Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • Feb 02 2021 21:00

    Georacer on master

    Added last_letter_lib as a subm… Deleted deprecated service code Deleted deprecated launcher fil… and 23 more (compare)

  • Feb 02 2021 21:00

    Georacer on ftc_journal

    Added last_letter_lib as a subm… Deleted deprecated service code Deleted deprecated launcher fil… and 23 more (compare)

  • Sep 08 2019 18:41

    Georacer on master

    Added support for randomized UA… A bit of polish in airdata calc… Added pause functionality and 4 more (compare)

  • Jun 05 2019 12:19

    Georacer on master

    Fixed custom message dependenci… (compare)

  • Dec 10 2018 14:16
    Georacer commented #63
  • Dec 10 2018 14:15

    Georacer on master

    gitflow-feature-stash: Bug_fix_… Merge pull request #63 from Mil… (compare)

  • Dec 10 2018 14:15
    Georacer closed #63
  • Dec 10 2018 14:11
    MiliStefanos opened #63
  • Dec 10 2018 13:53

    Georacer on time_lock

    First jab at Gazebo world step … (compare)

  • Dec 10 2018 13:52

    Georacer on migrate_to_gazebo

    (compare)

  • Dec 10 2018 13:52

    Georacer on migrate_to_gazebo

    Created test Gazebo Plugin ROS control of single thruster … Added Plugin for aerodynamics and 11 more (compare)

  • Dec 10 2018 13:47

    Georacer on proper_wind

    speedDot now uses relative spee… (compare)

  • Dec 10 2018 13:38

    Georacer on time_lock

    First jab at Gazebo world step … (compare)

  • Dec 10 2018 13:26

    Georacer on master

    Made Dashboard plugin compatibl… (compare)

  • Dec 10 2018 13:21
    MiliStefanos closed #61
  • Dec 10 2018 12:56

    Georacer on master

    (compare)

  • Dec 07 2018 16:25
    MiliStefanos opened #62
  • Dec 07 2018 16:04
    MiliStefanos opened #61
  • Dec 07 2018 16:04
    Georacer closed #60
  • Dec 07 2018 16:01
    MiliStefanos opened #60
Andrew Tridgell
@tridge
changing the order and direction in propulsion.yaml lets it takeoff, but it isn't stable
note you need "param set FRAME 1" to make it an X frame (it defaults to a + frame)
Andrew Tridgell
@tridge
with the motor order and rotation fixed it does takeoff better, but is badly tuned, and crashes quite soon
this is what I ended up with to match ardupilot order for a Quad-X: http://uav.tridgell.net/tmp/propulsion.yaml
George Zogopoulos-Papaliakos
@Georacer
thanks for giving it a shot. I'll look into eliminating the ground spin next.
So, what's the current launch command for an X-frame?
Does
sim_vehicle.sh -j8 -f last_letter:uav_name:=quadrotor -b sitl --console
launch an +-frame?
Since param set FRAME 1 is a MAVProxy command, I suspect that one would need a command for the sitl launch/build procedure.
George Zogopoulos-Papaliakos
@Georacer
Also, did you add the SITL time patch from Michael Day?
George Zogopoulos-Papaliakos
@Georacer
I merged mday's patch. Framerate is up, but there still are a lot of hick-ups/lagspikes. It's defnitely an improvement though.
Is there any update on this?
Futhermore, I also get the same behaviour as Tridge. The quad spins continuously out of control.
I may have unrealistically large moment on the motors; will investigate further.
Andrew Tridgell
@tridge
good morning
you need to start SITL once, then run "param set FRAME 1". Then restart
the frame type for quadcopters is a parameter, not a command line option
george: the real puzzle with the spinning on startup is that the motors are providing any torque at all. The motors should be off when it boots. They should only start to spin once the vehicle is armed and the throttle is raised
George Zogopoulos-Papaliakos
@Georacer
Good morning, Tridge.
The spinning might very well be due to the ground dynamics. I'll let you know when I have more info.
George Zogopoulos-Papaliakos
@Georacer
I believe it's a combination of both, in the end. The ground friction model doesn't handle lateral forces very well, whereas in multicopters, there's usually a moment which rotates the craft parallel to the ground.
George Zogopoulos-Papaliakos
@Georacer
I'll be on vacation next week, so I doubt I'll be able to put any work in last_letter.
Fredrik Hedberg
@fhedberg
@Georacer Hey! I'm working on APM support for a dual-swash helicopter tiltrotor (much like the V22 Osprey). I've flown it once fully converted in ACRO but would really want to use SITL for doing further development. I heard from @teamtiltrotor that you had a crude model working in last_letter?
George Zogopoulos-Papaliakos
@Georacer
Hello @fhedberg
last_letter development was at a pause during the last two weeks, but I believe some progress will happen from now on.
Indeed, I have done some promising work for generic wing+rotor configurations.
Fredrik Hedberg
@fhedberg
Sweet
George Zogopoulos-Papaliakos
@Georacer
Some bugs on the physics engine and lag on the APM - ROS udp communcation have held the whole thing back from being a reliable simulation resource
but I'm looking into it.
I've gotta go right now, but hopefully I'll log in again tonight. Where do you live? Are our time zones compatible?
Fredrik Hedberg
@fhedberg
Sounds great! I'm in Sweden, on CET.
George Zogopoulos-Papaliakos
@Georacer
Just -1 from me. Great.
We'll talk later. Cheers!
George Zogopoulos-Papaliakos
@Georacer
Hello @fhedberg . Are you available?
Fredrik Hedberg
@fhedberg
@Georacer Yes, for 15 min only unfortunately.
George Zogopoulos-Papaliakos
@Georacer
We'll make do.
The rest are for tomorrow.
Do you have any specific goals or questions?
Fredrik Hedberg
@fhedberg
So, i have implemented motor/mixing/tilt support in APM for a dual swashplate based tiltrotor, similar to the real V22 Osprey.
I added SITL (based on tridgesim;) support for tandem helicopter and it have been much helpful in testing.
So now I'm exploring the options for full tiltrotor simulation, that would require both dual swashplate rotor simulation and full aerodynamical control surfaces support.
George Zogopoulos-Papaliakos
@Georacer
I see.
Fredrik Hedberg
@fhedberg
I've flown in manual (ACRO/STABILIZE) before, but for doing further development, I think starting out with simulation will end up cheaper ;)
George Zogopoulos-Papaliakos
@Georacer
So you have a physics model for the swashplate or it is just thrust vectoring?
Fredrik Hedberg
@fhedberg
No physics model, it's an extension of @tridge's trad heli model in APM SITL
fhedberg/ardupilot@996a581
But that's for CH47 type tandem only, not tiltrotors.
George Zogopoulos-Papaliakos
@Georacer
Do you need to have a simulation for the swash plate dynamics or is a simple tilted brushless motor with a fixed propeller adequate?
Will your final vehicle have swash plates under the propellers?
Fredrik Hedberg
@fhedberg
My physical vehicle have dual swash plates yes, 600 size blades so it's over 2m wingspan.
But I suppose I can shim the swash plate "trustvectoring" into a titled brushless motor, if it can tilt in 2d.
George Zogopoulos-Papaliakos
@Georacer
That complicates things. Okay, two thoughts:
  1. I have to prioritize the team-tiltorotor development of a firefly Y6 model.