Tuesday, October 26, 2010

Driving Simulators.

Give a bunch of engineers an unlimited budget and say "Go Nuts."  They don't go to Vegas, they don't buy big houses, they build this monstrosity.


Maybe I could build a big X/Y table and slap on a Stewart Platform.


But don't forget about the Formula 1 Driving Sim.

5 comments:

  1. What are the attractive videos...
    You did a great work by posting such gorgeous videos. Thanks...

    ReplyDelete
  2. Impressive!

    Can you elaborate a bit more, how did you achieve the LAs control, given the associated mechanical constraints, imposed by the Stewart Platform?

    Did you use any Math simulation software to determine the number of rotations for each motor, in order to achieve a certain platform positioning?

    On the other way and since you are not using NXT servos (PF motors are not servos), so that you cant control exactly the number of intended turns, how did you control the system?

    ReplyDelete
  3. Thanks Conchas! Sure, I'll tell you what I know

    I have built a few of these with limited success. The hard part really was getting the linkages around the platform to work right. I recently built a newer version that works even better...
    http://1.bp.blogspot.com/_CVq5bbnuCpc/TIVYzSZAM3I/AAAAAAAAAUo/Rrzd9mm5bPg/s1600/Stewart.JPG

    I chose the PF motors just because they are so easy to mount and using the rectangular beams just worked perfectly.

    I must confess, the motion is completely faked. There is no rhyme or reason. I found some positions using a regular remote and then mimicked them with the programming and the HT IRLink.

    To get the positioning, I timed how long it takes to completely extend one of the LAs. Say it was 6 seconds. Then if I wanted to extend half way, I just ran it for 3 seconds. No magic to it at all. I considered putting an old legacy rotation sensor on each motor which wouldn't be too hard, but then I would have to deal with 6 sensors on 4 ports. I didn't have a true use for the platform, so I didn't put that much effort into it, but it certainly could be done.

    This is one of those pretty easy builds that can be done in about an hour, so I keep it in the back of my mind when I am thinking about other projects and how I could use it. One idea that I had was to put a head on it and make the head turn and move around.

    ReplyDelete
  4. This comment has been removed by the author.

    ReplyDelete
  5. Ok, thanks for the explanations!

    Yes I agree the PF motors are much easier to integrate into models. And the attachment in this specific MOC, is pretty simple and effective.

    Somehow, I suspected there was some "cheating" with the fluidity of movements seen in the video... :)

    Great work anyway!
    And simultaneously some coding challenge for the future. ;)

    ReplyDelete