Added support for HooperFly Hexa and Quad frames #1081
Conversation
|
Thx! |
|
Hi Felix,Yeah, I thought about that but I think there is good reason. The current simulation models are very generic and I was curious about making more detailed FDM descriptions for the two different quad frames as well the the hexacopter frames. The chords/radius and moment couples are different for each craft. I was hoping to make more refined physical models for simulation.I also noticed some odd simulation behavior that I want to investigate a bit more to see if it's a usage error or a real issue. It's related to swapping motor rotations and how the simulations differ. I will dig into some more and let you know what I find.I will probably do an octocopter model in the next few weeks as well.Happy to be involved with the project.Regards,Rich -------- Original Message -------- Thx! One thing: there is no need to duplicate the jsbsim files for racerpex and teensyfly with only name changes... It would make more sense to just make one quad and one hexa file and use them for both. —Reply to this email directly or view it on GitHub. |
|
Hi Felix,To your point, I didn't add the RacerPEX specific radius/moment point modifications before submitting. I will update the RacePEX jsbsim files with the appropriate data so that it's not a duplicate of the TeensyFly files. Thanks for the input. Much appreciated.Rich -------- Original Message -------- Thx! One thing: there is no need to duplicate the jsbsim files for racerpex and teensyfly with only name changes... It would make more sense to just make one quad and one hexa file and use them for both. —Reply to this email directly or view it on GitHub. |
|
@flixr Over the weekend, I successfully simulated an octocopter frame as well as reorganized some files. I put all the HooperFly related jsbsim/aircraft files into a HooperFly directory to isolate them from the simple example files. If there's interest in having a "simple_x_hexa" and a "simple_x_octo" file to compliment the current set of "simple_*" files? |
|
Yes, that would be great! |
|
@flixr, I've been busy making HooperFly related airframes definitions and associated simulator aircraft definitions. One of the things that I have noticed related to rotorcraft is the variability in motor layout. Specifically, a consistent method for the location of "motor 1" in a given frame type(e.g. tri, quad, hexa, octo, custom) and configuration(e.g. +, x, other). Since I build multiple frame types of all sizes and configurations, I have adopted a convention that "motor 1" is the motor to the left or on the positive Y axis given a standard cartesian X/Y coordinate system. I also assign "motor 1" a clockwise motor spin. Motor 2-N proceed in a clockwise direction about the origin of the coordinate system and alternate spin based on their position. Do you think this a reasonable methodology to use on the Paparazzi project? Looking at the current examples, it's a bit eclectic. |
|
@flixr, here's a link to a graphic: https://wiki.paparazziuav.org/wiki/User:Earthpatrol#Motor_Layout_Diagrams:_Quad.2C_Hexa.2C_Octo |
|
Numbering is fine. |
|
@flixr I'll put the little triangle on the front of the new icons like the one on the "rotorcraft" icon. Email address configured. Thanks. |
|
@esden @flixr Updated the icons with a "front" indicator. All of them are in the "+" configuration. What are your thoughts on having the icon reflect the frame orientation verbatim? Given the current icon template implementation, we could expand the icon names to include an orientation indicator(e.g. quadrotor & quadrotor_x, hexarotor & hexarotor_x, octorotor &octorotor_x); where "_x" denotes a frame in an "X" configuration. |
|
Thx! One remaining question: do you really want/need all the test_progs in each airframe file? |
|
Sure, so I wait until you have tuned this before merging? |
|
So you definitely don't need the test_progs firmware and |
|
@flixr yes, wait to merge. I will hopefully be integrating and flight testing this weekend. I will hack out the test_progs and rotorcraft_autopilot sections as well. |
|
@flixr I think it's good to merge. Thanks. |
|
@flixr it looks like the changes to acIcon.ml, acIcon.mli, and mapTrack.ml are the versions prior to the the addition of quadrotor_x, hexarotor_x, and octocopter_x. Do I need to do another pull request to get them added? |
|
Ups, sorry... added with 1e4f73f |
|
Hi @flixr. Thanks for the merge. Github is still of the opinion that there are commits that have not been merged --> Reference the "Closed with unmerged commits" entry in this thread. |
|
That is just because I didn't simply merge your commits but rebased/squashed them... |
|
@flixr that was my assumption but wanted to verify. I will update the wiki to include the new rotor icon types: quadrotor, hexarotor, octorotor, quadrotor_x, hexarotor_x, and octorotor_x. Thanks again for all the help. |
Added two new rotor icon types:
"hexarotor" - for use by hexacopter frames (uses a hoop for the body)
"quadrotor" - for use by quadrotor frames (uses a hoop for the body)
Added and tested simulator aircraft definitions:
Added and tested airframe configuration definitions: