Condor 2 Landscape autogeneration

Note": This isn't in alpha yet, more testing and docs to come shortly.

This is a set of tools, written in python, to generate as much of a Condor 2 landscape as possible for you. The ultimate goal is to take a few bits of information such as a lat/long and airport name, and product a flyable landscape in a few minutes.

If you specify: - The name of your landscape - The lat, long of the center of your landscape; - The number of tiles in x and y; - A source for the ortho imagery to use;

then the tools will automate some or all of the following steps for you. The ones marked with ✅ are done, those with ❌ require manual intervention (though may be automated in future!).

• ✅ Calculate UTM zone and tile corners
• ✅ Download height data
• ✅ Make height maps raster files
• ✅ Download aerial imagery and make textures
• ✅ Fetch tree and water data from OSM
• ✅ Make forest maps (one tree type only)
• ✅ Make textures with water alpha
• ✅ Make rudimentary thermal maps
• ✅ Export thermal map
• ❌ Color matching of imagery
• ❌ Generate forest and terrain hashes
• ✅ Make runways (depends on runway data availability)
• ❌ Add runway markings, add windsock
• ❌ Add buildings and other landscape objects

Texture imagery

The tool can fetch orthographic imagery from a variety of sources. Your choices are:

• A WMTS tile server (a web service that gives an image of a square of earth at a particular zoom level). Examples include USGS Earth Explorer (via arcgis), Google Maps, Bing.
• A WMS server (a service that provides images in a variety of types and projections)

If you use a public imaging service, you should be sure you have appropriate permissions, especially if you plan to redistribute the landscape.

In the next version I plan to include functionality to automatically generate synthetic textures, bypassing the need to fetch imagery (but foregoing the realism of ortho imaging).

Usage

Create a yaml configuration file following the template file.

Then either:

• Install all the python dependencies,
• and/or create a python virtual environment (recommended)
• or install docker and run with that (see below).

and run:

condor-autogen config.yaml generate

Non-automatable steps

• I don't know the hash algorithm, so can't make forest/terrain hashes. This may be secret on purpose.

Wish list

• AI tree placement
• AI buildings