|
Post by hexpod on Jun 21, 2016 18:26:53 GMT
I can notice that the combined pitch/surge and roll/sway cues are handled in a strange way in the x-sim kinematic equation.
- When you apply pitch first, the surge output is moving on the previously applied pitch plane.
- When you apply roll first, the sway output is moving on the previously applied roll plane.
Should normally the surge and sway move linearly on the x/y axis regardless to the pitch/roll angle, like in BFF software?
Do you see any reason why the surge/sway are moving on the pitch/roll angular planes ("off axis") or it's simply a bug???
What do you think guys?
|
|
|
Post by hexpod on Jun 22, 2016 11:23:50 GMT
I can notice that the combined pitch/surge and roll/sway cues are handled in a strange way in the x-sim kinematic equation.
- When you apply pitch first, the surge output is moving on the previously applied pitch plane.
- When you apply roll first, the sway output is moving on the previously applied roll plane.
Should normally the surge and sway move linearly on the x/y axis regardless to the pitch/roll angle, like in BFF software?
Do you see any reason why the surge/sway are moving on the pitch/roll angular planes ("off axis") or it's simply a bug???
What do you think guys?
bff vimeo.com/171742539 x-sim vimeo.com/171742499
In this test I've applied a pitch angle and I am moving the surge slider.
Thanos could you comment please what kind of kinematic is more appropriate to drive our simulators??
|
|
|
Post by tronicgr on Jun 22, 2016 11:35:35 GMT
From the videos you posted for comparison, the BFF software handles better the 6DOF math.
As you can see the x-sim plugin, limits the motion of the opposite motors a little bit resulting in different angle, while you get the full range for all motors in BFF software to allows the combined surge and pitch motion to be correct.
So I think it has to do with the limiters in x-sim plugin.
Thanks Thanos
|
|
|
Post by hexpod on Jun 22, 2016 12:14:49 GMT
From the videos you posted for comparison, the BFF software handles better the 6DOF math. So I think it has to do with the limiters in x-sim plugin. Thanks Thanos I want to be sure you got my question right.
watch please this clip:
vimeo.com/171751557
While moving the surge slider, the 3d model is moving the platform on the Pitch plane (the movement is NOT horizontal)
So basically speaking, the linear movements (surge,sway,heave) are linked with the angular positions (pitch, roll).
Don't you think they should be independent and the surge movement should be horizontal regardless to the platform's angular position??
Don't you think it has to do with the equation?
cheers
|
|
|
Post by tronicgr on Jun 22, 2016 17:54:43 GMT
Does it move same way in the 3D simulation window? If yes, then something is wrong in the math equations. Otherwise it could have something to do with the limiters.
Another thing that may affect it is that the Center of Gravity point stays stationary, it's not following the platform!!!!!! Try placing COG to 0 for XYZ
Thanks Thanos
|
|
|
Post by hexpod on Jun 22, 2016 18:14:10 GMT
Does it move same way in the 3D simulation window? If yes, then something is wrong in the math equations. Otherwise it could have something to do with the limiters. Another thing that may affect it is that the Center of Gravity point stays stationary, it's not following the platform!!!!!! Try placing COG to 0 for XYZ Thanks Thanos The CoG and also "heave offset" has no impact on this behavior.
Watch please the clip from my last post. The video shows 3d simulation window.
It seems it has nothing to do with my MOD. I've noticed the same behavior in the pervious versions.
cheers
|
|
|
Post by hexpod on Jun 23, 2016 13:04:41 GMT
Thanos,
This is how should look, in my opinion, the combined pitch/surge movement: vimeo.com/171928970
and NOT like we have for the moment: vimeo.com/171751557
Could you please have a closer look and confirm that we have a wrong math in the x-sim plugin?
That could be a starting point to think how it could be fixed.
Thanks
|
|
|
Post by tronicgr on Jun 24, 2016 14:11:11 GMT
I see, it happens in the 3D simulation as well. If you notice the COG calculation part is always aligned 90 degrees to the X/Y axis. Like the reference point is not on the base but the top platform.
Perhaps the COG equation uses the Top Platform points instead the Base points for the calculations.
Thanks Thanos
|
|
|
Post by hexpod on Jul 4, 2016 20:35:06 GMT
I see, it happens in the 3D simulation as well. If you notice the COG calculation part is always aligned 90 degrees to the X/Y axis. Like the reference point is not on the base but the top platform. Perhaps the COG equation uses the Top Platform points instead the Base points for the calculations. Thanks Thanos Hmmm... I've spend n hours (a lot!) trying to isolate the problem. No success. More I dig into it more I think it can be hidden everywhere.
We need a guy with mathematical experience in inverse kinematics. Do you still have contact to John? maybe he could help...
cheers
|
|
|
Post by hexpod on Jul 5, 2016 10:25:44 GMT
"Perhaps the COG equation uses the Top Platform points instead the Base points for the calculations."
Or maybe the opposite. That's why the CoG is fixed. In my opinion, the CoG should follow the DoF movements.
Is it logical for you?
Erasing all "gcenter" variables from the code doesn't correct this behavior. That's odd.
|
|
|
Post by tronicgr on Jul 5, 2016 11:13:14 GMT
"Perhaps the COG equation uses the Top Platform points instead the Base points for the calculations."
Or maybe the opposite. That's why the CoG is fixed. In my opinion, the CoG should follow the DoF movements.
Is it logical for you?
Erasing all "gcenter" variables from the code doesn't correct this behavior. That's odd. I think the COG should be recalculated every time based on the TOP platform location. Right now is calculated once at the start of the program using the BASE platform as reference. It is hard task indeed, I'm not good in math either and thus the reason I could not get these equations fixed. I tried to get in contact with John years ago, but he didn't respond, I hope he is ok, I remember he had heart health issues. Thanks Thanos
|
|