Waypoint navigation testing

[peabody124]

So there was a stiller day yesterday so I had the chance to test waypoint navigation a bit better.

https://vimeo.com/64079883

The flight plan is to fly a 10m box.  Full disclosure: I had throttle control in this in case anything went wrong.  From the video it seems like the quad was overtuned and thus kept overshooting at the 90 degree corners.  This is actually partly just a bad flight plan for a quad - a 90 degree turn at constant velocity is physically impossible so I should redo it with rounded corners.  However, the overo logging makes it really nice to see what happened, and clearly it is underdamped as it's oscillating pretty heavily around the desired path.

Still, reproducible and fixable limitations so I consider that a pretty successful test.

[metRo_]

This is awesome :) I'm waiting to get the F3 shield to start testing and learn about it:)

[Stefan Cenkov]

Today I tried to fly to WP 10m Box and looks good, I have noticed a difference of 1 m from WP2 to WP3, but I need to try some more.
 I'm use "james/development' branch.

@ peabody124 is normal WP0 that always gets in 0,0,0?

[metRo_]

That looks great:)

Can you share your quad setup?

[peabody124]

Wow, awesome.  Was that simulation or real flying?  It the later you got me beat!

WP0 should be (0,0,-10) not (0,0,) 

[Roberto Navoni]

Great Work Stefan :)

@ peabody124 

what need for accept Stefan pull request  ?

There are some VRBRAIN user that use windows that wait official revision of Ground Station that would join to doing beta testing . 

Check for example this beta tester :http://www.youtube.com/watch?feature=player_embedded&v=DTfT1KrxZHE

Now working only on Arducopter32 but could be work also on Taulab

Best

Roberto

2013/4/18 peabody124 <peabo...@gmail.com>

Wow, awesome.  Was that simulation or real flying?  It the later you got me beat!

WP0 should be (0,0,-10) not (0,0,) 

--
You received this message because you are subscribed to the Google Groups "phoenixpilot" group.
To unsubscribe from this group and stop receiving emails from it, send an email to phoenixpilot...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[peabody124]

I posted some comments in the PR previously https://github.com/TauLabs/TauLabs/pull/416 about refactoring out a few of the components.  I haven't seen follow up on that.

[Stefan Cenkov]

@ peabody124 this is real flight, the Waypoint instance(0) is 0,0,-10 , I'm wondering about WP editor. Read WP's I get 0,0,0 for WP1,
also if you change altitude and saving, reading always gives altitude 0.

[peabody124]

Wow, then congrats on a really clean flight!  I'll have to look into your settings.  I need to look at that waypoint editor issue - looks like a bug.

I also found yesterday over RF telemetry that uploading waypoints wasn't reliable - which was odd because I'd made it reliable in the past.

[peabody124]

Little bit more testing https://vimeo.com/84184234

[Reddog]

Is this for Multirotor or Fixed Wing?

BTW: Fantastic video!

[Reddog]

What units are the altitude and velocity figures you changed, measured in?

On Wednesday, January 15, 2014 12:57:08 PM UTC+10, peabody124 wrote:

[peabody124]

This was testing with multirotor. I think the fixed wing code probably still works and it did last time I simulated it. However, it has been a while since anyone has flown it practically so ...

The velocity is m/s between the waypoints. I actually thought there was a bug because my first test I had that at 0 m/s and it seemed to be in position hold :)

The altitude is a bit odd and I want to modify it. There is altitude (higher is positive) and relative down (further column to the right). I need to dig into the code because they weren't identical and were off by .2 meters. It's been forever since I worked on the waypoint editor so I don't recall.

[peabody124]

More testing with a more complicated path: http://buildandcrash.blogspot.com/2014/01/navigation-testing-and-post-crash.html and a pretty amusing crash with an important lesson.

Do NOT fly waypoints with the throttle in the low position.

[José Eduardo S. C. Xavier]

the weather here is not so good but I hope to be able to fly the quad in a few days and really want to test again PH and AH to try to get good results. Is f3 able to do waypoint navigation?

2014/1/20 peabody124 <peabo...@gmail.com>

[peabody124]

Well this is some testing on Sparky (F3) I did https://vimeo.com/84639710 but I dont' think it will do waypoints. There is a lot of memory that isn't efficiently allocated amongst the tasks so it might still be possible...

[VArcht]

Really hope you can get waypoints running on Sparky, just got mine flying this weekend and really looking forward to autonomous flight :)

[peabody124]

Your timing for that question was perfect. I was just tweaking things while working on https://github.com/TauLabs/TauLabs/issues/1071 . Since I already had settings adjusted I checked and it is possible to run GPS, Path Follower and Path Planner on Sparky. Now you probably won't be able to run lots of extra goodies like mavlink etc.

However, my first priority is to get https://github.com/TauLabs/TauLabs/pull/1054 merged and get people doing simple PH and RTH reproducibly and reliably.

[VArcht]

Sounds good, better I get acquainted with ph and rth first too. I see next is building for Win32 again so I'll update and see how it goes.