[1.9-1.10] Throttle Controlled Avionics

[smjjames]

Any idea what happens if they get out of physics range? Because three of them accidentially threw themselves skyward while trying to catch up with me, and also had to go up a slope (I'm still ascending in the first screenshot for the same reason since I was going pretty fast).

*a couple minutes later*

One of them apparently was able to slow down enough and was able to return, I don't know why the other two couldn't pretty sure they didn't run out of fuel.

[allista]

Any idea what happens if they get out of physics range? Because three of them accidentially threw themselves skyward while trying to catch up with me, and also had to go up a slope (I'm still ascending in the first screenshot for the same reason since I was going pretty fast).

http://img.photobucket.com/albums/v214/smjjames/testity_screenshot_2015-09-24--09-04-26_zpshu8ia5ct.png

http://img.photobucket.com/albums/v214/smjjames/testity_screenshot_2015-09-24--09-04-29_zpskbwvxwxt.png

*a couple minutes later*

http://img.photobucket.com/albums/v214/smjjames/testity_screenshot_2015-09-24--09-07-19_zpsls0lifiz.png

One of them apparently was able to slow down enough and was able to return, I don't know why the other two couldn't pretty sure they didn't run out of fuel.

As far as I know (and I don't know that far), while you're not switching between ships, vessels that are out of physics range get packed. This means any modules on parts (TCA included) become inactive, they don't get updated by Unity and all functionality is paused. If, however, you're trying to switch ships, out-of-range vessels go on rails. Then it depends: in atmosphere they just get destroyed. On Mun they should retain their orbits and should eventually smash into the ground...

- - - Updated - - -

Oh yeah, while doing testing, I've noticed that they tend to pair up, which is cool emergent behavior, the pairs aren't constant though, but the general pairing up is.

Also, flying in formation in low-grav, so far, so good.

http://img.photobucket.com/albums/v214/smjjames/testity_screenshot_2015-09-24--08-28-25_zpsztyqhrmt.png

Edit: In excess of 110m/s and still holding formation. They did fall back a little when I sped up quickly, but the followers quickly kept pace.

Edit2: Making a sudden stop did throw them off momentarily when I got to a waypoint, but didn't take long for them to catch up. Going to test this with the two atmosphere craft because they were having a real problem keeping up there.

Wow! Looks really cool

The 'pairing' is a typical example of self-organisation of a system with local constraints: each craft just tries to be out of the way of a weighted center of all the others, where weights are calculated from proximity and relative velocity. On the other hand, each craft tries to fly directly to the leader. These opposing actions combined give the 'formation' effect.

[komodo]

I would say that I'm at moderate level of building planes for FAR (never played planes without it), but right now (and in days to come) I'm mostly "testing" this

http://i.imgur.com/Ot7iNs5.jpg

http://i.imgur.com/EhMmBZ0.jpg

- - - Updated - - -

Test it in v13 -- should make difference

The problem was as follows: leader craft follows path and in the process automatically switches targets (wp1, wp2...); in the meantime the camera is focused on one of the followers which should be targeting the leader. But when the leader switched from wp1 to wp2, so did the active vessel (follower).

Wow, haven't seen this mod before, it looks like a winner for sure. I'll be trying it out as soon as I'm able to sit and play KSP again ><

Related note: an old gas chromatograph? It looks like it to my trained eye, although I admit I'm most familiar with HP models only... If so I entirely understand your pain

Edit: GC for sure! Then I zoomed in and no wonder I didn't recognize it! Instrumentation transverses every language barrier, hm?

Edited September 24, 2015 by komodo

[allista]

Wow, haven't seen this mod before, it looks like a winner for sure. I'll be trying it out as soon as I'm able to sit and play KSP again ><

Related note: an old gas chromatograph? It looks like it to my trained eye, although I admit I'm most familiar with HP models only... If so I entirely understand your pain

Edit: GC for sure! Then I zoomed in and no wonder I didn't recognize it! Instrumentation transverses every language barrier, hm?

Indeed a GC it is! An old one with attached custom made ADC->USB circuit and software >_<'

Edited September 24, 2015 by allista

[komodo]

Indeed a GC it is! An old one with attached custom made ADC->USB circuit and software >_<'

Anecdotally, we had a varian GC donated to my institution last year; sans the ADC board. Worse, they didn't even unplug it, just cut the wires. So we've got a great GC, with a bunch of unmarked analog signal wires... Appropriately, the auto sampler doesn't have its power supply, and thus is also a paperweight:confused:

Moral of the story, at least there is a USB plug!

[allista]

Anecdotally, we had a varian GC donated to my institution last year; sans the ADC board. Worse, they didn't even unplug it, just cut the wires. So we've got a great GC, with a bunch of unmarked analog signal wires... Appropriately, the auto sampler doesn't have its power supply, and thus is also a paperweight:confused:

Moral of the story, at least there is a USB plug!

Ouch!

And no one tried to fix the thing? At least figure out the wires using a multimeter and attach a generic ADC.

Electronic components shops usually have complete circuits + some recording soft dubbed 'digital oscillograph' like this one

Power supply shouldn't be a problem either, if you know the specs...

[allista]

Here's the

next version.

• Fixed autopilot resetting on collision. Needs testing.
  
• Improved collision prevention at descent/ascent. But can't guarantee that collisions will not happen at all. Needs testing.
  
• Warning: This is a debug version that will spam logs with messages and display CPS ship interactions.
  

As usual, I'm asking you for help:

I've created a google spreadsheet with the tests that came to mind. You are free to add other tests, comment and mark existing tests as done (with some screenshots, maybe).

I'm hoping that all tests will be done until the next weekend, so I could release all this at last

[smjjames]

Question on the debug version, what does the reload globals button reload and what does the reset button reset exactly?

[allista]

Question on the debug version, what does the reload globals button reload and what does the reset button reset exactly?

Sorry, didn't have time to document it yesterday

Reload Globals reload TCA.glob file; it is needed for tuning different parameters there at runtime.

Reset forces TCA of current vessel to completely reinitialize; this is what happens when e.g. some part is destroyed. Ideally, when you push Reset nothing in the behaviour of the craft should change.

- - - Updated - - -

Here's the next version.

I've added a real wedge formation of the followers to further decrease possibility of collisions...

[allista]

Here's the

next version.

• Fixed uncontrolled ascent bug.
  
• Improved collision prevention by actually predicting collisions and decreasing amount of maneuvering near the target.
  
• Improved waypoint navigation: the speed near intermediate waypoints depends on the sharpness of the turn to the next waypoint.
  
• Added waypoint management in-flight; no need to go to MapView each time.
  

[smjjames]

Here's the

next version.

• Added waypoint management in-flight; no need to go to MapView each time.
  

Thanks. I was like 'If MechJeb can do it, why not TCA?'. Also, what do you think is a reasonable number of craft to test the 'follow leader' routine and collision detection with? 10 or 4? 4 is probably a more likely number, but 10 seems to catch collision problems more often.

Edit: Infinite fuel and RCS is still on for this debug version.

Edited September 29, 2015 by smjjames

[allista]

Thanks. I was like 'If MechJeb can do it, why not TCA?'. Also, what do you think is a reasonable number of craft to test the 'follow leader' routine and collision detection with? 10 or 4? 4 is probably a more likely number, but 10 seems to catch collision problems more often.

Edit: Infinite fuel and RCS is still on for this debug version.

Oh, i forgot about it. I can disable it right now by commenting a single line and publish a .dll

I would say that 10 is a reasonable system overload; I've tested it on 8+leader different crafts.

[smjjames]

Oh, i forgot about it. I can disable it right now by commenting a single line and publish a .dll

I would say that 10 is a reasonable system overload; I've tested it on 8+leader different crafts.

Yeah, I did it with 20 once for fun and it starts getting laggy and the memory leaks (not from TCA) start happening more around 10, plus the setup starts getting tedius beyond 10 or so.

Also, I'm still getting this exception when the craft crashes into something and gets destroyed (was doing a walkthrough of KSC for fun) orI switch to debris:

NullReferenceException: Object reference not set to an instance of an object
  at ThrottleControlledAvionics.ThrottleControlledAvionics.get_CFG () [0x00000] in <filename unknown>:0 

  at ThrottleControlledAvionics.ThrottleControlledAvionics.WaypointOverlay () [0x00000] in <filename unknown>:0 

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at RenderingManager.OnGUI () [0x00000] in <filename unknown>:0 

[allista]

Here, release build without infinite fuels and with the attempt to fix that NRE:

https://www.dropbox.com/s/agskoyrs92wtg9x/ThrottleControlledAvionics.dll?dl=0

[smjjames]

Thanks, and just crashed (intentionally) into terrain and into a building and looks like the NRE is fixed.

Edit: Not completely, found a variant of it that happens when the leader craft has followers and the leader craft crashes and the followers try to find the last known location:

NullReferenceException: Object reference not set to an instance of an object
  at Vector3d.get_sqrMagnitude () [0x00000] in <filename unknown>:0 

  at ThrottleControlledAvionics.PointNavigator+<offset_updater>c__Iterator0.MoveNext () [0x00000] in <filename unknown>:0 

  at ThrottleControlledAvionics.PointNavigator.update_follower_offset () [0x00000] in <filename unknown>:0 

  at ThrottleControlledAvionics.PointNavigator.Update () [0x00000] in <filename unknown>:0 

  at ThrottleControlledAvionics.ModuleTCA.OnFixedUpdate () [0x00000] in <filename unknown>:0 

  at Part.ModulesOnFixedUpdate () [0x00000] in <filename unknown>:0 

  at Part.FixedUpdate () [0x00000] in <filename unknown>:0 

Edited September 30, 2015 by smjjames

[allista]

Here's the

next version.

• Fixed NREs on leader/followers crash, etc. (I hope )
  
• Improved collision prevention by:
  1) employing stop-turn-go strategy on sharp turns instead of smooth but chaotic maneuvering.
  2) making maneuvering into formation sequential: now only one follower at a time can move to its in-formation position; others will have to fly on parallel courses.
  3) making formation position depend on starting position of a follower instead of the rigid 'alphabetical' order.
  
• Made HSC control of manual engines adaptable. WIP.
  
• Added Squadron Mode: ability to assign a numerical ID to a bunch of vessels and then (if they're in range) control them all through TCA GUI of any of them. Naturally, not all TCA parameters are sent to squadron members, but autopilot commands are.
  

Edited October 1, 2015 by allista

[smjjames]

Here's the

next version.

• Fixed NREs on leader/followers crash, etc. (I hope )
  
• Improved collision prevention by:
  1) employing stop-turn-go strategy on sharp turns instead of smooth but chaotic maneuvering.
  2) making maneuvering into formation sequential: now only one follower at a time can move to its in-formation position; others will have to fly on parallel courses.
  3) making formation position depend on starting position of a follower instead of the rigid 'alphabetical' order.
  
• Made HSC control of manual engines adaptable. WIP.
  
• Added Squadron Mode: ability to assign a numerical ID to a bunch of vessels and then (if they're in range) control them all through TCA GUI of any of them. Naturally, not all TCA parameters are sent to squadron members, but autopilot commands are.
  

'Alphabetical order'? Does that mean it depends on the name of the craft?

[allista]

'Alphabetical order'? Does that mean it depends on the name of the craft?

No, it meant that it depended on vessel's Guid

Edit: thus it was possible to make a formation by applying only local rules, without communication with other followers.

[sebi.zzr]

Hello,i just installed this mod and i noticed that,every time i lift off (just couple meters),landing legs get auto retracted.At first i thought this is some kind of feature of this mod,but when landing,legs don't auto deploy.I couldn't find any settings for that,so could you disable "auto retract".Thank you.

[allista]

Hello,i just installed this mod and i noticed that,every time i lift off (just couple meters),landing legs get auto retracted.At first i thought this is some kind of feature of this mod,but when landing,legs don't auto deploy.I couldn't find any settings for that,so could you disable "auto retract".Thank you.

Autoretraction is indeed a feature, and autodeployment is also there if you use the Land autopilot. To be able to auto deploy legs on manual landing TCA should somehow guess if you're landing or just flying close to the ground or building. I'll think about how to handle such situations. Meanwhile, I'll add an option to disable autoretraction to the next beta release. Oh, and keep in mind: versions that I publish in this forum are BETAs. The latest stable version (which is much simpler and has less features) may be found on KerbalStuff.

[sebi.zzr]

Thank you for clarification,if i may suggest,the legs should "auto retract-deploy" little more late or higher.

But now i have another issue.I wanted to try the version from kerbalstuff and now the toolbar icon is gone(is there in menu,but not visible when selected to show).

I changed back to version i had,and problem still persist.I checked in ModuleManager.ConfigCache and all cockpits and engines are patched correctly.

Is this a known issue?

[allista]

Thank you for clarification,if i may suggest,the legs should "auto retract-deploy" little more late or higher.

But now i have another issue.I wanted to try the version from kerbalstuff and now the toolbar icon is gone(is there in menu,but not visible when selected to show).

I changed back to version i had,and problem still persist.I checked in ModuleManager.ConfigCache and all cockpits and engines are patched correctly.

Is this a known issue?

Unfortunately, yes. Apparently, it's a bug in the Toolbar itself. Try to rename or delete the GameData/toolbar-settings.dat file.

*also, you can change the delay of auto-retraction in GameData/ThrottleControlledAvionics/Plugins/PluginData/ThrottleControlledAvionics/TCA.glob file, in TLA section: GearTimer value in seconds.

[allista]

Now that's a real autopilot AI!

[allista]

Here's the

next BETA version.

• Fixed several squadron commands.
  
• Added automatic landing gear deployment on manual landing.
  *Manual landing is defined as a steady decrease in altitude with horizontal speed of no more than 1m/s. The gear is deployed near the ground.
  
• Reworked GUI to be a little more compact.
  
• Added 'advanced' options pane; moved throttle sensitivity, named configs and other rarely used controls there.
  

Note: this is a debug version which will show CPS and AutoLander state vectors.

Edit: no infinite fuels by default)

Edited October 2, 2015 by allista

[smjjames]

Bug with the set point text window in v18, if you delete all of the text, it goes and puts the current set point back in there. It might not be a bug, but it is somewhat annoying.