Pr liftdrag coefficients #690
Conversation
| this->cmaStall * (this->alpha + this->alphaStall)) | ||
| * cosSweepAngle; | ||
| cm = this->cm_alpha0 - this->cma * this->alphaStall + | ||
| this->cmaStall * (this->alpha + this->alphaStall); | ||
| // make sure cm is still less than 0 | ||
| cm = std::min(0.0, cm); |
There was a problem hiding this comment.
Does this assumption still hold with the defined cm_alpha0 ?
There was a problem hiding this comment.
Previously, cm behavior was symmetric around zero, while now we start at cm_alpha0. Given the simplified aerodynamics we use in this plugin, assuming now it is symmetric around cm_alpha0 makes sense. I'll add this to the pull request.
Note that for current simulation models, it doesn't matter, since those all set cm to 0 under all conditions.
There was a problem hiding this comment.
In fact, what I can find indicates that past stall, cm is typically more negative than cm_alpha0 (positive angle of attack), e.g. this. I'll improve on this too.
|
I added 3 commits to the pull request that address the feedback about Cm, a bug that slipped in and a docstring that was not up to date (the documentation of Gazebo 9 and later clarify that link.AddForceAtRelativePosition works relative to the CoG of the base link, not the origin). I haven't found the time yet to create or improve an example simulation model. |
|
@pjdewitte Please do, otherwise we have no way to test the newly introduced parameters |
|
@pjdewitte Would it be possible to share a timeline when you could add it to the models? It would be great that we could verify this PR by adding it to the models |
|
I should have thought about a testcase from the start. I'll adjust one of the models by the end of this week (24/01/2021). |
|
Flight log with plane model, using the committed new parameters: link |
|
@sfuhrer @RomanBapst FYI |
|
@pjdewitte Have you ever tested this with the techpod? It doesn't seem to be able to takeoff anymore |
This reverts commit 38224fb.
* liftdrag_plugin code cleanup * Add cd_alpha0, cm_alpha0 and cd_delta * Remove sweep correction: already covered by velocity in Lift-Drag plane * Deprecate controlJointRadToCL and a0 * Rename deprecated XML tags * Make cm post-stall behavior symmetric around cm_alpha0 * Fix: post-stall drag (pitch-down stall) not properly accounted for * Correct docstring of cp * Plane: rename elevon to aileron, to align with mixer file. Give elements unique name. * Plane: use new aero parameters Co-authored-by: Pieter-Jan Dewitte <pieterjan.dewitte@atmosuav.com>
This pull request adresses 1, 2, 3 and 5 from #684