Discussions

Auto land- How robust is it?

Posted by Cold Wombat on May 28, 2014 at 8:44am

I've read various reports over time that 'auto land' is (or was not so long ago) pretty fragile.  Is that still the case?  Can it yet make use of a proximity sensor to aid landing?  If not, is there much work involved in improving this?

I'm probably more interested in belly landing a wing rather than something with an undercarriage, but I'm not sure how much of a difference that makes.

Thanks,

CW

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

Join diydrones

Tags: auto land, landing
Email me when people reply –
Follow

Replies

  • iskess May 30, 2014 at 4:57pm
    Actually I find that it is safer to switch to Trainer mode as your panic button. Sometimes we have a tendency to over control the aircraft when abruptly transitioning to manual control.
  • 100KM
    Hein du Plessis May 29, 2014 at 11:19pm

    Thanks for the kind words, accepted. I consider myself a good "C" :)

  • 100KM
    Hein du Plessis May 29, 2014 at 10:25pm

    Very encouraging. I need to autoland a Mini Mugin. Hope it goes well...

  • Randy Ram May 29, 2014 at 8:00am

    I have tried it on multiple quads and it has been working great. Nice and smooth. Very happy with it.

    • 100KM
      Hein du Plessis > Randy Ram May 29, 2014 at 8:08am
      I think CW is referring to plane. Copter works great, agreed.
  • 3D Robotics
    Chris Anderson May 29, 2014 at 7:37am

    3.x added many improvements to autoland, and we now use it every day. The next big things coming are automatic landing pattern creation using rally points (so you don't have to set up the approach yourself in the mission planning) and Lidar-Lite support for more accurate proximity detection. 

    • 100KM
      Hein du Plessis > Chris Anderson May 29, 2014 at 7:39am

      Hi Chris I'm very excited about the 3.x improvements. What is the best practice to handle wind? If, say, we repeat a set mission every day, but the wind changes every day, should we save two missions with the same way points, each with a different approach?

  • Cold Wombat May 29, 2014 at 6:40am

    Thanks Hein & iskess.  That's given me enough confidence to pursue a few ideas I've got.

  • iskess May 28, 2014 at 11:15am
    It works very well if set up properly. You want a long gradual constant rate descent for best results. It doesn't handle crosswinds very well once the flare starts. It tends loose it's heading lock. All in all I'd give it a B-.
    Certainly room for improvement, but I am confident using it.
  • 100KM
    Hein du Plessis May 28, 2014 at 9:55am

    As far as I know it's pretty sorted, version 3.x has even more improvements. I'm busy with a project which sort of hinges on it... So here's hoping :)

This reply was deleted.

Activity