Optical flow stories


Look how well you can navigate using just el cheapo optical flow odometry & sonar.  Over carpet, 0.5 meters is all you get before the camera loses velocity detection.  Over the floor pattern, it can do 2 meters.



Went back to position->velocity->tilt feedback. That was done in 2007, then changed to position->tilt in 2008, when GPS got good enough.  For very precise indoor navigation, position->velocity->tilt feedback seems to be better.   This seems to be what Arducopter also uses.


Originally saw position->tilt in Mikrokopter when it was open source, thinking it was going to be the standard design. It was faster, but caused more toilet bowl effect.  That crazy German even used position->throttle for altitude.


Vision guidance doesn't have doppler shift, which makes velocity delayed. Also, the IMU drifts too much to limit velocity by clamping tilt at a reasonable value. You have to clamp velocity so a long movement doesn't go out of control.


Increased the optical flow frame rate to 50. There is always a hunt for the highest frame rate before the minimum speed becomes too high, yet not too slow to have too much error. Those improvements made it pretty damn stable.



There were some cheap optical flow floor covers at the dollar store. It takes painting 3 with a dry brush to create a reasonable flying space.




No better than unrolled garbage bags, but the most compact material. Can't imagine who would ever use them as table covers.










The 1st board which didn't have any defects. It worked on the 1st power up. It makes sense, because after that, you realized you needed a day job.



marcy2_96.jpg
Didn't have much luck with the  MB1043 but the MB1240 was golden.  That can range useful altitudes over carpet, with the propeller noise.  mm or cm accuracy didn't make any difference.

marcy2_97.jpg

marcy2_98.jpg

marcy2_99.jpg

marcy2_100.jpg


It's hard to conceive a business model for something that flies for only 5 minutes & needs a special floor.  The AR Drone did it by implementing an easy programming language & having enough spare clockcycles for developers to process the video.  Something smaller, with no spare clockcycles would be harder.




marcy2_95.jpg
E-mail me when people leave their comments –

You need to be a member of diydrones to add comments!

Join diydrones

Comments

  • Hi Jack,

    Nice job!
    I disagree with you about one of your last sentences:

    It's hard to conceive a business model for something that flies for only 5 minutes & needs a special floor. 

    It's easy, we can put special floors in too many places: indoor video recordings, theater & art performances, big warehouses or deposits, industry... it's a tool for five monites, so it's the standard time using tools.

This reply was deleted.