deciduous trees (winter): leafless branches don't reflect enough signal for accurate reconstruction. Hand-modeling trees in Design Mode is the standard workaround.
• Heavy snow cover: snow obscures roof edges and ground features.
• Very dark or highly reflective roof surfaces: may produce gaps in the surface model.

FAQs

Q: What is drone mapping?
A: A drone mapping processes raw drone images into a custom LIDAR and scaled top-down map image. It captures elevation and geometry across roofs, terrain, and surrounding structures. This output is used to support accurate design and modeling workflows.

Q: What types of drones are required?
A: You will need a drone that takes images with GPS positioning. This includes DJI Mini 3, DJI Air 2S, DJI Mavic 3 to name a few.

Q: Who can upload and process drone images?
A: Any user with access to Design Mode and project editing permissions can upload and process drone data.

Q: Does drone data replace standard imagery?
A: No. Drone data supplements standard imagery and can be toggled on or off as needed.

Q: Is drone data used automatically in all designs?
A: Drone data is available to use once processed, but designers can choose how and when to rely on it during modeling.

Q: Can I upload my own pre-processed drone imagery using a third-party software into my Aurora design?
A: Yes! If you have already mapped your raw drone in a separate software, you can upload a custom Map Image and LIDAR to your Aurora design. See above.

Q. Why did my image processing fail?
A. Most common reasons image processing might fail are:

1. Not enough images. There is a 10 image minimum. 
2. Not enough overlap 
3. Images are missing GPS data
4. Site area is outside the supported processing bounds (ex: your project project pin is too far from the property address)