• Which the release of FS2020 we see an explosition of activity on the forun and of course we are very happy to see this. But having all questions about FS2020 in one forum becomes a bit messy. So therefore we would like to ask you all to use the following guidelines when posting your questions:

    • Tag FS2020 specific questions with the MSFS2020 tag.
    • Questions about making 3D assets can be posted in the 3D asset design forum. Either post them in the subforum of the modelling tool you use or in the general forum if they are general.
    • Questions about aircraft design can be posted in the Aircraft design forum
    • Questions about airport design can be posted in the FS2020 airport design forum. Once airport development tools have been updated for FS2020 you can post tool speciifc questions in the subforums of those tools as well of course.
    • Questions about terrain design can be posted in the FS2020 terrain design forum.
    • Questions about SimConnect can be posted in the SimConnect forum.

    Any other question that is not specific to an aspect of development or tool can be posted in the General chat forum.

    By following these guidelines we make sure that the forums remain easy to read for everybody and also that the right people can find your post to answer it.

Wiggles and waggles

JonPatch

JonPatch

Resource contributor
Messages
443
Country
ca-britishcolumbia
Is there a solution to the issue that AI ships following an AFCAD taxiway waggle slightly when passing through nodes on a corner? AI aircraft as far as I can tell don't do this. Why the difference? And nope, I mean a solution other than using an FSX boat traffic file.
 
It would be great if someone has a solution to the problem Jon mentions. For reference, I've tried the following all of which had little or no effect: one would think that greatly increasing the yaw and or longitudinal moments of inertia in the aircraft.cfg file would damp this effect but it did not appear to do so. AI aircraft on the ground must be treated differently than piloted craft because the MOI settings have a distinct effect on the pilotable ferry I made. With the various MOI settings adjusted properly, I can get a quite realistic heel with course changes and appropriately slow responses to steering input.
Spreading the attachpoints much farther apart had no effect. Using my custom .air file which I had made for ferries had no effect on this problem. Greatly increasing the weight had no effect.

I wonder if the issue might have something to do with how FSX controls the velocity (speed and direction) of an AI aircraft. How does this work? Pure speculation, but maybe it could work like this: when the craft passes a node, or is within a certain range of a node, an acceleration is produced based on the angle between the current course and the upcoming course, and it is directed perpendicular to the upcoming course.

The problem might be that time is quantized, like this: at the next computation, the idea of the next calculation is not to just aim at the next node in the sequence but to return the craft to the track. Garmin would call the amount off the intended track the 'cross track error'. So maybe a computation is made to get the boat back on the track based on the CTE. But if the calculation rate is rather slow, by the time of the next calculation the boat may have not only gotten back on the track but might now be on the other side of the track. Again this would result an acceleration back towards the track, with the result that the course is a damped oscillation which dies out after several cycles, several course wiggles. If this is the way it works, or something like this, I can't see any obvious solution. So I hope I'm wrong!

AI boats do something like this when they reverse the direction of a route; they back up, stop, then accelerate and turn, but not 180 degrees as one would expect, but often 20 or 30 degrees beyond 180. They then can end up on the other side of the track and correct back to the proper course. This overshoot also happens with other dramatic course changes. My AI boat solution has been to put 3 or 4 closely spaced waypoints in a uniform arc where I want the boat to turn, and no other waypoints between these spots.
 
Hi,

I know I did it but unfortunately, migrating to my new PC, I lost all of my cfg extensions. None of my former customized FS9 AI ships work yet in FSX due to the absence of their cfg's ( the aircraft.cfg, the model.cfg and the panel.cfg).
I do remember though that I had the AI .air file adapted to have the ship heel over when turning and thus having smooth turns and not what you call wiggle waggle.
Wish I had the courage to start it all over again but I am still in the early stages of getting them all back.
Of course, in FSX I use AIBTC which obviously is not of interest to you.
Fyi: in FSX and with AIBTC, I had an AI Ferry pitch and roll crossing the channel (i.e. the straight between France and the UK) by putting in an extra so called airport with apron links all over at different elevations mimicking waves.

AI waves.jpg


Maybe something of what I said, can help pinpoint your problem?

Roby
 
Roby, do you remember what you did to the AI .air file to cause the effects you wanted?
The .air file overrides some of the .cfg parameters, but I can't remember which ones. If it has MOI values for example that override, that might be a simple solution.
I'll have to check my .air documentation... which is somewhere... I think I even have a file someone developed that has a lot of reverse engineered equations showing exactly how FSX uses the values.
 
Sorry, I have been distracted by the news of the B772 crash over Ukraine.
Yes, I think that I did something to the roll characteristics in the 1501 so that the vessel would lean over (heel) when turning.
Sorry, I will have to find back the air file entries somewhere in a zip file (if I did zip it up at the least:() I made, to be able to tell you how I did it (here should come the non existent embarrassment smiley on this site).
Pleased to know though that I am not the only one to have misplaced or lost some valuable files (old age?) ;).
But, once more, in order to check what I did, I have to find back the relevant vessel and its cfg and air file (the air files were gone as well or had received another extension and I do not remember which one out of the 73 I adapted). That will take some time:(.
Will do my best and revert asap.

Roby
 
One more thing: the rolling and pitching (adding extra airport apron path links) does work in FSX when using AIBTC, but NOT when using FS9 AFCAD routes, as it has nothing to do with the aircraft.cfg or the .air file!
 
Roby, does anything in the attached look familiar?
Of all the values one could play with, the Yaw Moment - Yaw Rate (Damping) looks like a candidate. I had left this at the default value for the pilotable boat.
The explanations in AAM seem to suggest that all the other variables would not deal with the problem directly. Perhaps indirectly by changing the turning moments, but I would think those would primarily affect the turning rate. That might be a good thing for a boat.

Perhaps some of the side force values might help, but figuring out just what they do from the equations looks a little daunting.
 

Attachments

  • air.jpg
    air.jpg
    153.6 KB · Views: 469
Thanks for digging through the archives, Roby, much appreciated. Since you've dropped such juicy hints, you'll be in a race with Larry to see who finds the solution first. ;-)
 
Well, I think I have found a solution to the wigwag; the sufficient conditions that is. What I don't know yet are the necessary conditions. Experimenting with the two AI boats in Vic+, none of the tweaks to the .air file that seemed to have any logic to them changed much if anything, so I again tried experimenting with the aircraft.cfg. Changes to the MOI, even huge ones had little or no effect. The more I watched the effect, the more it seemed that the problem was overly sensitive steering, so I reduced the rudder effectiveness (2.0 for the pilot boat) a little, then a little more --- maybe better--- then a lot. That seemed to do the trick for that boat, even using the original .air file. It's the old problem, if something does not work, do you need to do it more, or less, or in the opposite direction. (A common problem in medicine!) So far I've not fixed the problem on the cruise ship, but I think now it's just a matter of doing more of the same, maybe a lot more. I'll see if I need to tweak the .air file too. (I had also changed the aileron effectiveness for the pilot boat, but that by itself did not help, so I'll have to see if the cruise ship needs that adjustment too.)

Roby, does 'roll moment - yaw rate' in section 1101 of the .air file ring a bell? The claim is that 'neg tends to roll the AC level in turns'. Is that what you used to get the desired heel? Maybe a lot of negative so the heel is opposite to the turn?
 
Yes it does. I made a mistake when I said 1501. It was 1101 and roll was involved.
The vessel heeled over a lot afterwards.
I managed to get a pilotable FFG working again but unfortunately not the one that I adapted. But that one was pilotable and not AI. I am still looking for the AI one but real life things came in between.
 
You must log in or register to reply here.
Back
Top