• Dear visitors,

    The email issue has been finally solved.
    Thank you for your patience and happy browsing.

    Team ACM.

QUESTION Early progress and GPS-height question

Project.Dream

New Member
I finished my first rudimentary track which is based on a "Carrera 140 Profi", a german slot-car toy. Unfortunately I realized that the track is too small for faster cars so I will probably end this experiment. It was only made for learning to use the AC editor anyway so it is not a big setback for me.

I made a short video of me driving on the track, but the image quality is really bad because I had to use an external caputuring device.




Next is a question about building real world tracks: The first thing I did to be able to model a road was to get a satellite image and a relief map. To define the height of the road on each point I thought about using the conturs of constant height from google maps or to use a greyscale relief map (I tried a generator from the city skylines community). Both of them seemed to be imprecise or hard to utilize.
My idea then was to use the GPS device of a smartphone and to record the output while driving (see photos). I think the results looks promising - for the graphs I extracted one point every 2 seconds.
For this experiment I was driving about 2.5 km and had a height difference of about 147m. The upper graphs show the coordinates, the lower ones the error margins and the number of satellites that were available at the moment. I had hoped to also be able to extract the angle of the road surface (to precisely model steep corners), but the measurements were influenced by the g-forces while driving through a curve.
Do you think this could be a good method for getting more precise information about the track or is this nonsense?


Rats76.jpg GPS_0.jpg GPS_1.jpg
 

Gunnar333

Well-Known Member
I also did try to capture height information with smaprtphone GPS for my Borgloh hill climb track. But the results weren't good enough for modelling.
Capturing the height information with Google Sketchup was better.
For the angle of the road surface both methods are nonsense.

Best thing you can do is finding LIDAR Data. Search for Open LIDAR for the region you want to build the track and read the forum post how to convert it to your 3D software.
 

Johnr777

Moderator
I also did try to capture height information with smaprtphone GPS for my Borgloh hill climb track. But the results weren't good enough for modelling.
Capturing the height information with Google Sketchup was better.
For the angle of the road surface both methods are nonsense.

Best thing you can do is finding LIDAR Data. Search for Open LIDAR for the region you want to build the track and read the forum post how to convert it to your 3D software.
Or capture from google maps 3D view if available for that area, also known to be just as precise but very time consuming unfortunately. Examples and info over at the Kyalami thread.
 

Project.Dream

New Member
Best thing you can do is finding LIDAR Data. Search for Open LIDAR for the region you want to build the track and read the forum post how to convert it to your 3D software.
I did find LIDAR data and followed the tutorial on CloudCompare. The results are amazing, I attached a render of the generated low detail landscape with the highly detailed road (red). Took me days and large amounts of RAM to process the data but I am very happy with the model.
I probably can adjust the settings a little bit better to seperate the trees from the ground and maybe I will create another model just for the buildings...

VRa01a.jpg
 

Project.Dream

New Member
My next steps were to capture the real "track" on camera while driving to be able to see which hills, hollows and buildings would be visible and had to be modelled, as well as to see more details of the road and surroundings.
Then I modelled an extremely simplified version of the road (completely flat) to experience if the track would be interesting to drive on, or if the road would be too narrow to be fun (as my first track turned out to be completely impassable because of the tight curves). When looking at the Lidar-model I realized that it was very hard to set the road apart from the rest, so I first had to create a huge (~23,000 X 23,000px) satellite-image using google maps.

The first test-drive went well, but the road is indeed quite narrow and dangerous on high speeds. The bigger problem is that I noticed that the track is so large that the many buildings would probably take months or even years to model. Maybe I will try t0 build a road that is matched to the relief, as well as the relief itself, but I am not sure if I will be able to finish the project or give up soon.

VRsc.jpg
 
Top