• Blurring the line between real and virtual motorsports
  1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Working with aerial lidar

Discussion in 'Bob's Track Builder' started by Emery, Nov 19, 2014.

  1. So far the best free lidar data I've found comes from the US government agency NOAA. They have a web site at http://www.coast.noaa.gov/dataviewer/# where you can select an area and download the data. The best part about their site is you can choose the units & format for your area of interest.

    I've also found choosing only the "ground" points to be best for trackmaking. When you get the data into BTB, then you only have to classify the points between ground and road. Buildings & structures show up in the "ground" data as holes in the point cloud, so it's fairly easy to place structures accurately. Still, if something you need isn't in the "ground" data, like a building that you don't know the height of, then get all the points in a second download and delete the points you don't need.

    Poking around historic tracks ("ghost" tracks), often a property owner has purposely broken the pavement so hooligans don't go racing without permission. The majority of the surface might be in good shape except for rips down the middle or across the track every so often. Once you get a BTB track mesh lined up & track height calculated, then you can carefully navigate the track in 3D and delete the points that are obviously too high or too low. Recalculate the track height and presto, a restored racing surface!

    Other times, the earth has been moved to the extent that the point cloud is useless for the track surface (example: Bridgehampton Raceway turned into a golf course). So far I lay a narrow point-to-point track calculated to the point cloud as a guide, then loft another track into place using the narrow track and start laying down terrain. Test before considering this a final version.
    Last edited: Nov 19, 2014
    • Like Like x 1
  2. Attached is an example of classifying a point cloud from an aerial lidar. The right hand is a historical image I'm using as a guide. Green points are BTB "ground" and dark grey points are BTB "road" and I'm in the process of selecting more points for "ground". When you import a point cloud into BTB, the points are all classified as "road", so you recategorise them until it looks right. Meadowdale Classifying BTB point cloud.JPG
    • Beer Beer x 1
  3. Tracks with several width changes are probably best dealt with by using a single width adequate for covering the largest width. Use extra panels (ideally meeting modern standards of 1.5 meters) in the surface and change the material to grass when there isn't any paving.

    An example is Portland International Raceway. The main straight is 60' wide (plus pits!) then it narrows to 30' for most of the track, but opens up to 40' width more than once.

    I think you can leave the lidar categories (ground, road) as they should be and BTB will still calculate the correct road height.

    If you're using 1.5 meter panels based on lidar and exporting to ISI-based sims, make sure you adjust the following TDF values to avoid having an overly bumpy track:
    RoadBumpAmp = 0.001
    RoadBumpLen = 3.0
    MiscBumpAmp = 0.07​
    • Like Like x 1
  4. try this i couldn't find a road on the dem file or the image, althought it was marked on website.also seem to get a lot of crashes on dem tool,(out of memory, a a violation @ blah blah address). may i should move in a bit more.haven't a clue.
  5. Sorry, I haven't tried the DEM tool. Only been dealing with the LIDAR point clouds as CSV files.
  6. Good, I have a try, waiting that my selection arrive by email