Alert icon
We're changing our privacy policy. This stuff matters.  Learn more  Dismiss

Indoor vision-controlled helicopter

Loading...

Sign in or sign up now!
Alert icon
Upgrade to the latest Flash Player for improved playback performance. Upgrade now or more info.
16,917
Loading...
Alert icon
Sign in or sign up now!
Alert icon

Uploaded by on May 7, 2007

I demonstrate a simple PD controller on my vision-based helicopter control system.

A ring of IR lights shine on three reflectors on the target. The computer tracks them and computes the helicopter's pose and position 120 x's a second. You can see the spots on the black window in the computer screen, and the estimated pose as the white heli on the blue screen (hard to see compressed). The computer then sends radio control signals and flies the heli.

Here, it is trying to hover. The simple controller causes a lot of oscillation because the dynamics are "non-minimum phase." This means that it goes in the opposite direction from its command for a split second, then goes the way you told it. Motorcycles do the same thing.

The simple controller is pretty confused, but it still works. I am working on better control algorithms that should fly it more smoothly.

Category:

Howto & Style

Tags:

License:

Standard YouTube License

  • likes, 1 dislikes

Link to this comment:

Share to:
see all

All Comments (14)

Sign In or Sign Up now to post a comment!
  • What kind of interface do you use to emulate command PC to Helicopter?

  • Amazing, congratulation

  • what model helicopter did you use?

  • its quite easy idea but lot has to go in algorithm ... trying using labview vision system it makes things easy

  • fucking sick!!! this thing is impressive!!! congrats!!!!

  • Will you be publishing your control algorithms?

  • Congratulations! It'd take a lot of courage to put a lovely helicopter under the control of an untested program!

  • how is the interface hardware used? thks. lm2907 with LPT or DAC 8 bit conventional? many thk

  • Very impressive bit of a work

  • That is a good idea. I actually did an analysis to find the exact point of rotation and do exactly that. The problem is that the rotation point changes dynamically, depending on what exactly the heli is doing at the time. I got something that worked better in most cases, but if it ever perturbed an oscillation, the point moves up, and the oscillation becomes worse, and the point moves further up...

Loading...

Alert icon
0 / 00Unsaved Playlist Return to active list
    1. Your queue is empty. Add videos to your queue using this button:
      or sign in to load a different list.
    Loading...Loading...Saving...
    • Clear all videos from this list
    • Learn more